Emerging Spatial
Information Systems
and Applications

Brian N. Hilton
Claremont Graduate University, USA

Idea Group Publishing

Hershey • London • Melbourne • Singapore
ii

Acquisition Editor: Michelle Potter

Senior Managing Editor: Jennifer Neidig
Managing Editor: Sara Reed
Development Editor: Kristin Roth
Copy Editor: Susanna Svidunovich
Typesetter: Marko Primorac
Cover Design: Lisa Tosheff
Printed at: Integrated Book Technology

Published in the United States of America by

Idea Group Publishing (an imprint of Idea Group Inc.)
701 E. Chocolate Avenue
Hershey PA 17033
Tel: 717-533-8845
Fax: 717-533-8661
E-mail: [email protected]
Web site: http://www.idea-group.com

and in the United Kingdom by

Idea Group Publishing (an imprint of Idea Group Inc.)
3 Henrietta Street
Covent Garden
London WC2E 8LU
Tel: 44 20 7240 0856
Fax: 44 20 7379 3313
Web site: http://www.eurospan.co.uk

Copyright © 2007 by Idea Group Inc. All rights reserved. No part of this book may be reproduced in any
form or by any means, electronic or mechanical, including photocopying, without written permission from the
publisher.

Product or company names used in this book are for identification purposes only. Inclusion of the names of the
products or companies does not indicate a claim of ownership by IGI of the trademark or registered trademark.

Library of Congress Cataloging-in-Publication Data

Emerging spatial information systems and applications / Brian N. Hilton, editor.

p. cm.
Summary: "This book presents innovative spatial information systems developed for a specific problem or
decision-making situation. It also discusses key concepts and theories underlying current spatial information
systems and technology trends and emerging concepts that may impact spatial information system development
and applications in the future"--Provided by publisher.
Includes bibliographical references and index.
ISBN 1-59904-075-1 (hardcover) -- ISBN 1-59904-076-X (softcover) -- ISBN 1-59904-077-8 (ebook)
1. Geographic information systems. 2. Spatial analysis (Statistics) I. Hilton, Brian N., 1959-
G70.212.E44 2007
910.285--dc22
2006027711

British Cataloguing in Publication Data

A Cataloguing in Publication record for this book is available from the British Library.

All work contributed to this book is new, previously-unpublished material. The views expressed in this book are
those of the authors, but not necessarily of the publisher.
 

Dedication

To my wife, June, whose support,

patience, and love made this book possible.
v

Emerging Spatial
Information Systems
and Applications

Table of Contents

Foreword......................................................................................................... viii

Preface.............................................................................................................. xii

Section.I:.
Introduction

Chapter.I
Geospatial.Web.Services................................................................................... 1
Peisheng Zhao, George Mason University, USA
Genong Yu, George Mason University, USA
Liping Di, George Mason University, USA

Chapter.II
Spatial.Information.System.Development:.The.Role.of
Information.System.Design.Theory.and.Ontologies.................................... 36
Brian N. Hilton, Claremont Graduate University, USA
Richard J. Burkhard, San Jose State University, USA
Tarun Abhichandani, Claremont Graduate University, USA
 v

Section.II:.
Challenges.for.Spatial.Information.Systems

Chapter.III
Geospatially.Enabled.Directory.for.Emergency.Response
Interoperability............................................................................................... 63
Judith Woodhall, COMCARE, USA

Chapter.IV
The.Weather.Tool:.An.Agent-Based.Approach.to.Information.
Integration....................................................................................................... 85
Gunjan Kalra, Quantum Leap Innovations, USA

Chapter.V
GIME:.A.Geotechnical.Information.Exchange.Architecture
Using.Web.Services........................................................................................110
Wei-Shinn Ku, University of Southern California, USA
Roger Zimmermann, University of Southern California, USA

Chapter.VI
ScienceMaps:.An.Online.Resource.Portal.for.Standards-
Based.Science.Instruction.Using.Geographic.Information.
System.Technology........................................................................................ 133
June K. Hilton, Claremont High School, USA
David E. Drew, Claremont Graduate University, USA

Section.III:.
Decision-Making.Environments

Chapter.VII
Flexible.Spatial.Decision-Making.and.Support:........................................
Processes.and.Systems.................................................................................. 153
Shan Gao, University of Auckland, New Zealand
David Sundaram, University of Auckland, New Zealand

Chapter.VIII
Development.of.a.Web-Based.Intelligent.Spatial.Decision.
Support.System.(WEBISDSS):.
A.Case.Study.with.Snow.Removal.Operations.......................................... 184
Ramanathan Sugumaran, University of Northern Iowa, USA
Shriram Ilavajhala, University of Maryland, USA
Vijayan Sugumaran, Oakland University, USA
v

Chapter.IX
Conservation.Studio:.
Dynamically.Modeling.and.Mapping.Habitat.Suitability......................... 203
Todd G. Olson, Landix, Inc., USA
Brian N. Hilton, Claremont Graduate University, USA

Chapter.X
GIS-Based.Site.Suitability.Decision.Support.System.for.
Planning Confined Animal Feeding Operations in Iowa........................... 219
Ramanathan Sugumaran, University of Northern Iowa, USA
Brian Bakker, Aerial Services, Inc., USA

Section.II:.
Future.Trends.and.Technologies

Chapter.XI
Geo-Communication,.Web.Services,.and.Spatial.Data.Infrastructure:.
An.Approach.Through.Conceptual.Models................................................ 240
Lars Brodersen, Aalborg University, Denmark
Anders Nielsen, National Survey and Cadastre, Denmark

Chapter.XII
A.Data.Visualization.and.Interpretation.System.for.Sensor.
Networks........................................................................................................ 255
Fengxian Fan, Kunming University, People’s Republic of China

Chapter.XIII
Towards.a.Global.Real-Time.Enterprise.................................................... 271
Peter Ibach, Humboldt University Berlin, Germany
Miroslaw Malek, Humboldt University Berlin, Germany
Gerrit Tamm, University of Applied Sciences Erfurt, Germany

Chapter.XIV
Mining.Critical.Infrastructure.Information.from.Municipality.Data.Sets:
A.Knowledge-Driven.Approach.and.Its.Implications................................ 310
William J. Tolone, University of North Carolina at Charlotte, USA
Wei-Ning Xiang, University of North Carolina at Charlotte, USA
Anita Raja, University of North Carolina at Charlotte, USA
David Wilson, University of North Carolina at Charlotte, USA
Qianhong Tang, University of North Carolina at Charlotte, USA
Ken McWilliams, University of North Carolina at Charlotte, USA
Robert K. McNally, University of North Carolina at Charlotte, USA
 vii

Chapter XV
Cognitive Mapping and GIS for Community-Based
Resource Identification ................................................................................ 326
Lyn Kathlene, Colorado State University, USA

Chapter XVI
Spatial Reasoning for Human-Robot Teams ............................................. 351
David J. Bruemmer, Idaho National Laboratory, USA
Douglas A. Few, Idaho National Laboratory, USA
Curtis W. Nielsen, Idaho National Laboratory, USA

About the Authors ........................................................................................ 374

Index .............................................................................................................. 384

v

Foreword

Spatial information systems were created manually for many years. For example,
historically, the source of cholera in London in 1854 was found by John Snow, by
mapping where victims of the disease lived. The graph clearly showed them to be
close to the Broad Street Pump,1 one of the city’s water wells. Another example
is Zipf’s Law,2 which predicted that a number of sociological phenomena. His re-
gression model followed a straight line on log-log paper. For example, Zipf’s law
describes the rank of metropolitan statistical areas (SMSAs) in the census of cities
over 2,500 plotted against their population, or, for that matter, the number of copies
of a St. Louis newspaper bought in suburbs out to 150 miles.
What is new in our computer age is that storage capacity, computing speed, and
technology all grew to the point where large volumes of geographic and spatial
information can be used for understanding business and other phenomena.
Spatial information covers all sorts of data (e.g., demographics, customer locations,
real estate locations and values, and asset location). Sears, for example, uses spatial
information to find the optimal routing for its 1,000 delivery trucks that cover 70%
of the U.S. population. To use the system, they hand-coded millions of customers’
addresses, put them in their data base, and then use the results to determine routes
each day.
A spatial information system is more complex than a conventional back-office
database system. In many respects, it is like a CAD/CAM system where data is
 x

kept on each of many layers, and the layers can be superimposed on one another
as the user chooses. The spatial information system accesses spatial and attribute
information, analyzes it, and produces outputs with mapping and visual displays.
It needs to keep data on spatial boundaries and on attributes of the data. It includes
tools and models to manipulate the data and boundary information. Furthermore,
rather than just adding up columns of numbers, it requires understanding and using
numerical algorithms, statistics, and operations research optimization and simula-
tion techniques. In addition, spatial tools act on boundary layers, including union,
overlay, buffering, nearest neighbor, and spatial attraction. You need much more
expertise and capability to deal with spatial information systems data than with
ordinary databases.
Despite the complexity, or perhaps because of its complexity, spatial information
systems provide capabilities that offer competitive advantage. Of course, determining
the extent of that advantage is a difficult task. Pick3 points out that the up-front data
costs tend to be much higher (more data needs to be acquired) and some costs, such
as training needed for users with no experience, are difficult to estimate accurately.
Furthermore, many more of the benefits are intangibles compared to conventional
systems. For example, spatial information system applications tend to move up the
value chain of a firm as they are used for planning or decision support (see Section
III of this book). The value of the visualization aspects of spatial information sys-
tems is hard to quantify because we know little about how visual models improve
decision-making.
An important recent example of a spatial information system is the global position-
ing systems (GPS) in your car. I have one, and it changes the way that I find my
destination. Other new technologies associated with current spatial information
systems include RFID, mobile wireless, and server software that deliver spatial
information systems over the Internet. Enterprise applications, such as ORACLE,
now provided GIS add-ons.
The book presents a sampling of what is going on in spatial information systems from
a conceptual and application viewpoint. The book is divided into four sections:

I. Introduction
II. Challenges for Spatial Information Systems
III. Decision-Making Environments
IV. Future Trends and Technologies

Each of the sixteen chapters focus on a different application. Several themes do

recur. One is the use of Web services for geospatial applications. In the past, all
portions of a spatial information system resided in a single location. To undertake
a new project often required adding software at that location. The new idea, which
x

is gaining wide currency, is that the needed calculations can be subdivided into
pieces,4 each of which can be done by specialized programs, and that those programs
may reside at a number of locations. Thus, data is sent to a vendor who provides
the needed portion of the program, runs the calculation, and returns the computed
values as inputs to the next subroutine or module.
The book brings together work by a large number of researchers looking at applying
methods and techniques, including some that were developed for other purposes.
You will find discussions of ontologies, directories, semantic Webs, agent theory,
and robotics among others. In supporting decision-making, there are chapters on
snow removal operations, determining suitable habitats for endangered species,
and planning confined animal feeding operations. There are other intriguing ap-
plications such as:

• Using spatial information systems in teaching children content at the K-12

level by using a browser, without needing to teach them about the underlying
system
• Identifying community-based resources, to provide a database of existing
services, design the needed communications, identify missing resources, and
other societal implications
• Visualizing the distribution of rare plants and other endangered species by us-
ing sensor networks that bring together DBMS, spatial information systems,
and Web development technology

In short, the book leads you to start thinking about using spatial information sys-
tems in ways that almost none of us even conceived of only a few years ago. That
is progress.

Paul Gray
Professor Emeritus and Founding Chair
School of Information Systems and Technology
Claremont Graduate University
Claremont, CA 91711 USA
 x

Endnotes
1
Tufte, E. (1983). The visual display of quantitative information (Chap. 2).
Cheshire, CT: Graphics Press.
2
Goode, H. H., & Machol, R. E. (1957). Systems engineering. New York: Mc-
Graw-Hill.
3
Pick, J. B. (Ed.). (2005). Geographic information systems in business (p. 384).
Hershey, PA: Idea Group Publishing.
4
Sub-routines or modules in the old language of computing.

Paul.Gray is professor emeritus and founding chair of the School of Information Systems and Technol-
ogy at Claremont Graduate University, USA. Before coming to Claremont in 1983, he was a professor
at Stanford University, Georgia Tech, USC, and Southern Methodist. Prior to that, he worked for 18
years in research and development organizations, including 9 years at SRI International. He is the
author of three “first papers”: in crime in transportation, in telecommuting, and in group decision sup-
port systems. He is the author of over 130 journal articles and 13 books, including Manager’s Guide
to Making Decisions in Information Systems (Wiley, 2006). His honors include the LEO award for
lifetime achievement from AIS, a Fellow of both AIS and INFORMS, a winner of the NATO Systems
Science Prize, and Outstanding Information Systems Educator of 2000. He was president of TIMS in
1992-1993. He is the founding editor of Communications of AIS, serving from 1999 to 2005.
x

Preface

When a thing is new, people say: ‘It is not true.’ Later, when its truth
becomes obvious, they say: ‘It is not important.’ Finally, when its
importance cannot be denied, they say: ‘Anyway, it is not new.
~ William James, 1896

Several emerging phenomena and technologies, such as the increasing availability of

mature open source software and the continuing evolution of distributed computing,
are introducing a new dynamic into information system development. Specifically,
these phenomena and technologies are enabling the development of a variety of
innovative spatial information systems. This book contains several chapters that
present innovative spatial information systems that have been developed for a spe-
cific problem or decision-making situation. Also included are several chapters that
discuss key concepts and theories underlying current spatial information systems
as well as technology trends and emerging concepts that may impact spatial infor-
mation system development and applications in the future. Chapters are typically
presented as case studies, are grounded in information science research, and have
a specific practical application.
 x

Spatial. Informatics. and. Emerging. Technologies

Spatial informatics can be described as the sciences concerned with the collection,
manipulation, classification, analysis, storage, and retrieval of recorded spatial data,
information, and knowledge. These sciences are utilized in the development, man-
agement, and use of spatial information systems. Of particular interest to researchers
and practitioners alike, is the impact that emerging technologies may have on these
systems. Open source software and distributed computing (a brief overview of each
is presented below) are two of the many emerging phenomena and technologies that
are impacting the development of spatial information systems.

Open. Source. Software

Free and open source software is software that gives users the right to run, copy,
distribute, study, change, and improve it as they see fit, without the need to receive
permission from or make additional payments to any external group or person
(Bollinger, 2003). Perhaps two of the most well-known examples of open source
software are the operating system Linux and the Apache Web server. Linux is the
second most commonly-used operating system (Windows is number one) (Netcraft,
2001; The Economist, 2001) while Apache is the principal Web server in use today
(72% vs. Windows 22%) (Security Space, 2006).
Mature open source software solutions have expanded the number of acquisition
choices available to the organization beyond traditional closed source (commercial
or proprietary) software solutions. As a result, there now exists the possibility of
mixing open source and closed source solutions to best meet the information system
needs of the organization. There are four strategies (Bollinger, 2003) for mixing
open source and closed source software:

1... Distribution.mixing: Open source and closed source software can be stored
and transmitted together.
2... Execution.mixing: Open source and closed source software can run at the
same time on the same computer or network.
3... Application.mixing: Open source can rely on closed source software to pro-
vide it with services and visa versa.
4... Service.mixing: Open source can provide generic services to closed source
software and visa versa.
xv

Table 1. Comparison of open, mixed, and closed source software

Open.Source. Mixed.Source. Closed.Source.
Software Software Software
Operating System Linux Windows Windows
Web Server Apache IIS IIS
Database MySQL MySQL MSSQL
Scripting Language PHP PHP ASP

Table 1 illustrates the concept of open source/closed source software mixing in

relation to the typical Internet-based information system.
As illustrated in Table 1, this specific collection of open source software represents
the technologies that represent the open source Internet platform: LAMP = Linux
+ Apache + MySQL + (PHP | Perl | Python). The mixed source solution consists
of an operating system and Web server supplied by Microsoft, Inc.; the database
and scripting language are open source. The closed source solution is completely
proprietary and is based on additional products from Microsoft – MS SQL database
server and the Active Server Page scripting language. This is one example that il-
lustrates the impact of open source software on information system development;
in the chapters that follow, several more are presented.

Distributed. Computing

A distributed computing system consists of multiple software components on

multiple computers running as a single system (International Business Machines
Corporation, 2004). Furthermore, the computers in a distributed system can be
physically close together and connected by a local network, or they can be geo-
graphically distant and connected by a wide area network. A distributed system can
comprise any number of possible configurations: mainframes, personal computers,
workstations, minicomputers, and so forth. The goal of distributed computing is
to construct a system such that it appears and acts as a single computer. Numerous
emerging technologies are supporting the development of distributed computing
and applications. These include:

• Web. services have emerged as a popular standards-based framework for

accessing network applications. Web services consist of a set of messaging
protocols, programming standards, and network registration and discovery
 xv

facilities that expose business functions to authorized parties over the Internet
from any Web-connected device (Oracle Corporation, 2001). Basically, Web
services allow specific application logic to be exposed and used between
independent applications with a minimum knowledge of the Web service
and/or underlying application. The advent of Web services promises to let
organizations connect their applications to any number of other organizations
relatively inexpensively and easily (Hagel, 2002).
• Grid.computing has emerged as an important new field, distinguished from
conventional distributed computing by its focus on large-scale resource shar-
ing, innovative applications, and, in some cases, high-performance orientation
(Foster, Kesselman et al., 2001). Grid computing seeks to address many of the
problems associated with the continuing evolution of distributed computing,
in particular, controlled and coordinated resource sharing and use for prob-
lem-solving in dynamic, scalable, multi-institutional virtual organizations (a
set of individuals and/or institutions linked through resource sharing rules).
In this instance, resource sharing is not primarily file exchange, but direct ac-
cess to computers, software, data, and other resources, that is, those resources
required by the range of collaborative, problem-solving, and resource brokering
strategies that are emerging in industry, science, and engineering. Resource
sharing in this situation is highly controlled and coordinated, with resource
providers and consumers clearly defining the rules regarding what is shared,
who is allowed to share it, and the conditions under which sharing occurs.
• The.portable.Internet is a platform for high-speed data access using Internet
protocol (IP) and includes advanced short-range wireless technologies (within
30 meters) such as Bluetooth, medium-range wireless technologies (at least
150 meters) such as WiFi, and long-range wireless technologies (up to 50 ki-
lometers) such as WiMAX as well as several advanced techniques that make
more efficient use of the available spectrum, including spread spectrum, smart
antennae, agile radios, and mesh networks. A wireless IP platform can be used
to carry not only high-speed services, such as video entertainment and data
transfer, but also medium-speed services, such as Web-browsing, and low-
speed services such as voice or e-mail. As such, it is potentially substitutable
over a wide-range of existing networks and services, and could impact a large
number of current business models (International Telecommunications Union,
2004).

The emerging spatial information systems and applications that appear in the fol-
lowing chapters, encompass, in some manner, aspects of the open source software
and distributed computing technologies described above.
xv

Book. Organization

Section. I:. Introduction

This section presents information regarding key concepts and theories underlying
current spatial information systems. For instance, Chapter.I, by Zhao, Yu, and Di,
introduces all aspects of geospatial Web services from service-oriented architecture
to service implementation. It covers the life cycle of geospatial Web services in terms
of geospatial interoperable standards, including publish, discovery, invocation, and
orchestration. Semantic issues regarding geospatial data and services are discussed,
and the applications of standard-compliant geospatial Web services are reviewed.
Chapter.II, by Hilton, Burkhard, and Abhichandani, presents an approach to an
ontology-based information system design theory for spatial information system
development. This approach addresses the dynamic nature of information system
development at the beginning of the 21st century and addresses the question of
how to establish relationships between the various design components of a spatial
information system. An example of this approach is presented, along with examples
of the various ontologies utilized in the design of this particular spatial information
system.

Section. II:. Challenges. for. Spatial. Information. Systems

This section presents those innovative spatial information systems that have been
developed for a specific problem. As seen in Chapter.III, Judith Woodhall.dis-
cusses how the need for geospatially-enabled data messaging among emergency
response agencies can be enabled with the emergency provider access directory
(EPAD). She describes the directory, how it enables message routing, and its fit
into a boarder E-Safety network. She also discusses the architectural components
of the EPAD, specifically the geographic information system module, and how Web
services and open source products were used in the design to enhance the EPAD
service offering.
Chapter. IV, by Gunjan Kalra, discusses the process of providing information
in its most accurate, complete form to its users, and the difficulties faced by the
users of the current information systems. She describes the impact of prevalent
technologies such as the multi-agent systems and the Semantic Web in the area of
information supply via an example implementation and a model use case. She also
offers a potentially more efficient and robust approach to information integration
and supply process.
Ku and Zimmermann, in Chapter.V, present an information architecture using Web
services for exchanging and utilizing geotechnical information, which is of critical
 xv

interest to a large number of municipal, state, and federal agencies as well as private
enterprises involved with civil infrastructures. They propose an infrastructure of
Web services, which handles geotechnical data via an XML format, report on its
design, and share some initial experiences.
Chapter.VI, by June K. Hilton and David E. Drew, discusses ScienceMaps, an
online resource portal for standards-based science instruction using GIS technology.
ScienceMaps is unique in that it concentrates on using GIS to teach, not on teaching
GIS. Using an Internet-based GIS, ScienceMaps provides access to GIS technology
and data to anyone, anywhere, with access to an Internet browser.

Section. III:. Decision-Making. Environments

This section examines those spatial information systems that have been developed
for a specific decision-making situation or environment. Gao and Sundaram, in
Chapter.VII, draw from several relevant disciplines to overcome the problems
identified in various areas of spatial decision support and propose a generic spatial
decision-making process and a domain-independent spatial decision support system
(SDSS) framework and architecture to support this process. They develop a flexible
SDSS to demonstrate an environment in which decision-makers can utilize various
tools and explore different scenarios to derive a decision.
As seen in Chapter. VIII, Sugumaran, Ilavajhala, and Sugumaran discuss the
development of an intelligent Web-based spatial decision support system and dem-
onstrate it with a case study for planning snow removal operations. They illustrate
how traditional decision support system (DSS) and Web-based spatial DSS can be
further improved by integrating expert knowledge and utilizing intelligent software
components (such as expert systems and intelligent agents) to emulate the human
intelligence and decision-making.
In Chapter.IX, Todd G. Olson and Brian N. Hilton discuss Conservation Studio,
a spatial information system that automates the entire process of conservation
modeling, simulation, and planning. Conservation Studio consists of four software
modules: Data Acquisition Interface, Habitat Suitability Analyst, Conservation
Criteria Developer, and Implementation Modeler, the latter of which models the
outcome of using tradable conservation credits to conserve habitat resources in a
specified geographical plan area.
Chapter.X, by Sugumaran and Bakker, discusses the need for the development of
a decision support system to assist in the selection of an appropriate location for
the development of future confined animal feeding operations (CAFO) structures.
Furthermore, it presents the development of a decision support tool to aid CAFO
managers and producers in selecting appropriate locations for animal confinements
using geographic information system technology and CAFO regulations in Iowa.
xv

Section. IV:. Future. Trends. and. Technologies

This section highlights technology trends and emerging concepts and considers how
they may impact spatial information system development and/or applications in the
future. In Chapter.XI, Lars Brodersen and Anders Nielsen, present the relationships
and impacts between the various components of the spatial data infrastructure (SDI)
and geo-communication. They also discuss a model for the organization of the pas-
sive components of the infrastructure, that is, legislation, collaboration, standards,
models, specifications, Web services, and information.
Fengxian Fan, in Chapter.XII, explores an implementation to process and interpret
the data gathered by wireless sensor networks deployed to monitor rare plants and
other endangered species. The system she presents in this chapter combines database
management technology, geographic information system, and Web development
technology to visualize the data gathered by these wireless sensor networks.
Chapter.XIII, by Ibach, Malek, and Tamm, reviews the enabling technologies that
drive system development and also discusses market factors, security and privacy
concerns, and standardization processes that need to be taken into account concern-
ing the “global real-time enterprise”. The SEMALON (SEMAntic LOcation Net-
work) approach is proposed as a basic infrastructure for discovery and composition
of location-based services. A case study implementation for NOMADS Campus,
a distributed spatial information system on the campus at Humboldt University
Berlin, is presented.
Tolone, Xiang, Raja, Wilson, Tang, McWilliams, and McNally, in Chapter. IV,
propose a knowledge-driven methodology that facilitates the extraction of critical
infrastructure (CI) information from public domain, that is, open source, municipal
data sets. The proposed methodology was tested successfully on a municipality
in the Southeastern United States and is considered to be a viable choice for CIP
professionals in their efforts to gather CI information for scenario composition and
vulnerability assessment.
In Chapter.XV, Lyn Kathlene describes and analyzes the effectiveness of two meth-
odological techniques, cognitive mapping and geographical information systems
(GIS), for identifying social service resources. She also examines the processes used
to integrate hand-drawn map information into geo-coded data points and provides
recommendations for improving efficiency and precision.
In the final chapter, Chapter.XVI, by Bruemmer, Few, and Nielsen, research to
study and improve an operator’s ability to navigate or tele-operate a robot that is
distant from the operator through the use of a robot intelligence architecture and a
virtual 3D interface is presented. Their results suggest that performance is improved
when the robot assumes some of the navigational responsibilities or the interface
presents spatial information as it relates to the pose of the robot in the remote en-
vironment.
 xx

References

Bollinger, T. (2003). Use of Free and Open Source Software (FOSS) in the U.S.
Department of Defense. The MITRE Corporation.
Foster, I., Kesselman, C., et al. (2001). The anatomy of the grid: Enabling scalable
virtual organizations. International Journal of Supercomputer Applications,
15(3).
Hagel, J. (2002). The strategic value of Web services. New York: McKinsey and
Company.
International Business Machines Corporation. (2004). TXSeries™ for Multiplat-
formsConcepts and Planning Version 5.1. Armonk, NY: International Business
Machines Corporation.
International Telecommunications Union. (2004). ITU Internet reports: The portable
Internet. Geneva, Switzerland: International Telecommunications Union.
Netcraft (2001). Netcraft Web Server Survey — June 2001.
Oracle Corporation (2001). Oracle 9i Application Server: Web Services Technical
White Paper. Redwood Shores, CA: Oracle Corporation.
Security Space. (2006, January 1). Web Server Survey.
The Economist (2001). Survey of software: Out in the open. The Economist.
xx

Acknowledgments

I would like to thank the chapter authors for their hard work and fine
contributions. I would also like to thank those who participated in the
review process, both contributors as well as outside reviewers.

I would also like to acknowledge the faculty and staff in the School of
Information Systems and Technology at Claremont Graduate University.
I appreciated your support and assistance throughout this endeavor.
 xx

Section I

Introduction
xx
 Geospatal Web Servces 

Chapter.I

Geospatial.Web.Services
Pesheng Zhao, George Mason Unversty, USA
Genong Yu, George Mason Unversty, USA
Lpng D, George Mason Unversty, USA

Abstract

As Web service technologies mature in recent years, a growing number of geospatial

Web services designed to interoperate spatial information over the network have
emerged. Geospatial Web services are changing the way in which spatial informa-
tion systems and applications are designed, developed, and deployed. This chapter
introduces all aspects of geospatial Web services from service-oriented architecture
to service implementation. It covers the life cycle of geospatial Web services in terms
of geospatial interoperable standards, including publish, discovery, invocation, and
orchestration. To make geospatial Web services more intelligent, semantic issues
about geospatial data and services are discussed here. Furthermore, the applica-
tions of standard-compliant geospatial Web services are also reviewed.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Introduction

Web service technology promises standard-based information interoperability. There

are many different definitions for Web services (Andersson, Greenspun, & Grumet,
2003; Booth et al., 2004; Hirsch & Just, 2003; Mateos, Zunino, & Campo, 2005;
Skonnard, 2002; Vaughan-Nichols, 2002). In essence, a Web service is a modular,
self-describing, and self-contained software application which is discoverable and
accessible through standard interfaces over the network (Tsalgatidou & Pilioura,
2002).
The core technology associated with Web service is the standardization of data/mes-
sage exchange between applications or systems during every stage of their life cycle,
including transporting, invoking, and discovering (Akinci, 2004; Di, 2004a; Hecht,
2002). XML (eXtensible Markup Language) is used as the primary language to en-
code data/message in Web services since it hides the details of underlying transport
protocols and provides a platform-independent structured information format. Struc-
tured information can be exchanged using standard protocols, such as simple object
access protocol (SOAP), or XML-RPC (Gudgin, Hadley, Mendelsohn, Moreau, &
Nielsen, 2003; Winer, 1999). The public interface (functionality and input/output
parameters) of a Web service is described following a machine-processable format,
such as Web Service Description Language (WSDL) (Booth & Liu, 2005; Chinnici,
Haas, et al., 2005; Chinnici, Moreau, Ryman, & Weerawarana, 2005; Christensen,
Curbera, Meredith, & Weerawarana, 2001; Vedamuthu, 2005). A standard registry
or catalog is often used to publish and discover these Web services, such as UDDI
(Universal Description, Discovery and Integration) (Booth et al., 2004; Clement,
Hately, Riegen, & Rogers, 2004). These characters distinguish a Web service from
traditional proprietary distributed systems, such as distributed common object
model (DCOM) by Microsoft, java remote method invocation API (RMI) by Sun,
and common object request broker architecture (CORBA) by Object Management
Group (OMG).
The major benefit of Web services is the interoperability enabled by those standards;
in other words, Web services are capable of collaborating process control and shar-
ing data and information across applications over different platforms (Di, 2005; Di,
Zhao, Yang, Yu, & Yue, 2005; Kralidis, 2005). A Web service hides all the details
of implementation under a well-defined interface, and thus other applications or
services can invoke such a Web service through the standard interface. Such type of
interoperation is not just limited within one organization, but also can be conducted
across organizations. From the technical point of view, the advantages of using Web
services can be summarized as: (1) enabling the sharing of computational resources
(hardware, software, and data/information holdings) across the organization boundary;
(2) easy to maintain and wrap legacy system since the modularity of Web service
allows the partial updating and change to existing systems; (3) independent from

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

platforms and operating systems since Web services interact with clients or other
Web services through standard messages; and (4) independent from programming
languages and implementations as long as the claimed interfaces are obeyed (Akinci,
2004; Di, 2004a; Kralidis, 2005; Zhao et al., 2004). Other advantages are: real-time
data access, on-demand/customized information delivery, value-added composition,
easy integration, and extended collaboration (Kralidis, 2005). These benefits are
achieved at the slightest cost of performance compared to proprietary distributed
computing approaches. Overall, Web services are more suitable for handling huge-
amount, distributed, and diverse geospatial data/information as compared to the
conventional approach of geospatial data dissemination.
Traditionally, geospatial information is delivered to end users through media such
as magnetic tapes (e.g., 8mm storage tape, computer compatible tape), optical disks
(e.g., CD-ROM, DVD), and electronic transfer (e.g., file transfer protocol, hypertext
transfer protocol, simple mail transfer protocol). Often a catalog is provided for
searching the data in granules or some pre-determined units, and a matched list of
geographic information is given for downloading or mailing. The problems with such
approaches to deliver geospatial information are numerous (Di & McDonald, 2005;
Di, Yang, Deng, Deng, & McDonald, 2002; Kralidis, 2005). Firstly, the geospatial
information is delivered in file-based format and is not customizable. Geospatial
information may be packaged in a provider-defined size, which potentially lead to
overload of network (with larger dataset than user desires) or to under-sized data
(with many small datasets to meet user’s requirement that leave a lot of work on
converting and merging these datasets). Geospatial information may not be updated
timely with the most current status as the data change at the provider’s site. Delay of
obtaining data may not be desirable when the study is very time-sensitive. Secondly,
the process to obtain geographic information is labor-intensive. With the traditional
approach, a user may spend a lot of time in many steps: (1) identifying data source
and its catalog, (2) interactively searching the dataset with multiple attempts to
narrow down the desired dataset, (3) downloading the data or waiting for the data
media to be delivered, (4) converting and importing the data into the system the user
uses, and (5) processing the data and presenting the results. All of these processes
cannot be automated or prescribed with batch-processing plans. Thirdly, geospatial
information is not interoperable. The information may be stored in vendor’s propri-
etary formats, and a specific processing has to be taken care of before it can be used
in user’s system. The messages communicated between different vendors’ systems
always cannot be directly processed by other machines or systems. Finally, it is very
difficult to share the value-added geospatial information. A user may be in need of
a processing function that the provided software package does not possess and then
be pushed to purchase a suite of software to get the specific information. All these
issues with the traditional geospatial information dissemination and processing are
what Web services promise to solve, by leveraging Web service capabilities: (1)
modularity for the sharing of specific function, (2) standard interfaces for the in-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

teroperability of applications, (3) on-demand capability for the customized, real-time

value-added geospatial information delivery, and (4) self-contained functionality
for the loosely-coupled operation on dataset. The marriage of Web service with
geospatial information leads to geospatial Web services.
Generally, a geospatial Web service can be viewed as a modular Web application
that provides services on geospatial data, information, or knowledge (Di, Zhao,
Yang, Yu, & Yue, 2005). It refers to the use of Web services technologies to manage,
analyze, and distribute spatial information. Furthermore, a geospatial Web service
can be sorted and searched through its geospatial characteristics, such as location,
area, neighborhood, and other spatial features. As a Web service, a geospatial Web
service involves three actors: user entity (consumer), provider entity (provider), and
register entity (broker) (Booth et al., 2004). Basic operations during the life cycle
of a geospatial Web service include publication, discovery, binding, invoking, and
execution. The interfaces of some geospatial Web services have been standard-
ized. The most important players for such standardization efforts are ISO/TC211
and Open Geospatial Consortium (OGC). Because of the influential specifications
from OGC, geospatial Web services have an add-on unique geospatial flavor on
the message-communicating other than those popular W3C Web services standards
such as WSDL, SOAP and UDDI. Geospatial Web services focus on standard in-
terface directly through HTTP protocol, which is now recognized as more related
to representation state transfer (REST) (Booth et al., 2004; Fielding, 2000). Further
details about the OGC standards will be discussed in “Standards for Geospatial
Web Services.” Geospatial Web services are changing the way in which spatial
information systems and applications are designed, developed, and deployed. The
added characters give a flexible extension of functionality required to enable the
interoperable operation of geospatial data and their processing.
In this chapter, following the lead of standards and the ultimate goal of interoper-
ability, we will cover all aspects of geospatial Web services from conceptual level to
technology level. In “Standards for Geospatial Web Services,” the interoperability
standards, especially those from ISO and OGC, will be introduced. In “Geospatial
Web Service Implementation,” examples will be given to illustrate the life cycle
of Geospatial Web services and guide readers how to design, create, deploy, and
execute Geospatial Web services. “Geospatial Web Service Discovery” discusses
the discovery of geospatial Web service with three approaches — registry, index,
and peer-to-peer. Registry, or service catalog, will be the main focus. “Geospatial
Web Service Orchestration” presents the composition of Web services or assem-
bling individual services into a more useful process flow. The section “Geospatial
Web Service Solutions” introduces some Geospatial Web service solutions that are
widespread in use. “Semantic Issues on Geospatial Web Service” touches the most
advanced aspects of Geospatial Web services, semantics. Finally, a short summary
and future direction are given.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Standards. for. Geospatial. Web. Services

Geospatial interoperability is achieved through the use of standards by different

standard-setting organizations, including government organizations (e.g., FGDC),
international standard bodies (e.g., ISO), and industry associations (e.g., OGC) (Groot
& McLaughlin, 2000; Kralidis, 2005). Major bodies for standards of geospatial
Web services are ISO/TC211, OGC, FGDC/NSDI (EOP, 2003; Evans & Bambacus,
2005; ISO/TC211, 2005; OGC, 2005a). The general Web service standard stacks
can be partitioned into three parts: data (message encoding), interface (transport
protocol), and metadata (ontology), which cover all the aspects of interoperations
(Alameh, 2001; Nebert, 2005). These are similar in geospatial Web services. At the
data level, the standards specify the message encoding and data formatting that are
used for communicating between Web services and applications. At the interface
level, the standards define common interfaces for both applications/Web services
and human users. At the metadata level, a set of consensus data types and descrip-
tions are associated with each Web service or data.
One should notice that OGC geospatial Web services are different from Web ser-
vices in the e-business world. OGC geospatial Web services have been developed
in parallel with the evolution of W3C and OASIS Web services. Standards from
W3C and OASIS, such as WSDL, SOAP, and UDDI, are the standards for Web
services in the e-business world. The OGC geospatial Web services do not comply
with these Web service standards. Recently, OGC started to explore the possibilities
in bridging the gaps and implementing OGC Web services using W3C/OASIS Web
services (Duschene & Sonnet, 2005a, 2005b; Lieberman, Reich, & Vretanos, 2003;
Sonnet, 2004; Sonnet & Savage, 2003). Another move in standardization of geospatial
Web services is the exploration of the semantic Web. OGC initiated the geospatial
semantic Web interoperability experiment in April, 2005 (Bacharach, 2005).

Web.Service.Standard.Stacks

Web services are an emerging technology for which standards are currently in great
need. Many organizations are participating in the standardization of data, interfaces,
and metadata, such as World Wide Web Consortium (W3C), Organization for the
Advancement of Structured Information Standards (OASIS), and Internet Engineer-
ing Task Force (IETF) (Alameh, 2001; IETF, 2005; Kralidis, 2005; Nebert, 2005;
OASIS, 2005; W3C, 2005b; Wilkes, 2005). Figure 1 shows the protocols for Web
services that are either widely-adopted or in development. The bold-faced proto-
cols formed the de facto core standards for Web services. XML is the underlying
language used to encode messages, metadata, schema, and interfaces. XSD (XML
schema) replaces document type definition (DTD) to define information structure
(W3C, 2005a). SOAP is a commonly-used message encoding protocol (Gudgin,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Figure 1. Web service protocol stack (Source: Tsalgatidou & Pilioura, 2002; Wilkes,
2005)

Hadley, Mendelsohn, Moreau, & Nielsen, 2003). WSDL is used for describing each
Web service, including input/output, operations, and underlying transport protocols
(Christensen, Curbera, Meredith, & Weerawarana, 2001). UDDI is a protocol for
building a directory of Web services (Clement, Hately, Riegen, & Rogers, 2004).
Other protocols are mostly in the stage of early adoption, experimentation, or speci-
fication proposal (Wilkes, 2005).

Geospatial.Web.Service.Standard.Stacks

OGC and ISO/TC211 are the major players in standardizing geospatial Web services.
OGC specifications focus on developing implementation standards while ISO/TC
211 is concentrating on developing theoretical/abstract standards. ISO/TC 211
standards specify methods, tools, and services for acquiring, processing, analyz-
ing, accessing, presenting, and transferring spatial information between different
users, systems, and locations. OGC specifications support the full integration of
“geo-enabled” Web services into mainstream computing to make complex spa-
tial information and services accessible and useful with all kinds of applications.
Through the cooperation between ISO TC 211 and OGC, most of approved OGC
implementation standards are either already ISO standards or in the process to
become the ISO standards (Di, 2003).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Figure 2. Geospatial Web service standard stack (Source: ISO/TC211, 2005; Nebert,
2005; OGC, 2005a)

Since 1999, OGC has finished phase 1, 2, and 3 of Open Web Services (OWS)
initiatives in 2002, 2004, and 2005 that addresses geospatial interoperability re-
quirements and standards to enhance the discovery, retrieval, and use of geospatial
information and geoprocessing services. Under these OGC initiatives, a series of
geospatial Web services specifications have been published, such as Web cover-
age service (WCS) (Evans, 2003), Web feature service (WFS) (Vretanos, 2005),
Web map service (WMS) (de La Beaujardiere, 2004), Web image classification

Figure 3. Overall architecture of OGC geospatial Web services (Source: Nebert,

2005)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

service (WICS) (Yang & Whiteside, 2005), Web coordinate transformation service
(WCTS) (Whiteside, Müller, Fellah, & Warmerdam, 2005). Those specifications
are the foundation for geospatial interoperability (Di, 2005b). Figure 2 shows most
standards and specifications for geospatial Web services.
Similar to the standards for general Web services, these standards and specifications
can also be grouped into three groups: data, interface, and metadata. They play
different roles in the overall architecture of OGC geospatial Web service. The data
standards specify the storage file format of geospatial data. The interface specifica-
tions enable the interactions between geospatial Web services. The metadata are
used to describe geospatial data and geospatial Web services. Figure 3 shows the
overall architecture of OGC geospatial Web services.

Data Standards

The standardization for geospatial data formats sets a good basis for geospatial
Web services to communicate with each other in an understandable format. These
efforts have resulted in many specifications and standards for data storage and ex-
change, such as digital line graph (DLG), digital raster graph (DRG), spatial data
transfer standard (SDTS), and hierarchical data format for Earth observation system
(HDFEOS) (Alameh, 2001; GSFC, 2005). In compliance with ISO 19118 for the
transport and storage of geospatial information, the OGC Geography Markup Lan-
guage (GML) provides an open, portable, and vendor-neural framework to define
geographic features and datasets (Cox, Daisey, Lake, Portele, & Whiteside, 2004).
GML uses XML text to encode geometry and properties of geographic feature, spa-
tial reference system, and feature collections. The GML also absorbs great graphic
quality of scalable vector graphics (SVG). These data formats are commonly used
for the delivery of geospatial data with geospatial Web services.

Interface Standards

OGC has been dedicated to the standardization of interfaces of geospatial Web ser-
vices to enable the interoperability. All OGC Web services are based on a common
model and share some unified characteristics: (1) some common operation request
and response contents (e.g., getCapabilities operation), (2) some unified parameters
for operation requests and responses, and (3) exclusive adoption of XML and key-
value-pair (KVP) in encoding (Whiteside, 2005).
Under such a common specification, a series of specifications were developed and
adopted for delivering and analyzing geospatial data over the Web. CAT defines
a catalog service for discovery, browsing, and querying of distributed and hetero-
geneous catalog servers (Nebert & Whiteside, 2004). Coordinate transformation

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

service (CT) specifies interfaces for positioning, coordinate systems, and their
transformation (Daly, 2001). WCS defines common interfaces for grid analysis and
processing (Burry, 2001; Di, 2005b). WFS uses GML to encode geographic features
and defines common interfaces to handle geographic vector information (Vretanos,
2005). WMS enables the display of registered and superimposed map-like views of
geographic information in some image formats (e.g., JPEG) (de La Beaujardiere,
2004). Web map context (WMC) supports the creation and manipulation of a
context map that consists of several WMS maps (Sonnet, 2005). Location service
(OpenLS) describes an open platform for location-based application servers and
outlines related activities (Mabrouk, 2005).

Metadata Standards

To enable human interpretation and machine processing of geospatial data and geo-
spatial Web services, another important standardization endeavor is the definition of
a common metadata (description about data or services). For geographic informa-
tion, ISO19115 provides abstract guidelines for geographic information metadata
and ISO19119 for geospatial services metadata (ISO/TC211, 2002b, 2003a). ISO
19139 defines the XML schema for ISO19115 (ISO/TC211, 2003b).
ISO 19115 defines metadata for geospatial data product in fourteen parts: (1) meta-
data entity set (MD_Metadata) is the mandatory part that includes identification,
constraints, data quality, maintenance, spatial representation, reference system,
content, portrayal catalog, distribution, metadata extension, and application schema;
(2) identification (MD_Identification) uniquely identifies the data by defining format,
graphic overview, specific uses, constraints, keywords, maintenance and aggregate
information; (3) constraint (MD_Constraints) defines the restrictions placed on the
data; (4) data quality (DQ_DataQuality) contains quality of the dataset and infor-
mation about the sources and production processes; (5) maintenance (MD_Main-
tenanceInformation) describes the scope and frequency of updating; (6) spatial
representation (MD_SpatialRepresentation) points out the mechanism to represent
spatial information; (7) reference system (MD_ReferenceSystem) describes spatial
and temporal reference system; (8) content (MD_ContentInformation) identifies the
feature catalog; (9) portrayal catalog (MD_PortrayalCatalogReference) gives the
type for displaying data; (10) distribution (MD_Distribution) describes the distribu-
tor of the data; (11) metadata extension (MD_MetadataExtentionInformaiton) is for
user-specified extensions; (12) application schema (MD_ApplicationSchemaInfor-
mation) is for the schema used to build a dataset; (13) extent (EX_Extent) describes
the spatial and temporal extent; and (14) citation and responsible party (CI_Citation
and CI_ResponsibleParty) provides citation information.
ISO 19119 presents taxonomy of geospatial services and gives a list of example
geospatial services in each service category. Major geographic services are: (1) geo-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Zhao, Yu, & D

graphic human interaction, for example, catalog viewer, geographic feature editor,
(2) geographic model/information management, for example, feature access, map
access, catalog service, (3) geographic workflow/task management, for example,
chain definition service, workflow enactment service, (4) geographic processing
(spatial, thematic, temporal, and metadata), for example, coordinate conversion
service, thematic classification service, temporal reference system transformation
service, statistical calculation service, (5) geographic communication, for example,
encoding service, transfer service, and (6) geographic system management.

Geospatial. Web. Service. Implementation

There are many tools available for the implementation and deployment of Web
services. Theoretically, any programming languages can be used in developing
Web services.

Web.Service.Implementation

Web service, technically and typically, uses WSDL to describe interfaces, encodes
communicate messages using SOAP, and publishes itself in a UDDI registry. A typi-
cal process to create a Web service from scratch: (1) starts with the WSDL design
using Unified Modeling Language (UML) or other visual design tools, (2) decides
the message styles (either RPC encoded/literal or document encoded/literal), (3)
develops the program, and (4) ends with the publication and testing of the Web
services in a SOAP container (Peltz, 2003).
Another approach is to adapt the existing programs and expose some of their func-
tions as Web services (Peltz, 2003). The task to deploy such Web services mainly
involves the creation of adapters that make the legacy system compatible with the
Web services. For example, a developer may create an HTTP adapter to convert
the input and output between SOAP and HTTP messages, or may develop a Java
adapter to call the C/C++ component through Java native interface (JNI) if the pub-
lished Web service container is a Java server page container (e.g., Axis for Java).
Other design patterns may be considered to efficiently re-use the Web services and
balance the loading of network traffic, such as MVC (model-view-controller), and
asynchronous messaging (Peltz, 2003).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Geospatial.Web.Service.Implementation

In the implementation of geospatial Web services, similar approaches can be ap-

plied. In a narrower sense, geospatial Web services are limited to those following
geospatial standards, such as OGC WFS, WCS, and WICS. In a broader sense,
geospatial Web services embrace Web services that handle geospatial data and pro-
vide services to processing geospatial data. Examples are the Web services adapted
from geographic resource analysis support system (GRASS) at LAITS (GRASS,
2005; LAITS, 2005).

Implement Standard Geospatial Web Services

A typical implementation practice for OGC geospatial Web services may start with
the analysis of the geospatial data to be served and determine the proper interfaces
to be implemented, especially those optional operations and parameters. Secondly,
all the internal functions should be coded and tested modularly. Thirdly, these re-
quired interfaces (operations) and their parameters should be exposed by abiding the
relevant specifications. Finally, test and debugging should be carried out. A series
of quality assurance and performance evaluation should be conducted (Di, 2004b;
Kolodziej, 2004; Zhao et al., 2004).
It is essential to implement the specific interfaces for developing OGC Web services.
Some of the interfaces are mandatory. Others are optional. For example, the opera-
tions of GetCapabilities, and GetMap in WMS are mandatory, and its GetFeatureInfo
operation is optional (de La Beaujardiere, 2004; Kolodziej, 2004). Furthermore,
the parameters for each operation are partly required and partly optional. For ex-
ample, the GetMap operation of a WMS require parameters of “VERSION=1.3.0”,
“REQUEST=GetMap”, “LAYER=?”, “STYLES=?”, “CRS=?”, “BBOX=?,?,?,?”,
“WIDTH=?”, “HEIGHT=?”, and “FORMAT=?”. Other KVP (key value pairs)
are optional, such as “TRANSPARENT=TRUE|FALSE”, “BGCOLOR=?”, and
“EXCEPTIONS=?”.
Currently, OGC Web service supports only HTTP GET and/or HTTP PUT binding
at the transport level by default. This is one aspect that differs from the W3C/OA-
SIS Web services. OGC is trying to bridge this gap by making some experiments
on SOAP binding. OGC Web services can be published through common gateway
interface (CGI) (a standard to interact with external applications through an infor-
mation servers, such as HTTP or Web servers), servlets (a Java-enabled server),
active server pages (ASP) (a Microsoft-based approach), Java server pages (JSP)
(a Java-enabled approach), and ASP.NET (a Web page service upon a Microsoft
.NET framework).
When there are existing routines or Web services, the adaptation of these geopro-
cessing services or functional modules to comply with OGC specifications may be
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

required. There are three possible cases: (1) to adapt a Web service, an OGC-com-
pliant translator (using XSL/XSLT) may be sufficient since the encoding is already
XML-based, (2) to adapt a server program, extra operations may be required to
meet the specifications, and (3) to adapt a client, corresponding operations may be
implemented to access the OGC-compliant Web services (Kolodziej, 2004; Zhao
et al., 2004; Zhao, Nusser, & Miller, 2002).

Implement Customized Geospatial Web Services

In most cases, we may still be required to deploy some specific geospatial Web
services that complete certain geospatial operations. If so, the developers may need
some thinking on determining which operations are to be published, how to manage
the intermediary and temporary file spaces, how to achieve the best performance in
a distributed environment, and how to secure the system at large. This is especially
true for these legacy geographic information systems. Their rich functions may be
demanded in a service-oriented environment. Here we use one case that we adapted
the GRASS functions and publish most of the operations as geospatial Web services at
LAITS. These services are required for the effective demonstration on the intelligent
chaining of Web services using a large repository of geospatial Web services.
GRASS has a huge number of operations. It is a comprehensive system of more than
350 programs for raster/vector geographic data manipulation, image processing, and
map production. The functions range over spatial analysis, map generation (on paper
and monitor), data visualization (2-dimensions, 2.5-dimentions, and 3-dimensions),
data modeling, database management, and data manipulation (GRASS, 2005). In
order to make GRASS functions as Web services, the first issue is to transfer these
functions from a desktop environment to a networked environment. As a stand-
alone system, the GRASS starts with a series of global variable initialization and
keeps several states over the execution of programs. These need to be adapted to
handle multiple routines running simultaneously. This can be solved by wrapping
the initialization with each function to be explored, or every function would start
with a set of own initialization. The second issue is what functions are proper to be
deployed as Web services, and what is the basic level of operations to be exposed,
command level or combined script. The command level is an individual program in
GRASS to complete a small operation, such as Fast Fourier Transform, statistical
summation, and optimum-index-factor table calculation. The combined script may
complete a series of operations by editing and executing a script file of multiple
steps of commands. The solutions are: (1) The Web service should be mainly based
on the command level as this is the lowest level of function units. At this level, the
creation of Web services from each command is straightforward, so as to ease the
maintenance when new functions are added into GRASS. Combined operation can
be easily achieved by combining several Web services through workflow manager,
such as BPEL engines; and (2) Some commands are not exposed as Web services,
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

especially those purely dealing with disk storage and monitor display. Finally, the
use of customized geospatial Web services by an OGC-compliant Web service or
an OGC-compliant client is a big issue. The input and output geospatial data are
not exposed through an OGC-compliant data services. They are merely a universal
resource identification (URI). This can be resolved by providing a series of wrap
services to enable the feeding of geospatial data from an OGC-compliant data- pro-
viding service and the delivering of geospatial data to an OGC-compliant data server.
By doing so, the intermediate processes are invisible to OGC-compliant geospatial
services, but they can be easily handled by following Web service standards.

Geospatial. Web. Service. Discovery

One of the first steps in consuming geospatial Web services is to find the desired
geospatial Web services. Web service discovery is to locate a machine-processable
description of a Web service that meets desired functional criteria. There are mainly
three types of approach: Registry, Index, and Peer-to-Peer (Booth et al., 2004). Simi-
larly, geospatial Web services can be searched through these three approaches.

Search.by.Registry

A registry is a catalog or a directory that is authoritative on the description of each

registered Web service or metadata. The “catalog” plays a “directory” role: Pro-
viders advertise the availability of their resources through metadata, and users can
then query the metadata to discover interesting resources and determine how to
access those resources at run time (Nebert & Whiteside, 2004; Zhao et al., 2004).
The success of a registry search highly depends upon what metadata is recorded
for each geospatial Web service and how they are managed. If limited metadata
information is registered, a search may give an explosive number of returns because
of the broad matching. Heavy requirements on the metadata information would be
a burden for the provider to submit all the details. There are many efforts in build-
ing the core metadata that are necessary for describing a geospatial Web services.
Leading standards are ISO19115 for data and ISO19119 for service (ISO/TC211,
2002a, 2002b, 2003a).
Internally, a registry can be organized into a database, either a relational database
system or an XML-based database. Once all the metadata information is entered
into the registry, users or applications can query the registry with certain constraints
clauses or criteria and retrieve matched service descriptions. Figure 4 illustrates
major actors who interact with the registry and main operations which the actors
perform on the registry.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Figure 4. Geospatial Web Registry Server

In a distributed infrastructure, the registry is presented as a Web service which

targets at interoperation with client applications or other Web services (Alameh,
2001). In such a case, certain interoperable interfaces should be followed for the
design of the registry service. UDDI is a standard registry to manage Web services
with simple object access protocol (SOAP) as messaging protocol and Web Service

Figure 5. A simplified illustration of ebXML information model (Source: OASIS/

ebXML, 2003)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Description Language (WSDL) as Web service description language. And the OGC
Catalog Service for Web (CSW) is a standard specification specifically designed for
cataloging geospatial Web services (Martell, 2004; Voges & Zenkler, 2005; Wei et
al., 2005; Zhao et al., 2004).
The information model of CSW is based on ebXML information model (ebRIM)
(see Figure 5) (OASIS/ebXML, 2003). Table 1 lists classes used in this informa-
tion model. The ebRIM is designed to be extensible through several mechanisms,
either creating an ExtrinsicObject or adding a Slot. Therefore, CSW can easily be
extended to describe different objects, such as organization, data, and services.
For geospatial data and geospatial Web services, it is logical to extend the internal
description capability of an CSW registry server with metadata described by ISO
19115 and ISO 19119 (Wei et al., 2005).

Table 1. Classes in an ebRIM

Class Description
RegistryObject The base object for the registry
RepositoryItem Any object in a repository for storage and safekeeping
Slot Arbitrary attributes to the RegistryObject
Association Define the many-to-many relationship between objects in the information
model
ExternalIdentifier Additional identifier information to a RegistryObject
ExternalLink An URI to external RegistryObject
ClassificationScheme A scheme to classify RegisttryObject instances
ClassificationNode A child node to a ClassificationScheme
Classification A taxonomy value in a classification scheme
RegistryPackage A group of Registry
AuditableEvent An audit trail for RegistryObject instances
User A registered user
PostalAddress A postal address
E-mailAddress An e-mail address
Organization For example, submitting organization
Service Service information
ServiceBinding Technical information on a specific way to access a specific interface of a
Service
SpecificationLink A link referring to the technical specification that describes how to use the
service

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Table 2. CSW interfaces and their operations (Source: Nebert & Whiteside,
2004)
Interface Operation Description
OGC_Service getCapabilities() To retrieve metadata describing what the Catalog Service
can provide
Discovery query() To execute a query that searches the cataloged metadata
and returns results satisfying the search criteria
present() To retrieve selected metatdata for specific records
describeRecordType() To retrieve type definitions used for that dataset
getDomain() To retrieve the domain (allowed values) of a metadata
property or request parameter
Manager transaction() To request a specific set of “insert”, “update”, and
“delete” action
harvestResource() To retrieve a resource from a specified location and to
optionally create one or more entries for that resource
Session Not discussed. Possible operations are: cancel(), close(), initialize(), and status().
BrokedAccess No discussed. Possible operation is order().

The CSW interfaces are well-defined that unify the operations for the registration
of new geospatial data/Web services and the query and retrieval of item descrip-
tion. Table 2 lists the interfaces and operations defined by CSW. This specification
is still in evolving.

Search.by.Index

A search for geospatial Web services can be done more actively by using an index
approach. The index approach harvests the Web services by crawling through all
Web sites and servers and collects their metadata information into an indexer. This
approach is a bottom-up search in which several dedicated crawlers keep on scrub-
bing through all Web sites to find geographical data and geospatial Web services.
Figure 6 shows a proposed architecture for such an index search engine. The OGC
Registry can be used to manage these collected metadata, and the query interfaces
can follow the OGC Registry standard as well. The difference is that such an ap-
proach relies upon a passive publication of geospatial Web service and geospatial
data; the crawlers of the search engine scrub through the Web to find this data instead
of providers’ submission.
The most challenge in this approach is how to determine if a link or Web service is
geospatial or non-geospatial. The accuracy of identifying geospatial data and geo-
spatial Web service depends upon the metadata associated with each data or Web
service. For standard geospatial data (e.g., HDF-EOS file), an extraction of internal
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Figure 6. Crawler-based geospatial Web service discovery

geospatial information will gather its geographic extent, time, and coordinate sys-
tems. For standard-compliant geospatial Web services (such as OGC WFS, WCS),
further contents can be extracted by calling its standard interface, for example,
getCapabilities for OGC-compliant geospatial Web services. For semantically-de-
scribed geospatial Web service or data (such as associated taxonomy information
in Ontology Language for Web [OWL]), a reasoner can be embedded to extract
geospatial knowledge, such as class hierarchy of the data.

Search.by.Peer-to-Peer

Peer-to-peer (P2P) computing treats each computer as equally as another computer

in the network. It is a different view from client/server architecture. Grid comput-
ing is considered as a P2P. The P2P search architecture does not need a centralized
registry or a collected indexer. Each node in the P2P responds to the queries it
receives (Booth et al., 2004). Once a user submits a query, the query is passed on
to each individual node in the P2P network and returns the matched results through
relaying. Such a search system may suffer performance inefficiencies, for example,
lag time in propagation of queries and relaying of matches, and incomplete routing
through the network.

Geospatial. Web. Services. Orchestration

Assembling individual geospatial Web services into a service chain for representing
a more complicated geospatial model and process flow proves to be essential for

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

complex geospatial applications and knowledge discovery (Di, 2004a; Di, 2004c;
Di, 2005; Di, 2005a). Such a process is called geospatial Web service orchestration.
Service orchestration introduces a new method of application development and has
great potential to reduce the development time and effort for new applications.

Processes.of.Orchestration

Geospatial Web service orchestration, in the context of this chapter, is nearly ex-
changeable with choreography, to embrace any service chaining activities, including

Figure 7. Typical composite processes for service chaining

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

both top-down (orchestration) and bottom-up (choreography) service composition.

Typical processes in service chain include Sequence, Split, Split + Join, Choice,
Any-Order, Condition, If-Then-Else, Iterate, Repeat-While, and Repeat-Until. Fig-
ure 7 illustrates some typical composite processes by aggregating Web services or
orchestrating (conducting) heterogeneous Web services.

Orchestration.Approaches

Geospatial Web service orchestration can be completed in three approaches: trans-

parent, translucent, and opaque (Alameh, 2001; Di, 2004b; ISO/TC211, 2002b).

Transparent Geospatial Web Services Orchestration

\With transparent geospatial Web service orchestration, the user plays a central role
in finding all the required geospatial Web services and geospatial data. Once the user
identifies all the required services or data, a composite process can be achieved by
two ways: The user invokes the services one by one in a user-controlled sequence (see
Figure 8(a)), or the user creates a composite process in Business Process Execution
Language (BPEL) or OWL-based Web service ontology (OWL-S) and execute the
composite process using a proper engine (see Figure 8(b) for a composite sequence)
(Andrews et al., 2003; Arkin et al., 2005; Martin et al., 2004). For the former, the user
takes care of the sequence of requests as well as the transfer of information between
geospatial Web services. For the latter, the user pre-composes the sequence and the
information or message is directly relayed between geospatial Web services.

Translucent Geospatial Web Service Orchestration

A workflow management service or a composition service is used in such a case

to coordinate all the invocation and scheduling of component Web services. The
user or expert only needs to prepare an abstract composite service and store it in a
composite service repository. The user later on can view and adjust the component
Web service using a composite service editor/viewer.

Opaque Geospatial Services Orchestration

From the user viewpoint, an opaque geospatial Web service chain acts exactly the
same as a simple Web service — the user sets all the required parameters, submits
the request, and gets back the results without knowing what are the component
Web services. From the system viewpoint, the opaque Web service chain can be a

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Zhao, Yu, & D

Figure 8. Transparent geospatial Web service chaining (Source: Alameh, 2001)

Figure 9. Translucent geospatial Web service chaining (Source: Alameh, 2001)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Figure 10. Opaque geospatial Web service chaining (Source: Alameh, 2001)

pre-defined composite Web service or a dynamically-chained Web service. For the

former, the component Web services are fixed for the service chain even though
the user does not know what they are. These can be pre-defined by experts. For the
latter, the component Web services can only be fixed when the specific request is
fixed. Some intelligent reasoning is used to complete the chaining instantly. This is
especially useful in serving earth-science data when the data needs some intermittent
process to meet customized requirements of users (Di et al., 2005).

Geospatial. Web. Service. Solutions

The emerging standards enable the sharing of spatial data (data service) and process-
ing functions (Web services) over the Web. Envisioning the prospect of avoiding the
headache of formatting, geo-referencing, and integrating spatial data with existing
data, all sectors (governments, industries, educational institutes, and individuals)
are getting involved in publishing their geospatial Web services following these
standards (Sayar, Pierce, & Fox, 2005). More than 260 products have claimed to be
complaint with or implementing OGC specifications or interfaces (OGC, 2005b).
Figure 11 shows part of the summary report for the compliant or implementing
OGC specifications and services. Web map service (WMS), Web feature service
(WFS), and Geography Markup Language (GML) are among the most popular
specifications. By further analyzing the products claimed to be compliant with OGC

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Figure 11. Part of the statistical summary of OGC specification or interface imple-
mentations (Source: OGC, 2005b)

specifications, this section gives users a head-start guide to know about commercial
and open-source programs.

Server-Side.Solutions

More and more geospatial software products have been published using standards-
compliant specifications, for example, GeoMedia WebMap (Intergraph), and Ar-
cWeb services (ESRI). Table 3 lists some of the server-side products that support
OGC specifications, including WMS, WFS, catalog service specification (CAT),
GML, simple feature — SQL (SFS), and WCS. To achieve more interoperability,
several data services from governments start to implement these specifications.
Geospatial-one-stop (geodata.gov) is one of them which support a wide suite of
OGC specifications.

Client-Side.Solutions

On the client side, more and more client software have claimed that they support
the access of geospatial data and services through open geospatial standards, for

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Table 3. List of standard-compliant geospatial web-services (server side) (Source:

OGC, 2005b)
Specification Product
WMS 1.0, 1.1, MapServer 3.5, 4.2; CarbonTools 2; Drill Down Server Web services; TakukGIS Developer
1.1.1 Kernel; POLYGIS MAPSERVER 9.1; SuperGIS 1.2; SuperWebGIS 1.2; e-NeXt server
6.2; GIS Broker 3.3; TerraVision WMS Server 1.1; PostGIS / PostgreSQL 0.8.0; rasdaman
5.1; XIS (Extensible information Systems) 1.1; WebPix WMS Web service 1.0; Geomatica
- WebServer Suite 10.0; shp2gml GeoBroker 1.1; Oracle Application Server MapViewer,
10g Release 2 (10.1.2); COP-WDS 1.1; SpatialFX 4.0; Navici; GIS/LINE WMS 1.0; DMAP
GIDB Portal OGC Web services WMS 1.1.1, 1.1.0; MIT Ortho Server ; MiraMon Map
Server 5.0; MapXtreme 2004 6.1, Java Edition 4.7;WEGA-MARS 3.1, 4.1; LuciadMap 4.2,
5; deegree owsProxy 0.5, deegree Web Map Service 1.1.2;Gothic Integrator: Java Edition
(GI:JE) WMS, Gothic WebMapper v4-1a-12a-2; JGisBr - JGisCatalogo V2; WMSControl.
Net 1.0b; RedSpider Catalog 2.0, Web 3.0, 3.1, 3.3;International Interfaces AWMS Server
and Cookbook 1; GeoMedia; XtraWMS 0.38; JCarnacGIS 2.0; FasterMap 1.0; Geologica
Map Server 1.1; GeoServer 1.2; GeoSecurity 2.0; GIS Portal 2.0; Global Aerial and Satellite
Image Service 3.0g-business integrator 2.0; Exposure Image Server 1, Spatial Server 1;
SclMapProxy 1; SclMapServer 2.30; GenaWare 8; sisIMS 2004.0301; GE Smallworld
Internet Application Server; PlanAccess WMS 1.5; FreeStyle Suite 1.15; NAVImap 1;
ArcIMS 3.0, 4.0, 4.0.1, 9.0; GIS Portal Tool Kit 2.0; iSMART 4.3; Demis Map Server 2.0,
3.0, 4.0; XMap Web 3.0; CubeServ – Web Map Server 2.10.7, Cascading Web Map Server
4.0.2; sdi.suite terraCatalog 2.0, terrainServer 1.0; ArcIMS to OGC Connector 1.2; Map
Manager 3.2; Service manager 3.0; Web Enterprise Suite 3.1; Web Feature Portrayal Service
1.2; SpatialAce Web Map Server 2.0; CARIS Chartserver 2.0, Spatial Fusion 2.5.1, 3.0,
3.2, Cascading Web Map Server 3.2; HPDWeb 1.1; Cadcorp GeognoSIS.NET V6.1, SIS
Feature Server 6, Map Server 5.2, 6, 6.1; GeoMapServer 1.0; Autodesk WMS Extention 6.3;
SICAD/open – IMS 6.0, SICAD/Raster – IMS 6.0, SICAD/SD – IMS 6.0; GeoServer 4.1
WFS 1.0 MapServer 4.2; CarbonTools 2; TakukGIS Developer Kernel 8.x; SuperWebGIS 1.2;
Geomatica-WebServer Suite 10.0; GEOINTEGRATION 2.0; GIS/LINE WFS 1.0; LuciadMap
5; deegree Web Feature Service 1.2.1; Gothic WebMapper v4-1a-12a-2; RedSpider Catalog
2.0, Web 3.0, 3.3; GeoMedia; JCarnacGIS 2.0; GeoServer 1.2; g.business integrator 2.0,
organizer 2.0; FreeStyle Suite 1.15; Cartalinea Geographic Data Server (GDS) 1.4; ArcIMS
4.0, 4.0.1, 9.0, 9.1; GIS Toolk Kit 2.0; CubeServ – Web Feature Server 2.10.7; sdi.suite
terraCatalog 2.0; Meta Manager – WFS 5.0; Service Manager 3.0; CARIS Spatial Fusion
3.2; Cadcorp GeognoSIS.NET V6.1; LIDS Application Server 6.5.3
GML 1.0, 2.0, MapServer 4.2; CarbonTools 2; TatukGIS Developer Kernel 8.x;SuperWebGIS 1.2; WFS
2.1.1, 2.1.2, 3.0, Server 1.0; PCI Web Prototypes 0.1; shp2gml GeoBroker 1.1; Envinsa 3.5; MapInfo Routing
3.1 J Server 3.2; MapMaker J Server 3.0; MapXtreme 2004 6.1, Java Edition 4.7; LuciadMap
5; degree Web Feature Service 1.2.1; Gothic WebMapper v4-1a-12a-2; RedSpider Web 3.3;
GeoMedia; JCarnacGIS 2.0; FasterMap 1-0; g.buisiness integrator 2.0; SclFeatureServer
1.3; Cartalinea Geographic Data Server (GDS) 1.4; ArcIMS; iSMART 4.3; CubeServ – Web
Map Server; Map manager 3.2; XchainJ 1.1; LIDS Application Server 6.5.3
SFS TatukGIS Developer Kernel 8.x; PostGIS / PostgreSQL 0.8.0; Oracle Application Server Map
Viewer, 10g, Oracle8i Spatial 8.1.x; MapInfo SpatialWare 4.5-4.8; JGisBr-JGisCatalogo v2;
GeoMedia Data Server for Oracle Object Model Server 05.01; IBM DB2 Spatial Extender
7.1, 8.1; GeoTask Server on IBM DB2, on Oracle 9i; sisNET 2004; ArcSDE; TOPOBASE
Geodataserver; AEDIDB (option AEDIDB-SimpleFeature) 3.5
CAT 1.0, 2.0 deegree Web Catalog Service; JGisBr- JGisCatalogo V2; RedSpdier Catalog 2.0; SMMS
GeoConnect 5.x; ogcatsrv 1.2; ArcIMS Metadata Server; GIS Portal Tool Kit 2.0�con terra
Catalog Server 1; Meta Miner 3.1
WCS 0.7, 1.0 Geomatica –WebServer Suite 10.0; MiraMon Map Server 5.0; LuciadMap 5; deegree
Web Coverage Service 1.1.5; RedSpider Catalog 2.0, Web 3.1, 3.3; ArcGIS Server 9.0;
GIS Portal Tool Kit 2.0

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Table 4. List of standard-compliant geospatial web-services (client-side only)

(Source: OGC, 2005b)
Specification Product
WMS 1.0, 1.1, Gaia 2; J2ME WMS Client 1.0.1; Skylab GPS Simulator 1.1; WebMap Composer 2.03;
1.1.1 Arc3WMS 1.0; XIS View Point 1.1; NACMAP 4.0; Internet Mapping Framework 1.0.0;
Java based wmsClient 1; deegree iGeoPortal 0.3; deeJUMP 1.0; RedSpider Studio 3.0;
GeoMedia Viewer; XtraZWMSClient 0.27; Community Mapbuilder mapbuilder-lib;
g.business navigator 2.0; Mapplets 5.0; sclMapJSClient; WebWraptor 1.0; sisVIEW
1.0; ArcExplorer Web; Chameleon 1.x; CubeXPLOR 2.10.7; sdi.suite mapClient 1.0;
Mapbender 1.4, 2.0; CARIS WMS Client 1.0; Cadcorp apSIS 6; SICAD-IS D-HTML
Client 6.0, HTML Client 6.0, Java Client 6.0
WFS 1.0 Gaia 2; WebMap Composer 2.03; deegree iGeoPortal 0.3; deeJUMP 1.0; RedSpider
Studio 3.0; GeoMedia Viewer; g.business organizer 2.0; SclMapJSClient; ArcGIS Data
Interoperability Extension 9.0; Chameleon 1.x
GML 1.0, 2.0, 2.1.1, Gaia 2; GO Loader 0.3; deegree iGeoPortal 0.3; RedSpider Studio 3.0; GeoMedia Viewer;
2.1.2, 3.0, 3.1 sisVIEW 1.0; ArcGIS Data Interoperability Extension 9.0;CubeXPLOR 2.10.7;
SFS Munsys Cadasrral/Drainage/Electricity/Roads/Sewer/Water; MapInfo Professional 8.0;
JGisBr-JGisEditor V2; HiRDB Spatial Search Plug-in 01-01; Spatial Database Engine
for DB2 Datajoiner (Infomix or Oracle) 3.0.2; Cadcorp SIS 5.2
CAT 1.0, 2.0 JGisBr-JGisEditor V2; ogcatclt 1.2; Catalog Server for Informix 1; terraCatalog 1.2

example, ArcExplorer Web (ESRI), Interoperability Add-on for ArcGIS (ESRI),

MapInfo Professional (MapInfo), and GeoMedia (Intergraph). Most server-side
vendors have their own clients to access their standards-compliant servers with
similar names. Table 4 gives some extra client applications that implement the
OGC specifications.

Open-Source.Solutions

One noticeable aspect is that open source software products are emerging as a key
force in developing standard-compliant geospatial Web services and clients. For the
server side, examples of geospatial Web services are LAITS OGC WCS, WICS, and
WMS from George Mason University (http://laits.gmu.edu), MapServer (Univer-
sity of Minnesota), and GeoTools Web Map Server (http://www.geotools.org). For
the client side, examples of standards-supported clients include the LAITS client
(George Mason University) and the MapServer client (University of Minnesota),
and so forth. These offer researchers the opportunity to explore the mechanisms in
designing, developing, implementing, and testing of geospatial Web services.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Semantic. Issues. on. Geospatial. Web. Service

With ever-growing geospatial Web services and geospatial data, it is very important to
find the proper data and service matching specific requirements. Metadata is always
used in describing and discovering geospatial Web services. However, mismatching
may arise due to semantic ambiguity, or the same term used in different domains
may lead to a different meaning (Sivashanmugam et al., 2003). This section reviews
the semantic issues and their possible solutions for geospatial Web services.

Semantic.Issues

Metadata for geospatial Web services can be classified into three levels: syntactic,
structural, and semantic (Sheth, 1999, 2003; Sivashanmugam et al., 2003). Syntactic
metadata describes basic sources and simple authorship information about geospatial
Web services, such as data file format, creation date, sources, file size, and authors.
Structural metadata describes the structure of geospatial data and the functions of
the geospatial Web services, such as XML schema, data organization types (vector
or raster), and functional types (data conversion, manipulation). Semantic metadata
defines the context or domain of the geospatial data and geospatial services, such
as thematic type for data (e.g., landslide in geoscience, population distribution in
economic geography) and functional domain types (e.g., vegetation index computa-
tion in remote sensing, land surface temperature estimation in climate).

Figure 12. Data and their semantic hierarchy (Source: Sheth, 1999, 2003; Siv-
ashanmugam et al., 2003)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Semantic issues for interoperability have been recognized in many distributed multiple
data systems (Kashyap & Sheth, 1996; Sheth, 1991; Sivashanmugam et al., 2003).
Similar semantic issues rise for geospatial Web service chaining. For example, a
geospatial service chain is required to compute a landslide susceptibility map, given
a large repository of geospatial Web services. A landslide susceptibility model (as
a Web service) may take seven input datasets: surface slope, surface aspect, base
rock slope, base rock aspect, depth to base rock, soil moisture, and vegetation. All
these input data types may not be directly available at data repositories, but they
can be derived from some other Web services. Therefore, the first task is to find the
Web services with output matching the required data type. The following problems
may occur if semantics is not considered.

Semantic Issue 1: Ambiguous Match Due to Lack of Semantics

Each output of the services may be represented by a URI (universal resource iden-
tification) string, or other basic data types (e.g., integer, double, or single). If the
match criteria are only defined as the match on data types (e.g., the only search
criterion being that the output of a service is a URI string), the returned results may
be overwhelming, containing numerous improper matches. This is because a URI
string may carry different meaning.

Semantic Issue 2: Ambiguous Match Due to Incomplete Semantics

If the search criteria is based upon the service structure (e.g., a service with one
input of DEM and one output of slope), the returned results may not satisfy the exact
requirement; an improper computation algorithm may have been used to derive the
slope. There are many algorithms for slope calculations: neighborhood method, qua-
dratic surface method, maximum slope method, and maximum downhill method. If
a desired method is quadratic surface method, the match with other methods would
result in the failure of searching.

Semantic Issue 3: Missed Match Due to Lack of Relationship Semantics

A search for a vegetation cover map data service may fail to retrieve any match
although a land cover map data service may exist in the repository. It is common
knowledge that a land cover map can be actually used in place of a vegetation map
since the vegetation categories (forest land, grass land, cultivated land) available in
the land cover map may be sufficient for this purpose. Without a clear definition of
the relationship between vegetation cover and land cover maps in the semantics of
the metadata, the search would not be able to make this intelligent association.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Semantic.Solutions

One solution to the above problems is to add semantic information to each geospatial
data and geospatial Web services using a proper common ontology. An ontology is
a specific set of vocabulary and relationships that have explicit assumptions about
the intended meanings and relationships of the vocabulary in the set (Sivashan-
mugam et al., 2003). It is a technology that enables the computer-understandable
description of Web services. Emerging standards for ontology are OWL, OWL-S,
and Semantic Web services Language (SWRL) (Dean & Schreiber, 2004; Horrocks,
Patel-Schneider, Boley, Tabet, Grosof, & Dean, 2003; Martin et al., 2004). These
standards evolve the description of Web services from a syntactical description to
a semantic description, mainly based on description logic. These can be used to
describe the taxonomy of geospatial data and geospatial Web services and define
their relationships with one another. Then, a reasoner can be used to validate the
logic and induce the relationship underlying every term. A match can be made
using this knowledge by matching both data semantics for input and output and
functional semantics for the Web services (Di, 2004c; Di, 2005a; Di et al., 2005;
Sivashanmugam et al., 2003; Zhao et al., 2004).

Conclusion

Web services give promising prospects in interoperability across applications and

organizational boundary, independence from languages, and ease of maintenance.
Geospatial data and service dissemination can take advantage of Web services to
achieve higher usability and efficiencies. This chapter covers all aspects of the life
cycle for geospatial Web services, including standards, implementation, discovery,
and orchestration. It also reviews example solutions and applications leveraging
the standards for geospatial Web services. Semantic issues and possible solutions
are discussed.
Web service is an emerging technology. Standards, specifications, and practices for
Web service are evolving. The same is also true to geospatial Web services. Many
issues in the geospatial Web services still need to be explored. The following lists
a few:

1. A.Web.service.is.originally.designed.to.be.stateless: However, transaction is

required in many cases because geospatial modeling is a complicated process
requiring multiple interactions. How to evolve the geospatial Web service to
deal with transactions should be further studied.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

2. Web.service.uses.plain.XML.as.the.base.to.communicate.between.appli-
cations:.Compared to other distributed computing approaches, Web service
is relatively poor in performance and lack of security. These issues should be
researched for geospatial Web services as well.
3. Semantics.for.geospatial.Web.services.is.in.development:.Many research
issues need to be solved before an operational geospatial semantic Web can
be established.

Acknowledgments

The authors of this chapter, as well as some of the example systems and services
discussed in the chapter, were supported fully or partially with grants from the
NASA Earth Science Data and Information System Project (ESDISP) (NCC5-645,
PI: Dr. Liping Di), NASA Earth Science Technology Office (ESTO) (NAG-13409,
PI: Dr. Liping Di), NASA REASoN program (NNG04GE61A, PI: Dr. Liping Di),
and National Geospatial-intelligence Agency (NGA) University Research Initiative
(NURI) (HM1582-04-1-2021, PI: Dr. Liping Di). We also sincerely thank to our
colleagues at Laboratory for Advanced Information Technology and Standards,
George Mason University. Numerous discussions with our colleagues at LAITS
have brought forth much inspiration and enlightenment and shaped the chapter.

References

Akinci, H. (2004, 12-23 July). Geospatial Web services for e-municipality. Paper
presented at the XXth ISPRS Congress, Istanbul, Turkey.
Alameh, N. S. (2001). Scalable and extensible infrastructures for distributing
interoperable geographic information services on the Internet. Cambridge,
MA: Massachusetts Institute of Technology..
Andersson, E., Greenspun, P., & Grumet, A. (2006). Software engineering for in-
ternet applications. Cambridge, MA: The MIT Press.
Andrews, T., Curbera, F., Dholakia, H., Goland, Y., Klein, J., Leymann, F., et al.
(2003). Business Process Execution Language for Web services (Version 1.1).
Retrieved from http://dev2dev.bea.com/webservices/BPEL4WS.html
Arkin, A., Askary, S., Bloch, B., Curbera, F., Goland, Y., Kartha, N., et al. (2005,
July 13). Web Services Business Process Execution Language Version 2.0

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

(Working Draft). Retrieved from http://www.oasis-open.org/apps/org/work-

group/wsbpel/
Bacharach, S. (2005). OGC to begin geospatial semantic Web interoperability
experiment. Retrieved August 20, 2005, from http://www.opengeospatial.
org/press?page=pressrelease&year=0&prid=220
Booth, D., Haas, H., McCabe, F., Newcomer, E., Champion, M., Ferris, C., et al.
(2004, February 11). Web services architecture (W3C Working Group Note).
Retrieved August 8, 2005, from http://www.w3.org/TR/2004/NOTE-wsdl20-
primer-20050803/
Booth, D., & Liu, C. K. (2005, August 3). Web Services Description Language
(WSDL) Version 2.0 Part 0: Primer (W3C Working Draft). Retrieved from
http://www.w3.org/TR/2005/WD-wsdl20-primer-20050803/
Burry, L. (2001). OpenGIS implementation specification: Coordinate tansforma-
tion services, revision 1.00. Retrieved August 20, 2005, from http://portal.
opengeospatial.org/files/?artifact_id=6628
Chinnici, R., Haas, H., Lewis, A., Moreau, J. -J., Orchard, D., & Weerawarana, S.
(2005, August 3). Web Services Description Language (WSDL) Version 2.0
Part 2: Adjuncts (W3C Working Draft). Retrieved from http://www.w3.org/
TR/2005/WD-wsdl20-adjuncts-20050803/
Chinnici, R., Moreau, J.-J., Ryman, A., & Weerawarana, S. (2005, August 3). Web
Services Description Language (WSDL) Version 2.0 Part 1: Core Language
(W3C Working Draft). Retrieved from http://www.w3.org/TR/2005/WD-
wsdl20-20050803/
Christensen, E., Curbera, F., Meredith, G., & Weerawarana, S. (2001, March 15).
Web Services Description Language (WSDL) 1.1 (W3C Note). Retrieved from
http://www.w3.org/TR/2001/NOTE-wsdl-20010315
Clement, L., Hately, A., Riegen, C. Y., & Rogers, T. (2004, October 19). UDDI Ver-
sion 3.0.2 (UDDI Spec Technical Committee Draft). Retrieved from http://uddi.
org/pubs/uddi-v3.0.2-20041019.htm
Cox, S., Daisey, P., Lake, R., Portele, C., & Whiteside, A. (2004). OpenGIS Ge-
ography Markup Language (GML) implementation specification. Retrieved
August 20, 2005, from http://schemas.opengis.net/gml/3.1.0/base/
Daly, M. (2001). OpenGIS implementation specification: Coordinate transforma-
tion services, revision 1.00. Retrieved August 20, 2005, from http://portal.
opengeospatial.org/files/?artifact_id=999
Dean, M., & Schreiber, G. (2004). OWL Web Ontology Language reference. Retrieved
from http://www.w3.org/TR/2004/REC-owl-ref-20040210/
de La Beaujardiere, J. (2004). OpenGIS Web map service, Version 1.3. Retrieved Au-
gust 20, 2005, from http://portal.opengeospatial.org/files?artifact_id=5316

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Zhao, Yu, & D

Di, L. (2003, July 21-25). The development of remote-sensing related standards at

FGDC, OGC, and ISO TC 211. In Proceedings of 2003 IEEE International Geo-
science and Remote Sensing Symposium (IGARSS 2003), Toulouse, France.
Di, L. (2004a). Distributed geospatial information services-architectures, standards,
and research issues. In The International Archives of Photogrammetry, Remote
Sensing, and Spatial Information Sciences (Vol. XXXV, Part 2, Commission
II).
Di, L. (2004b). GeoBrain — A Web services based geospatial knowledge building
system. In Proceedings of the Fourth annual NASA Earth Science Technology
Conference, Palo Alto, CA.
Di, L. (2004c, September 20-24). A framework for intelligent geospatial knowl-
edge discovery at Web service environment..In Proceedings of 2004 IEEE
International Geoscience and Remote Sensing Symposium (IGARSS 2004),
Anchorage , AK.
Di, L. (2005, July 25-29). Customizable virtual geospatial products at Web/grid
service environment. In Proceedings of 2005 IEEE International Geoscience
and Remote Sensing Symposium (IGARSS 2005), Seoul, Korea.
Di, L. (2005). A framework for construction of Web-service based intelligent
geospatial knowledge systems. Journal of Geographic Information Science,
11(1), 24-28.
Di, L. (in press). The open GIS Web service specifications for interoperable access
and services of NASA EOS Data. In J. Qu., et al. (Eds.), Earth science satellite
remote sensing. Springer-Verlag.
Di, L., & McDonald, K. (2005). The NASA HDF-EOS Web GIS software suite
(NWGISS). In J. Qu., et al. (Eds.), Earth science satellite remote sensing.
Springer-Verlag.
Di, L., Yang, W., Deng, M., Deng, D., & McDonald, K. (2002). Interoperable ac-
cess of remote sensing data through NWGISS. In Proceedings of 2002 IEEE
International Geoscience and Remote Sensing Symposium (IGARSS 2002),
Toronto, Canada (pp. 255-257).
Di, L., Zhao, P., Yang, W., Yu, G., & Yue, P. (2005, July 25-29). Intelligent geospa-
tial Web services. In Proceedings of 2005 IEEE International Geoscience and
Remote Sensing Symposium (IGARSS 2005), Seoul, Korea.
Duschene, P., & Sonnet, J. (2005a). OGC WCS change request: Support for WSDL
& SOAP, Version 0.1.0. Retrieved August 20, 2005, from http://portal.open-
geospatial.org/files/?artifact_id=9540
Duschene, P., & Sonnet, J. (2005b). OGC WMS change request: Support for WSDL
& SOAP. Retrieved August 20, 2005, from http://portal.opengeospatial.org/
files/?artifact_id=9541

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

EOP (2003). Coordinate geographic data acquisition and acess; the national spatial
data infrastructure, Federal Register, 68, 10619-10633.
Evans, J. D. (2003). OpenGIS Web coverage service (WCS) Version 1.0.0. Re-
trieved August 20, 2005, from https://portal.opengeospatial.org/files/?artifact_
id=3837
Evans, J. D., & Bambacus, M. J. (2005, 25-29 July). NASA’s earth-sun system gate-
way: An open standards-based portal to geospatial data and services. Paper
presented at the IEEE 25th Anniversary International Geoscience and Remote
Sensing Symposium, Seoul, Korea.
Fielding, R. T. (2000). Architectural styles and the design of network-based software
architectures. Irvine, CA: University of California, Irvine.
GRASS (2005). Geographic resource analysis support system. Retrieved August
10, 2005, from http://grass.itc.it/
Groot, R., & McLaughlin, J. D. (2000). Geospatial data infrastructure: Concepts,
cases, and good practice. Oxford, England; New York: Oxford University
Press.
GSFC (2005). HDF-EOS. Retrieved August 20, 2005, from http://hdfeos.feos.gsfc.
nasa.gov/
Gudgin, M., Hadley, M., Mendelsohn, N., Moreau, J. -J., & Nielsen, H. F. (2003,
June 24). SOAP Version 1.2 Part 1: Messaging framework (W3C Recommen-
dation). Retrieved from http://www.w3.org/TR/soap12/
Hecht, L., Jr. (2002). Web services are the future of geoprocessing. Retrieved August
14, 2006, from http://www.geoplane.com/gw/2002/02/06/206apng.asp
Hirsch, F., & Just, M. (2003, May 5). XML key management (XKMS 2.0) Require-
ments (W3C Note). Retrieved from http://www.w3.org/TR/2003/NOTE-
xkms2-20030505
Horrocks, I., Patel-Schneider, P. F., Boley, H., Tabet, S., Grosof, B., & Dean, M.
(2003). SWRL: A Semantic Web Rule Language combining OWL and RuleML.
Retrieved from http://www.daml.org/2003/11/swrl/
IETF. (2005). The Internet Engineering Task Force. Retrieved August 20, 2005,
from http://www.ietf.org/
ISO/TC211. (2002a). Draft new work item proposal: Geographic information
— metadata: Part 2 - Metadata for imagery and gridded data. Retrieved from
http://www.isotc211.org/protdoc/211n1337/211n1337.pdf
ISO/TC211. (2002b). Geographic information services. Retrieved from http://www.
ncits.org/ref-docs/DIS19119.PDF
ISO/TC211. (2003a). Geographic information — metadata. Retrieved from http://
www.ncits.org/ref-docs/FIDIS_19115.pdf

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

ISO/TC211. (2003b). ISO 19139 dataset implementation model and schema, Version
0.7. Retrieved August 20, 2005, from http://metadata.dgiwg.org/ISO19115/
ISO19115_v0_7detail.htm
ISO/TC211. (2005). ISO/TC211 Geographic information/geomatics. Retrieved
August 20, 2005, from http://www.isotc211.org/
Kashyap, V., & Sheth, A. P. (1996). Semantic and schematic similarities between
database objects: A context-based approach. VLDB Journal, 5(4), 276-304.
Kolodziej, K. (2004). OpenGIS Web map server cookbook, Version 1.0.2. Re-
trieved August 20, 2005, from http://portal.opengeospatial.org/files/?artifact_
id=7769
Kralidis, A. (2005). Geospatial Web services: An evolution of geospatial data in-
frastructure. Ottawa, Ontario, Canada: Carleton University.
LAITS (2005). Mobilization of NASA EOS data and information through Web services
and knowledge management technologies for higher-education teaching and
research. Retrieved August 20, 2005, from http://geobrain.laits.gmu.edu
Lieberman, J., Reich, L., & Vretanos, P. (2003). OWS 1.2 UDDI experiment,
Version 0.5. Retrieved August 20, 2005, from http://portal.opengeospatial.
org/files/?artifact_id=1317
Mabrouk, M. (2005). OpenGIS® location service (OpenLS) implementation specifica-
tion: Core services, Version 1.1. Retrieved August 20, 2005, from http://portal.
opengeospatial.org/files/?artifact_id=8836
Martell, R. (2004). OGC™ catalogue services — ebRIM (ISO/TS 15000-3) pro-
file of CSW. Retrieved from https://portal.opengeospatial.org/files/index.
php?artifact_id=6495
Martin, D., Burstein, M., Hobbs, J., Lassila, O., McDermott, D., McIlraith, S., et
al. (2004, November 22). OWL-S: Semantic markup for Web services ( W3C
Member Submission). Retrieved from http://www.w3.org/Submission/2004/
SUBM-OWL-S-20041122/
Mateos, C., Zunino, A., & Campo, M. (2005). Integrating intelligent mobile agents
with Web services. International Journal of Web Services Research, 2(2),
85-103.
Nebert, D. (2005). Current standards, specifications, and practices of relevance
to SDI development. Retrieved August 20, 2005, from http://www.anzlic.org.
au/pubinfo/2413322725
Nebert, D., & Whiteside, A. (2004). OpenGIS catalogue services specifica-
tion. Retrieved from http://portal.opengeospatial.org/org/files/?artifact_
id=5929&version=1

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

OASIS. (2005). Organization for the Advancement of Structured Information Stan-

dards. Retrieved August 20, 2005, from http://www.oasis-open.org/home/index.
php
OASIS/ebXML. (2003). OASIS/ebXML registry informaiton model V2.5. Retrieved
from http://www.oasis-open.org/committees/regrep/documents/2.5/specs/
ebrim-2.5.pdf
OGC. (2004). Critical infrastructure protection initiative, Phase 2 (CIPI2). Retrieved
August 20, 2005, from http://www.opengeospatial.org/initiatives/?iid=66
OGC. (2005a). OpenGIS specifications. Retrieved August 20, 2005, from http://www.
opengeospatial.org/specs/download.php?view=IS&keyword=&excel=1
OGC. (2005b). Resources — registered products. Retrieved from http://www.open-
geospatial.org/resources/?page=products
Peltz, C. (2003). Best practices for Web services development. Retrieved August 20,
2005, from http://devresource.hp.com/drc/resources/WSBestPractices/index.
jsp
Peltz, C., & Rogers, C. (2003). Leveraging open source for Web servies develop-
ment. Retrieved August 20, 2005, from http://devresource.hp.com/drc/techni-
cal_white_papers/wsopenrc.pdf
Sayar, A., Pierce, M., & Fox, G. (2005). OGC compatible geographical information
services. Bloomington, IN: Indiana University.
Sheth, A. P. (1991). Semantic issues in multidatabase systems. SIGMOD Record,
20(4), 5-9.
Sheth, A. P. (1999). Changing focus on interoperability in information systems: From
system, syntax, structure to semantics. In M. F. Goodchild, M. J. Egenhofer,
R. Fegeas, & C. A. Kottman (Eds.), Interoperating geographic information
systems (pp. 5-30). Boston: Kluwer Academic Pub.
Sheth, A. P. (2003). Semantic meta data for enterprise information integration. DM
Review, 13, 52-54.
Sivashanmugam, K., Sheth, A., Miller, J., Verma, K., Aggarwal, R., & Rajasekaran,
P. (2003). Metadata and semantics for Web services and processes. In W.
Benn, P. Dadam, S. Kirn & R. Unland (Eds.), Datenbanken und Informations-
systeme: Festschrift zum 60. Geburtstag von Gunter Schlageter (pp. 245-271).
Hagen, Germany: Fachbereich Informatik.
Skonnard, A. (2002). The XML files: The birth of Web services. Retrieved August
14, 2006, from http://msdn.microsoft.com/msdnmag/issues/02/10/xmlfiles/
Sonnet, J. (2004). OWS 2 common architecture: WSDL SOAP UDDI, Version
1.0.0. Retrieved August 20, 2005, from https://portal.opengeospatial.org/
files/?artifact_id=8348

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Zhao, Yu, & D

Sonnet, J. (2005). OpenGIS Web map context implementation specification, Ver-

sion 1.1. Retrieved August 20, 2005, from https://portal.opengeospatial.
org/files/?artifact_id=8618
Sonnet, J., & Savage, C. (2003). OWS 1.2 SOAP experiment report. Retrieved August
20, 2005, from http://portal.opengeospatial.org/files/?artifact_id=1337
Tsalgatidou, A., & Pilioura, T. (2002). An overview of standards and related technol-
ogy in Web services. Distributed and Parallel Databases, 12, 135-162.
Vaughan-Nichols, S. J. (2002). Web services: Beyond the hype. Computer, 35(2),
18-21.
Vedamuthu, A. S. (2006). Web Services Description Language (WSDL) Version 2.0
SOAP 1.1 binding (W3C Working Draft). Retrieved August 14, 2006, from
http://www.w3org/TR/wsdl20-soap11-binding/
Voges, U., & Zenkler, K. (2005). OpenGIS catalog services specification 2.0 —
ISO19115/ISO19119 application profile for CSW 2.0, Version 0.9.3. Retrieved
from https://portal.opengeospatial.org/files/?artifact_id=8305
Vretanos, P. A. (2005). OpenGIS Web feature service implementation specification,
Version 1.1.0. Retrieved August 20, 2005, from https://portal.opengeospatial.
org/files/?artifact_id=8339
W3C. (2005a). W3C XML schema. Retrieved August 20, 2005, from http://www.
w3.org/XML/Schema
W3C. (2005b). World Wide Web Consortium. Retrieved August 20, 2005, from
http://www.w3.org/
Wei, Y., Di, L., Zhao, B., Liao, G., Chen, A., Bai, Y., et al. (2005, 25-29 July). The
design and implementation of a grid-enabled catalogue service. Paper presented
at the IEEE 25th Anniversary International Geoscience and Remote Sensing
Symposium, Seoul, Korea.
Whiteside, A. (2005). OGC Web services common specification. Retrieved August
20, 2005, from https://portal.opengeospatial.org/files/?artifact_id=8798
Whiteside, A., Müller, M. U., Fellah, S., & Warmerdam, F. (2005). OGC Web co-
ordinate transformation service (WCTS) draft implementation specification,
Version 0.3.0. Retrieved August 20, 2005, from https://portal.opengeospatial.
org/files/?artifact_id=8847
Wilkes, L. (2005). The Web services protocol stack. Retrieved August 20, 2005,
from http://roadmap.cpdiforum.com/reports/protocols/index.php
Winer, D. (1999). XML-RPC specification. Retrieved from http://www.xmlrpc.
com/spec
Yang, W., & Whiteside, A. (2005). OGC Web image classification service (WICS)
impementation specification, Version 0.3.3. Retrieved August 20, 2005, from
http://portal.opengeospatial.org/files/?artifact_id=8981

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatal Web Servces 

Zhao, P., Chen, A., Liu, Y., Di, L., Yang, W., & Li, P. (2004, 12-13 November). Grid
metadata catalog service-based OGC Web registry service. Paper presented
at the 12th ACM International Workshop on Geographic Information Systems
(ACM-GIS 2004), Washington DC.
Zhao, P., Nusser, S., & Miller, L. (2002). Design of field wrappers for mobile
field data collection. Paper presented at the Proceedings of the Tenth ACM
International Symposium on Advances in Geographic Information Systems
(ACM-GIS 2002), McLean, VA.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

Chapter.II

Spatial.Information.
System.Development:
The.Role.of.Information.System.
Design.Theory.and.Ontologies
Bran N. Hlton, Claremont Graduate Unversty, USA
Rchard J. Burkhard, San Jose State Unversty, USA
Tarun Abhchandan, Claremont Graduate Unversty, USA

Abstract

An approach to an ontology-based information system design theory for spatial

information system development is presented. This approach addresses the dynamic
nature of information system development at the beginning of the 21st century and
addresses the question of how to establish relationships between the various design
components of a spatial information system. It should also help to automate and guide
the design process while at the same time improve the quality of the process along
with its outputs. An example of this approach is presented, along with examples of
the various ontologies utilized in the design of this particular spatial information
system. Finally, a method to mitigate the issues regarding the organization and
management of a growing library of ontologies is discussed.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Introduction

Spatial informatics, as a field of study within the information sciences, is concerned

with the development of spatial information systems for a specific problem or deci-
sion-making environment. Moreover, the development of these spatial information
systems is being influenced by many factors: the increasing breadth and depth of
various emerging technologies, the nearly exponential increase in the quantity of
spatial data, and the growing awareness in many domains of the impact these systems
can have within these domains. As a result, there is a need for an approach to spatial
information system development that addresses these, and other, influences.
Action research and improvement research are two approaches for conducting
research in the information sciences that employ methods in which the explicitly-
expected outcomes include the solving of a problem or the building of a system
(Truex, 2001). These research approaches differ from explanatory or descriptive
research, which focus on explaining phenomena rather than solving problems.
Furthermore, “whereas explanation of research methods and traditions are based in
natural science and are well established, improvement research methods and tradi-
tions are relatively new and are closer to engineering or computer science modes
of research” (Truex, 2001, p. 7). Information system design theory (ISDT) is one
example of these approaches that has at its core three key elements: a set of user
requirements, a set of system features, and a set of principles for guiding the process
of development. Specifically, an ISDT can be thought of as a complete package of
guidance for designers and practitioners facing particular sets of circumstances that
are based in theory (Markus, Majchrzak, & Gasser, 2002).
An ontology is a hierarchical structuring of knowledge, similar to a dictionary
or glossary, but with greater detail and structure. An ontology consists of a set of
concepts, axioms, and relations, and provides a number of potential benefits in
processing knowledge, including the externalization of domain knowledge from
operational knowledge, sharing of common understanding of subjects among human
and also among computer programs, and the reuse of domain knowledge (Lee et
al., 2004). The impact of ontologies in the information sciences can be seen, on the
methodological side, through the adoption of a highly interdisciplinary approach to
ontology development, while on the architectural side, in the role that an ontology
can play in an information system, leading to the perspective of ontology-driven
information systems (Guarino, 1998).
Increasing international competitiveness and technological advances have prompted
the need for organizations to (Green, Kennedy, & McGown, 2002):

• Exploit emerging technologies more rapidly

• Reduce design time-scales

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

• Provide “right first time” design

• Innovate more frequently and produce more innovative products
• Improve the reliability of products and systems

Consequently, industry and academia have been directing research towards devel-
oping methods to automate and guide the design process while at the same time
improving the quality of the process along with its outputs (Green et al., 2002).
This chapter follows in this vein.

Information. System. Design. Theory

Design.Research

The challenge of design for spatial information systems reflects many of the design
issues that are seen in the information systems discipline as a whole, as well as the
related disciplines of engineering and computer science. A key influence on design
in information systems is the work of Herbert Simon, whose concepts influence
the development of many aspects of design science theory and the disciplines that
contribute to spatial informatics. Simon argued that design sciences for artificial,
or created, systems would find their substance in the generation of artifacts (Simon,
1996) that are attempts to meet the purposes of the designers. These artifacts, such
as information system software elements, are synthesized from various contributing
areas and typically emulate natural systems in some way. In spatial information sys-
tems such as geographic information systems (GIS), an example of a natural system
is found in the high-level view of a terrain as obtained from an aircraft or satellite.
This emulation can be supplemented with multiple dimensions of quantitative data
from other sources such as geometric projection systems, which are combined into
the artifact of the GIS software element. In this model, the GIS exemplifies com-
puter systems that are designed to emulate human activities, particularly thought
processes (Freeman & Hart, 2004).
The view that such artifacts should be a central focus of effort, both in theory and
in practice, in information systems research has gained considerable attention
(Orlikowski & Iacono, 2001). The design science approach to information systems
research “seeks to extend the boundaries of human and organizational capabilities
by creating new and innovative artifacts” such as information systems applica-
tions implemented in computer software (Hevner, March, Park, & Ram, 2004). In
information system design research, increased understanding of the problem area

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

is expected to result from the activities of building the artifact itself (Vaishnavi &
Kuechler, 2004)
Simon and others maintain that the development of such artifact-based systems will
be most efficient if the developers follow a pattern of several general classes of activ-
ity. The first of these activities is creating design alternatives and representing these
alternatives by means of appropriate representational models. The spatial systems
specialty within the field of information systems has developed a variety of such
models, many of which are discussed in the following chapters. Next, a series of
solution approaches are generated followed by assessment of usability and human
factors such as interface design. Finally, the actual artifacts are developed, usually
in the form of a series of prototypes, which in turn are evaluated. Many variants of
this iterative process are familiar within design-based disciplines (NSF, 2004).
The task of designing spatial systems presents special opportunities to emulate
natural systems by allowing visual and cognitive ordering of information to enhance
its communication power. Such information visualization tools allow the user to
examine the “terrain” of information (Bederson, 2001; Shneiderman, Card, Norman,
Tremaine, & Waldrop, 2001). The analytical potential of information visualization
methods is important in any context where a large body of data forms the basis
for decision-making. In particular, visualization tools have special relevance in
situations in which the decision maker faces information overload and hidden or
ambiguous patterns in the decision-basis information, such as what one might face
in the following situations.
For example, spatial information system artifacts can enable pattern discovery
through visual inspection that can be superior to other, non-spatial information sys-
tem methods (Gershon, Eick, & Card, 1998). Pattern discovery through interactive
techniques as provided in spatial information systems can augment the information
search process in unexpected ways (Kreuseler & Schumann, 1999). In addition, the
visualization of spatial information can allow users to intelligently develop specific
questions to be asked via more traditional mechanisms, such as search engines
(Shneiderman, 1999).
The unique opportunity for a spatial information system to emulate and extend
natural human processes places it in a special position to develop as an important
specialty within the information systems field, as well as to contribute to the body
of work in information system design theory.

Information.System.Design.Theory

As presented by Walls, Widmeyer, and El Sawy, 1992), the concern of researchers

in the Information Systems discipline is the design of systems and the development
of system design theories that address the question of how to establish relation-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Hlton, Burkhard, & Abhchandan

ships between components of a system to achieve a specific result. An information

system design theory (ISDT), then, refers to an integrated prescription consisting
of a particular class of user requirements, a type of system solution (with distinc-
tive features), and a set of effective development practices (Markus et al., 2002).
Here, Walls et al. refer to meta-requirements (user requirements), which describe
the specific goals that are applicable to any system of the type to which the theory
relates. Meta-design (system solution) describes the specific artifacts that are ap-
plicable to the meta-requirements, while the design method (effective development
practices) describes the procedures to be utilized in building a system of the type to
which the theory relates. Underlying these three features (meta-requirements, meta-
design, design method) of an ISDT are kernel theories (an academic theory and/or
a practitioner theory-in-use) that enable the formulation of empirically-testable
hypotheses relating the design theory to expected outcomes. The ISDT addresses
both the product of the design effort, which is typically an artifact or instantiation,
and the process of design. In this way, the method of design can be adapted to the
meta-requirements of the product. These design theory characteristics help to
establish why the application of one particular ISDT to a specific situation yields
better results than the application of a different ISDT. For example, “it should be
possible to establish empirically that the application of Decision Support System
design theory to a particular set of requirements produces better results than applying
Transaction Processing System design theory to the same requirements” (Markus et

Figure 1. Components of an information system design theory

Design Product Design Process

Kernel Kernel
Theores Theores

Meta-
Requrements

Meta- Desgn
Desgn Method

Testable Desgn Testable Desgn

Product Hypothess Process Hypothess

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

al., 2002, p. 181). Figure 1 illustrates the relationships between the components of
an ISDT where the meta-requirements and meta-design form the core of the design
product, while the design method forms the core of the design process.

Ontologies

Ontology.Fundamentals

Ontology development is a descriptive activity that seeks to provide taxonomies

and definitive descriptions of the kinds and structures of objects, properties, events,
processes, and relations in every area of reality (Smith, 2003). The main purpose of
developing ontologies is to provide machine-processable semantics of information
sources that can be communicated between different agents (software and humans)
(Fensel, 2003). Ontologies can serve as a common foundation for building and as-
sessing systems in particular domains (Embley, 2004; Sugumaran & Storey, 2002).
They offer specifications of various kinds and general interrelationships (Geroi-
menko, 2003) that can be utilized in designing cohesive systems. Appropriately, an
ontology is defined as a formal, explicit specification of a shared conceptualization
(Gruber, 1993). It is important to note that these conceptualizations can be shared
between agents (software and human) and that they serve a specific context in a
particular domain.

Protégé.Ontology.Editor.and.Knowledge..............................
Acquisition.System

Protégé is an integrated software tool used by system developers and domain ex-
perts to develop knowledge-based systems (Stanford Medical Informatics, 2003).
Such systems are enhanced by ontology-derived common vocabularies that allow
researchers to share information in a domain, and the best examples of such ontol-
ogy-based systems include machine-interpretable definitions of basic concepts in
the domain and relations among them (Noy & McGuinness, 2003). As such, appli-
cations developed with Protégé are used in problem solving and decision-making
in a particular domain. Noy and McGuinness (2003) identified five major reasons
for developing an ontology. They are:

• Sharing common understanding of the structure of information among people

or software agents

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

• Enabling reuse of domain knowledge

• Making explicit domain assumptions
• Separating the domain knowledge from the operational knowledge
• Analyzing domain knowledge

Noy and McGuinness further state that problem-solving methods, domain-indepen-

dent applications, and software agents use ontologies and knowledge bases built
from ontologies as data. Figure 2 is a basic example of an ontology developed using
Protégé that describes a newspaper that defines classes, the class hierarchy, slots
and slot-value restrictions, relationships between classes, and properties of these
relationships.
An ontology includes a formal explicit description of concepts (classes) in a par-
ticular domain or field. Attributes (slots) describe various features of each concept.
These slot-values may have restrictions (constraints). An ontology, together with
a set of individual instances, comprises a knowledge base. Since instances may be
the best description of unique classes, there is often little distinction between the
ontology-defined class and the instances of the larger knowledge base.

Figure 2. Protégé ontology editor and knowledge acquisition system — newspaper

example

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Application.of.Information.System..............................
Design.Theory.and.Ontologies.for.
Spatial. Information. System. Development

The approach to spatial information system development presented here builds on

the foundation of design research, specifically, ISDT and the use of ontologies to
address the question of how to establish relationships between the various design
components of a spatial information system. The example presented below is that
of a particular spatial information system, a proximate commuting system.

Proximate.Commuting

Transportation demand management (TDM) is a general term for strategies that

result in the efficient use of transportation resources. One TDM strategy is commute
trip reduction (CTR), a program that gives commuters resources and incentives
to reduce their automobile trips (Victoria Transport Policy Institute, 2005). These
programs, such as rideshare matching, alternative scheduling, and tele-work, may
be encouraged or required by local, regional, state, or federal policies.
One CTR program, proximate commuting, allows employees to shift to work sites
that are closest to their homes. Here, spatial analysis is performed that identifies the
commutes of employees of multi-location organizations that could be reduced by
working at a closer location. Proximate commuting is a potential employer-based
travel-demand management program under which large, decentralized employers
reassign each voluntary participant to a job location (with the same employer) closer
to the participant’s residence to reduce commuting distances (Rodriguez, 2001). A

Table 1. Design product — proximate commuting system

Feature Definition Realization.in.a.Spatial.Information.System

Kernel theories Core theories that support Multi-criteria decision-making 1
and guide the design Spatial decision support systems 1
Spatial cognition 1
Meta-requirements Meta-requirements describe Geographic human interaction services 2
the specific goals that are Geographic model/information management services 2
applicable to any system of Geographic workflow/task management services 2
the type to which the theory Geographic processing services 2
relates. Geographic communication services 2
Goal: Clarity 3
Goal: Efficiency 3

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

Table 1. continued
Meta-design Meta-design describes the Geographic viewer 2
specific artifacts that are Feature access service 2
applicable to the meta- Coverage access service 2
requirements. Coverage access service — sensor 2
Route determination service 2
Proximity analysis service 2
Geographic annotation service 2
Transfer service 2
Messaging service 2
Remote file and execution management 2
Feature: Maps 3
Feature: Place directories 3
Feature: Page form and performance 3

Testable hypothesis Hypotheses that are open to It is possible to develop a proximate commuting
empirical testing system using the ISDT for SDSS.
A proximate commuting system built using the ISDT
for SDSS will be more open, useful, and innovative.
Note: 1 concepts and theories ontology; 2 geographic services ontology; 3 advanced traveler information systems
ontology

spatial information system, a group of procedures that provide data input, storage and
retrieval, mapping, and spatial analysis for both spatial and attribute data to support
the decision-making activities of the organization (Grimshaw, 2000), is particularly
suited to perform the spatial analysis required for proximate commuting. Tables 1
and 2 present an instance of an ontology-based ISDT, for a proximate commuting
system prototype that was developed by the lead author (Hilton, 2003).

Table 2. Design process — proximate commuting system

Feature Definition Realization.in.a.Spatial.Information.System
Kernel theories Core theories that support Human-centered design 1
and guide the design Quality in use 1
process Prototyping 1

Design method Design method describes Understand and specify the context of use 1
the procedures to be Specify the user and organizational requirements 1
utilized in building a Produce design solutions 1
system of the type to which Evaluate designs against requirements 1
the theory relates. User evaluation of functionality, reliability, usability,
and efficiency 1
Iterative development process 1
Utilize service organizer folder 2

Testable hypothesis Hypotheses that are open Systems built using the ISDT for SDSS will better meet
to empirical testing the needs of both the user and organization.
The level of adoption for a system built using the ISDT
for SDSS will be higher than other approaches.
Note: 1 concepts and theories ontology; 2 geographic services ontology

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Ontologies. for. Spatial. Information...................

System. Development

Following the guidelines for building spatial information system ontologies (Linkova,
Nedbal, & Rimnac, 2005), several ontologies were developed and used in the cre-
ation of the ISDT for proximate commuting outlined above. These ontologies are
accessible through a Web-based portal for spatial information system development.
This portal contains the following ontologies, as well as, those developed by other
researchers related to spatial information system development (Drexel University,
2005; Jet Propulsion Laboratory — NASA, 2005; Umweltbundesamt — Austria,
2005; University of Muenster, 2005).

Spatial.Information.System.Ontology

The use of ontologies in information systems (Guarino, 1998) and in spatial infor-
mation systems has been discussed in Fonseca, Egenhofer, Agouris, and Câmara

Figure 3. Spatial information system ontology — proximate commuting system

instance

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

(2002), Fonseca, Egenhofer, Davis, and Câmara (2002), Lin and Ludäscher (2003),
and Weissenberg, Voisard, and Gartmann (2004). The following example builds on
the ideas and concepts of these researchers and presents a methodology for spatial
information system development. This methodology assists developers in determining
how best to organize and incorporate the various ontologies into a coherent, useful,
ontology-based ISDT for spatial information systems. Figure 3 is an illustration of
this ontology as developed in Protégé.
In the above example, various ontologies related to spatial information system
development have been merged into an instance of a spatial information system
ontology using the application domain ontology as the organizing framework. These
ontologies, detailed below, include concepts and theories, geographic services, and
advanced traveler information systems. As seen in this example, an instance of an
Intelligent transportation system, a proximate commuting system, has been designed
using these ontologies.

Concepts. and. Theories. Ontology

Spatial.Decision.Support.Systems

Decision support systems (DSS) are “interactive, computer-based systems that aid
users in judgment and choice activities. They provide data storage and retrieval but
enhance the traditional information access and retrieval functions with support for
model building and model-based reasoning. They support framing, modeling, and
problem solving” (Druzdzel & Flynn, 2000, p. 794).
One instance of a DSS is the spatial decision support systems (SDSS), which are
decision support systems where the spatial properties of the data to be analyzed play
a major role in decision-making (Seffino, Medeiros, Rocha, & Yi, 1999). A SDSS
can also be defined as an interactive, computer-based system designed to support a
user or group of users in achieving a higher effectiveness of decision making while
solving a semi-structured spatial decision problem (Malczewski, 1997).

Information.Systems.Development.Methodologies

A number of methodologies are used to develop and support information systems.

The two most well known of these methodologies are the waterfall model and the
spiral model. The waterfall model, first developed by Royce and popularized by
Boehm, was named because of the way the model was depicted (Abrams, 1998).
This methodology, known today as the systems development life cycle, is a common

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

method for systems development in many organizations, featuring several phases

that mark the progress of the systems analysis and design effort (Hoffer, George, &
Valacich, 1998). The spiral model depicts the systems development life cycle as a
series of cycles in which the development team cycles through each of the phases
at varying levels of detail (Hoffer et al., 1998).
To some observers, the waterfall model appears to focus on a linear-bounded process,
while the spiral model emphasizes cyclic recurring activities. Others have observed
that the waterfall can be rolled up or the spiral can be unrolled, thereby mapping
one model to the other (Abrams, 1998). The cyclic or iterative nature of these ap-
proaches is especially characteristic of rapid application development methods such
as prototyping (Hoffer et al., 1998).
The prototyping development methodology has been described as a revolutionary
rather than an evolutionary change in the system development process (Naumann
& Jenkins, 1982). Prototyping is a four-step interactive process between the user
and the builder. Here, an initial version is defined, constructed, and used quickly; as
problems and misfits are discovered, revisions and enhancements are made to the
working system in its user’s environment (Naumann & Jenkins, 1982).

Human-Centered.Design

One instance of an information systems development methodology is human-centered

design. As defined by the International Standards Organization, human-centered
design is an approach to interactive system development that focuses specifically
on making systems useable (International Organization for Standardization, 1997).
Human-centered design integrates a number of disciplines including information
science and the behavioral sciences. Human-centered design has the following
characteristics:

• Active involvement of users and a clear understanding of user task require-

ments
• Appropriate allocation of function between users and technology
• Iteration of design solutions
• Multi-disciplinary design

The International Organization for Standardization has outlined a methodology

(International Organization for Standardization, 1997), which includes specific
steps to take during system development to address the above characteristics. These
human-centered design activities include:

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

• Understanding and specifying the context of use

• Specifying the user and organizational requirements
• Producing design solutions
• Evaluating designs against requirements

This human-centered design process iterates through these activities until the system
meets specified functional, user, and organizational requirements.

Quality.in.Use

Quality in use is defined as the extent to which an entity satisfies stated and implied
needs when used under stated conditions (International Organization for Standardiza-
tion, 1997). Put another way, quality in use is the combined effect of the software
quality characteristics for the user (Bevan, 1999). As such, quality in use is the user’s
view of the quality of a system containing software measured in terms of the result
of using the software rather than properties of the software itself. In addition, the
relationship of quality in use to the other software quality characteristics depends
on the type of user, for instance:

• For the end user, quality in use is mainly a result of functionality, reliability,
usability, and efficiency.
• For the person maintaining the software, quality in use is a result of maintain-
ability.
• For the person porting the software, quality in use is a result of portability.

Quality in use should be the major design objective for an interactive product so
that the product can be used for its intended purpose. Increased quality in use brings
significant benefits including the following:

• Increased efficiency
• Improved productivity
• Reduced errors
• Reduced training
• Improved acceptance

Figure 4 is an illustration of this ontology as developed in Protégé.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Figure 4. Concepts and theories ontology

Geographic.Services.Ontology

The International Organization for Standardization has established a structured set

of standards for information concerning objects or phenomena that are directly or
indirectly associated with a location relative to the Earth (International Organiza-
tion for Standardization, 2000). The mandate for the International Organization for
Standardization Technical Committee (ISO/TC) 211 is to develop an integrated set
of standards for geographic information that:

• Increases the understanding and usage of geographic information

• Increases the availability, access, integration, and sharing of geographic in-
formation
• Promotes the efficient, effective, and economic use of digital geographic
information and associated hardware and software systems
• Contributes to a unified approach to addressing global ecological and humani-
tarian problems

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Hlton, Burkhard, & Abhchandan

The ISO/TC 211 expects that innovative, new, and unknown technologies and appli-
cation domains will present a challenge to the process of geographic standardization.
However, where standardization was once a process for recognizing and codifying
the status quo of technology, it is now beginning to define the requirements and
implementations of new technology (International Organization for Standardiza-
tion, 2000).
Accordingly, the OGC, in conjunction with the ISO/TC 211, has developed an
international standard that provides a framework for developers to create software
that enables users to access and process geographic data from a variety of sources
across a generic computing interface within an open information technology envi-
ronment (Open GIS Consortium Inc., 2002). The OGC expects that spatial infor-
mation system and software developers will use these standards to provide general
and specialized services for all geographic information. In fact, the geographic
services architecture specified in this international standard was developed to meet
the following purposes:

• Provide an abstract framework to allow coordinated development of specific

services

Figure 5. Geographic services ontology

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

• Enable interoperable data services through interface standardization

• Support development of a service catalog through the definition of service
metadata
• Allow separation of data instances and service instances
• Enable use of one provider's service on another provider's data
• Define an abstract framework, which can be implemented in multiple ways

The geographic services architecture specified in this international standard contains

six classes of geospatial technology services. Contained within each of these classes
are many geospatial design elements (instances); the complete list of geographic
services as outlined by the OGC can be found in Open GIS Consortium Inc. (2002).
Figure 5 is an illustration of this ontology as developed in Protégé.

Advanced.Traveler.Information.Systems.Ontology

Intelligent transportation systems (ITS) encompass a broad range of wireless and

wireline communications-based information, control, and electronics technolo-
gies. When integrated into the transportation system infrastructure, and in vehicles
themselves, these technologies help monitor and manage traffic flow, reduce con-
gestion, provide alternate routes to travelers, enhance productivity, and save lives,
time, and money (Intelligent Transportation Society of America, 2005). Examples
of ITS include:

• Advanced traveler information systems deliver data directly to travelers,

enabling them to make choices about alternate routes or modes of transporta-
tion.
• Advanced traffic management systems employ a variety of detectors, cameras,
and communication systems to monitor traffic, optimize signal timings on
major arterials, and control the flow of traffic.
• Incident management systems provide traffic operators with the tools to respond
to accidents, hazardous spills, and other emergencies.

The advanced traveler information systems (ATIS) ontology is a compilation of

several ATIS customer demand factors (traveler characteristics, trip types, and modes
of transportation) and Transitweb design resources, both outlined as follows.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

ATIS.Customer.Demand.Factors

ATIS can be defined in two ways: (1) real-time network information, whether traf-
fic or transit; and (2) traveler information, such as route guidance or destination
information, provided on advanced technologies, such as mobile phones enabled
by wireless application protocol, personal digital assistants, the Internet, and in-
vehicle computers. Consequently, traveler demand for ATIS traffic services can be
based on four factors:

1. Regional traffic context

2. Quality of ATIS services
3. Individual trip characteristics
4. Characteristics of the traveler

The regional traffic context includes attributes of the region, such as highway-road-
way network and capacity, levels of traffic congestion, and future highway-roadway
expansion plans. Information quality determines whether, how frequently, and with
what level of confidence travelers consult traveler information. The trip purpose,
the time and length of the trip, and the particular route or route choices available
to the individual traveler all significantly affect whether individuals consult traffic
information. The fourth factor includes user values and attitude characteristics,
which are important determinants of use patterns, behavioral responses, and valu-
ation of ATIS (Lappin, 2000).

Traveler.Characteristics

A recent study (Pucher & Renne, 2003) found that the private car continues to
dominate urban travel among every segment of the American population, includ-
ing the poor, minorities, and the elderly. This study, drawing on the 2001 National
Household Travel Survey, confirms most of the same travel trends and variations
among socioeconomic groups documented by previous surveys. While the private
car continues to dominate travel, there are important variations in auto ownership
and travel behavior by household income, race/ethnicity, gender, and age. Overall,
the poor, racial, and ethnic minorities, and the elderly have much lower mobility
rates than the general population. Moreover, the poor, Blacks, and Hispanics are
far more likely to use public transit than other groups (minorities and low-income
households account for 63% of public transit riders). Different socioeconomic groups
also have different rates of carpooling, taxi use, bicycling, and walking as well as
travel different distances at different times of day.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Trip.Types.and.Modes.of.Transportation

The Nationwide Personal Transportation Survey (Research Triangle Institute, 1997)

is an inventory of daily personal travel and is an authoritative resource for national
data on daily trips including:

• Purpose of the trip (work, shopping, etc.)

• Means of transportation used (car, bus, subway, walk, etc.)
• How long the trip took, that is, travel time
• Time of day the trip took place
• Day of week the trip took place, and, if a private vehicle trip
• Number of people in the vehicle, that is, vehicle occupancy
• Driver characteristics (age, sex, worker status, education level, etc.)
• Vehicle attributes (make, model, model year, amount of miles driven in a
year)

As such, this resource is used to examine travel behavior at the individual and house-
hold level including: characteristics of travel (use of the various modes, amount
and purpose of travel by time of day and day of week, and vehicle occupancy), the
relationship between demographics and travel (changes in personal mobility among
women, older Americans, and low-income households), and the public’s perceptions
of the transportation system.

TransitWeb.Design.Resources

TransitWeb, a service of the United States Department of Transportation, is an

online portal designed to bring together a variety of resources to help transit agen-
cies better communicate with their customers through Web sites and ITS-related
technologies (United States Department of Transportation, 2005a). One of these
resources, The Usability Guidelines for Transit Web Sites (United States Depart-
ment of Transportation, 2005b), outlines five basic principles for achieving usable
transit Web sites. These principles are:

• Efficiency: Minimize user time and effort to reach information

•. Clarity: Make the available content and options as clear as possible to the
user
•. Proportionality: Keep design elements proportional to their functionality

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

•. Consistency: Use consistent design elements on all pages

•. Organization: Put related things together

There are many features that transit Web sites should provide to support these prin-
ciples and TransitWeb has identified several of these design features:

•. Content.organization: Organizing, distributing, and linking content throughout

the site
•. Navigation: Design of the menu and other aids to site navigation common to
all pages
•. Home.page: Design of the home page of the Web site
•. Maps: Design of system and route maps
•. Itinerary.maker: Design of an interactive itinerary planner
•. Schedules: Design of pages for accessing and displaying route schedules
•. Place.directories: Design of directories of streets, points of interest, towns,
or other geographic features intended to provide users with a means to select
an acceptable route when planning a trip

Figure 6. Advanced traveler information systems ontology

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

•. Page. form. and. performance: Design and characteristics of site pages in

general
•. Content:.Form, format, and appearance of site content in general
•. Links: Presence, position, and appearance of links in general

In addition, TransitWeb provides specific details regarding the qualities of these fea-
tures. These qualities were transformed into the form of questions to guide the design
process. Figure 6 is an illustration of this ontology as developed in Protégé.

Application.Domains.Ontology

The application domains ontology, a high-level classification schema, represents a

large number of subject areas and encompasses a wide-range of disciplines specific
to spatial information systems and applications. For example, the metadata++ model
and system was developed as a domain-specific digital library for use in natural
resource management in conjunction with the USDA Forest Service, the USDI
Bureau of Land Management, and the USDI Fish and Wildlife Service (Delcambre
et al., 2005). Similar application domain ontologies have been developed in this

Figure 7. Application domains ontology

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

manner (Fonseca, Egenhofer, Agouris, & Câmara, 2002; Fonseca, Egenhofer, Davis,
& Câmara, 2002; Lin & Ludäscher, 2003; Stoimenov, Stanimirovic, & Djordjevic-
Kajan, 2005; Weissenberg et al., 2004). Figure 7 is an illustration of this ontology
as developed in Protégé.

Conclusion. and. Future. Research

The example presented above regarding the use and application of ontologies was
relatively straightforward to organize and manage. However, as the number of on-
tologies grows, these issues (organization and management) will become of greater
concern. Below is a discussion of one method to address these concerns.

Management.of.Ontology.Libraries

An ontology library system is a library system that offers various functions for
managing, adapting, and standardizing groups of ontologies and fulfills the needs

Figure 8. Spatial information system ontology library

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

for re-use of ontologies (Ding & Fensel, 2001). The primary objective of ontology
management systems is to provide holistic control over management activities for
ontological information by externalizing them from application programs (Lee et al.,
2004). The mapping of ontologies to conceptual schemas is made using three differ-
ent levels of abstractions: formal, domain, and application levels (Fonseca, Davis,
& Câmara, 2003). At the formal level, highly abstract concepts are used to express
the schema and the ontologies. At the domain level, the schema is regarded as an
instance of a generic model. At the application level, the focus is on the particular
case of spatial application. Figure 8 illustrates an instance of a spatial information
system ontology library with respect to the spatial information system described in
the preceding sections.
The Protégé Web browser, seen in Figure 8, allows users to browse, share, and
edit their protégé ontologies over the Internet. It also provides the functionality to
perform a text-based search of the knowledge base.
The spatial information system presented above is one example of the use of ISDT
and ontologies for the design and development of systems of this type. As the use
of these systems grows, it is hoped that this method of system design and develop-
ment will prove valuable.

References

Abrams, M. D. (1998). Security engineering in an evolutionary acquisition envi-

ronment. Paper presented at the New Security Paradigms Workshop, Char-
lottsville, VA.
Bederson, B. B. (2001). Making visualization work. (Presentation): CS/HCI Depts,
University of Maryland.
Bevan, N. (1999). Quality in use: Meeting user needs for quality. The Journal of
Systems and Software, 49, 89-96.
Delcambre, L., Nielsen, M., Tolle, T., Weaver, M., Maier, D., & Price, S. (2005).
NSF project: Harvesting information to sustain our forests. Paper presented
at the National Digital Government Conference 2005, Atlanta, GA.
Ding, Y., & Fensel, D. (2001). Ontology library systems: The key to successful ontology
re-use. Paper presented at the Infrastructure and Applications for the Semantic
Web, International Semantic Web Working Symposium, Stanford, CA.
Drexel University (2005). List of ontologies for geographic information standards
provided by ISO. Retrieved from http://loki.cae.drexel.edu%7Ewbs/ontol-
ogy/list.htm

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

Druzdzel, M. J., & Flynn, R. R. (2000). Decision support systems: Encyclopedia of

library and information science. New York: Marcel Dekker, Inc.
Embley, D. W. (2004). Toward semantic understanding — an approach based on
information extraction ontolgies. Paper presented at the Proceedings of the
Fifteenth Conference on Australasian Database, Dunedin, New Zealand.
Fensel, D. (2003). Ontologies: A silver bullet for knowledge management and
electronic commerce. Berlin, DE: Springer-Verlag.
Fonseca, F., Davis, C., & Câmara, G. (2003). Bridging ontologies and conceptual
schemas in geographic information integration. GeoInformatica, 7(4), 355-
378.
Fonseca, F., Egenhofer, M., Agouris, P., & Câmara, G. (2002). Using ontologies
for integrated geographic information systems. Transactions in GIS, 6(3),
231-257.
Fonseca, F., Egenhofer, M., Davis, C., & Câmara, G. (2002). Semantic granularity
in ontology-driven geographic information systems. Annals of Mathematics
and Artificial Intelligence, 36(1-2), 121-151.
Freeman, D., & Hart, D. (2004). A science of design for software-intensive systems.
Communications of the ACM, 47(8), 19-21.
Geroimenko, V. (2003). Dictionary of XML technologies and the semantic Web.
Berlin, DE: Springer Verlag.
Gershon, N., Eick, S., & Card, S. K. (1998, March-April). Information visualiza-
tion. Interactions, 5(2), 9-15.
Green, G., Kennedy, P., & McGown, A. (2002). Management of multi-method en-
gineering design research: A case study. Journal of Engineering Technology
Management, 19, 131-140.
Grimshaw, D. (2000). Bringing geographical information systems into business.
New York: John Wiley and Sons, Inc.
Gruber, T. R. (1993). A translation approach to portable ontology specification.
Knowledge Acquisition, 5, 199-220.
Guarino, N. (1998). Formal ontology and information systems. Paper presented at
the 1st International Conference on Formal Ontology in Information Systems
(FOIS ‘98), Trento, IT.
Hevner, A., March, S., Park, J. A, & Ram, S. (March, 2004). Design science in
information systems research. MIS Quarterly, 28(1), 75-105.
Hilton, B. (2003). The impact of open source software and Web services on informa-
tion system development: An investigation of proximate computing. Claremont,
CA: School of Information Science, Claremont Graduate University.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Hoffer, J. A., George, J. F., & Valacich, J. S. (1998). Modern systems analysis and
design. Reading, MA: Addison-Wesley.
Intelligent Transportation Society of America. (2005). What is ITS? Retrieved
from http://www.itsa.org/subject.nsf/vLookupAboutITSA/What+is+ITS!Op
enDocument
International Organization for Standardization. (1997). Human-centered design
processes for interactive systems (ISO 13407). Geneva.
International Organization for Standardization. (2000). Draft Business Plan of
ISO/TC 211 — geographic information/geomatics. Geneva.
Jet Propulsion Laboratory — NASA (2005). Semantic Web for Earth and Envi-
ronmental Terminology Ontologies. Retrieved from http://sweet.jpl.nasa.
gov/ontology/
Kreuseler, M., & Schumann, H. (1999). Information visualization using a new focus
+ context technique in combination with dynamic clustering of information
space. In Proceedings of ACM PNPIVM 1999, Kansas City, MO (pp. 1-5).
New York: ACM Press.
Lappin, J. (2000). What have we learned about intelligent transportation systems
and customer satisfaction? In J. E. Lappin (Ed.), What have we learned about
intelligent transportation systems? For the ITS Joint Program Office, Federal
Highway Administration (JPO Doc. No. 13390). Washington, DC: United States
Department of Transportation, Federal Highway Administration.
Lee, J., Goodwin, R., Akkiraju, R., Ranganathan, A., Verma, K., & Goh, S. W.
(2004). Towards enterprise-scale ontology management. Hawthorne, NY:
IBM T. J. Watson Research Center.
Lin, K., & Ludäscher, B. (2003). A system for semantic integration of geologic
maps via ontologies. Paper presented at the Semantic Web Technologies for
Searching and Retrieving Scientific Data, Sanibel Island, FL.
Linkova, Z., Nedbal, R., & Rimnac, M. (2005). Building ontologies for GIS (No.
932). Prague, CZ: Institute of Computer Science, Academy of Sciences of the
Czech Republic.
Malczewski, J. (1997, October 6, 1998). Spatial decision support systems, NCGIA
core curriculum in GIScience. Retrieved from http://www.ncgia.ucsb.edu/
giscc/units/u127/u127.html
Markus, L., Majchrzak, A., & Gasser, L. (2002). A design theory for systems that
support emergent knowledge processes. MIS Quarterly, 26(3), 179-212.
Naumann, J. D., & Jenkins, A. M. (1982). Prototyping: The new paradigm for
systems development. MIS Quarterly, 6(3), 29-44.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Hlton, Burkhard, & Abhchandan

Noy, N. F., & McGuinness, D. L. (2003). Ontology development 101: A guide to

creating your first ontology. Retrieved from http://protege.stanford.edu/
NSF. (2004). Science of design for information systems: Report of the NSF work-
shop, Seattle, 2003. ACM SIGMOD Record, 33(1), 133-137.
Open GIS Consortium Inc. (2002). The OpenGIS Abstract Specification, Topic
12: OpenGIS Service Architecture, Version 4.3. Wayland, MA: Open GIS
Consortium, Inc.
Orlikowski, W., & Iacono, C. (2001). Desperately seeking the “IT” in IT research
— a call to theorizing the IT artifact. Information Systems Research, 12(2),
121-134.
Pucher, J., & Renne, J. (2003). Socioeconomics of urban travel: Evidence from the
2001 NHTS. Transportation Quarterly, 57(3), (49–77).
Research Triangle Institute. (1997). Nationwide personal transportation survey
(No. FHWA-PL-98-002). Research Triangle Park, NC: Research Triangle
Institute.
Rodriguez, D. A. (2001). Proximate commuting: Hype or potential? An evaluation
(No. Transportation Research Record 1675). Washington, DC: Transportation
Research Board, National Research Council.
Seffino, L. A., Medeiros, C. B., Rocha, J. V., & Yi, B. (1999). WOODSS — a spatial
decision support system based on workflows. Decision Support Systems, 27,
105-123.
Shneiderman, B. (1999). Crossing the information visualization chasm. Retrieved
March 15, 2004, from http://www.cs.umd.edu/hcil/pubs/presentations/info-
viz-chasmslides/
Shneiderman, B., Card, S. K., Norman, D. A., Tremaine, M., & Waldrop, M. M.
(2001). CHI@20: Fighting our way from marginality to power. In ACM CHI
2002 Conference on Human Factors in Computing Systems (pp. 688-691).
Minneapolis, MN: ACM Press.
Simon, H. (1996). The sciences of the artificial (3rd ed.). Cambridge, MA: MIT
Press.
Smith, B. (2003). Ontology. In L. Floridi (Ed.), Blackwell guide to the philosophy
of computing and information (pp. 155-166). Oxford, UK: Blackwell.
Stanford Medical Informatics. (2003). Protégé-2000 user’s guide. Retrieved from
http://protege.stanford.edu/
Stoimenov, L., Stanimirovic, A., & Djordjevic-Kajan, S. (2005). Development of
GIS interoperability infrastructure in local community environment. Paper
presented at the FIG Working Week 2005, and the 8th International Conference
for Global Spatial Data Infrastructure (GSDI-8), Cairo, EG.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Informaton System Development 

Sugumaran, V., & Storey, V. C. (2002). Ontologies for conceptual modeling: Their
creation, use, and management. Data Knowledge and Engineering, 42(3),
251-271.
Truex, D. (2001). Three issues concerning relevance in IS research: Epistemology,
audience, and method. Communications of AIS, 6(24), 1-10.
Umweltbundesamt — Austria. (2005). Umweltbundesamt: CEDEX Protege. Retrieved
from http://www.unweltbundesamt.at/umweeltdaten/schnittsellen/cedex/ce-
dex_protege/?&tempL=1
United States Department of Transportation. (2005a). TransitWeb: A Website de-
sign and traveler information resource for transit agencies. Retrieved from
http://www.transitweb.its.dot.gov/Introduction.asp
United States Department of Transportation (2005b). Website Usability Guidelines
— an online handbook for making public transit Web sites easy to use. Retrieved
from http://www.transitweb.its.dot.gov/guidelines/complete_guidelines.asp
University of Muenster. (2005). Ontologies developed in ACE-GIS. Retrieved from
http://musil.uni-muenster.de/onto/ACE/
Vaishnavi, V., & Kuechler, W. (2004, January 20). Design research in information
systems. Retrieved from http://www.isworld.org/Researchdesign/drisISworld.
htm
Victoria Transport Policy Institute. (2005). Online transportation demand manage-
ment encyclopedia. Retrieved from http://www.vtpi.org/tdm/index.php
Walls, J. G., Widmeyer, G. R., & El Sawy, O. A. (1992). Building an information
system design theory for vigilant EIS. Information Systems Research, 3(1),
36-59.
Weissenberg, N., Voisard, A., & Gartmann, R. (2004). Using ontologies in personal-
ized mobile applications. Paper presented at the 12th Annual ACM International
Workshop on Geographic Information Systems, Washington, DC.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton, Burkhard, & Abhchandan

Section II

Challenges for Spatial

Information Systems

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

Chapter.III

Geospatially.Enabled.
Directory.for.Emergency.
Response.Interoperability
Judth Woodhall, COMCARE, USA

Abstract

This chapter discusses how the need for geospatially-enabled data messaging
among emergency response agencies can be enabled with the Emergency Provider
Access Directory (EPAD). It describes the directory, how it enables message rout-
ing, and its fit into a broader E-Safety Network. It also discusses the architectural
components of the EPAD, specifically the geographic information system (GIS)
module, and how Web services and open source products were used in the design
to enhance the EPAD service offering, an offering that has the potential to bring
emergency agencies one step closer to realizing interagency interoperability for
advanced response to emergency events.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

Introduction

In an era when technology can bring news, current events, and entertainment to the
farthest reaches of the world, many emergency response agencies cannot share data
with one another, even if they are in the same jurisdiction. Yet, emergencies demand
real-time data. Most of today’s efforts to improve interoperability have been focused
on wireless voice communications. Led by state and local public safety experts and
supported by the Department of Homeland Security (DHS), SAFECOM program,
and significant federal funding, wireless first responder interoperability is moving
forward. While voice or radio interoperability is a critical need for responders at the
scene, voice represents only one side of the interoperability equation.
COMCARE is a national non-profit alliance dedicated to advancing emergency
response by promoting modern, interoperable emergency communications systems,
and the development of new procedures, training, and tools to maximize value for
emergency responders. COMCARE encourages cooperation across professional,
jurisdictional, and geographic lines, and works to integrate the emergency response
professions, government, private industry, and the public. Its vision is to create an
environment of borderless, geographically-targeted information sharing to achieve
the most advanced response to emergencies. To that end, it has created the E-Safety
Network, a framework for establishing a unified emergency Web services informa-
tion architecture that ties together the various data systems used by all emergency
response organizations.1

Background

Even after September 11, 2001, Columbine, the Northeast blackout, the California
wildfires, and the recent hurricane seasons, there is still no national comprehensive
emergency infrastructure that allows agencies to communicate data with one another
across professions and across jurisdictions. Time and time again, important infor-
mation about an emergency event does not reach the right people at the right time,
resulting in unnecessary loss of life and property. During the 2003 SARS (severe
acute respiratory syndrome) outbreak in Toronto, first responders and emergency
medical personnel learned of the outbreak through the media, not from public health
officials, hospitals, or other medical organizations (SARS Commission, 2004).
During the Oklahoma City bombing incident, responders lacked important informa-
tion needed to understand damage from the blast and where victims might be found.
According to the former assistant chief of the Oklahoma City Fire Department, it
would have been useful to have blueprints, personnel lists, and data estimating blast
effects and impacts to the buildings early in the response process. Other commu-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

nications problems also hindered response efforts. Because of overload and chaos,
voice communications were unreliable. In addition, the 9-1-1 center did not have a
direct line to the Emergency Medical Services (EMS) agency, and hospitals were
not even in the communications loop. According to the Police Department Final
Report, this situation highlighted the need for electronic data links between emer-
gency agencies and the need to extend data access into the field (Manzi, Powers,
& Zetterman, 2002).
On January 6, 2005, at 2:40 AM, two freight trains collided in Graniteville, South
Carolina, releasing an estimated 11,500 gallons of chlorine gas (CDC, 2005). Two
hours later, the public information director was asked to issue a federal Emergency
Alert System (EAS) warning (Lipowicz, 2005). The director used the telephone
to issue the warning to local radio and TV stations, but he could not be certain
it actually reached anyone. Because of the early hour, many radio stations were
not staffed, and there was no guarantee that residents were listening to the radio
or watching TV. The Federal Communications Commission (FCC) mandates that
all TV and radio stations are required to carry any presidential warning message.
However, when operated regionally, it relies on voluntary participation from radio
and broadcast television stations (FCC, 2004). And its reach is limited. The system
does not include all the other communications devices that consumers have (from
cable TV to personal computers with Internet access to telephones, cell phones, and
beepers). Most residents did not know about the spill until they woke up and had
limited time to react. Unfortunately for them, chlorine gas is so deadly that it was
used as a chemical weapon in World War I. The gas is heavier than air and sinks to
the ground. Unknowingly, some Graniteville residents dropped down to the ground
to get air as they might in a smoky fire (Hajek, Shapiro, Cooper, & Lewis, 2005).
This mistake cost them their lives. A better, more targeted, and comprehensive public
warning system could have helped (Lipowicz, 2005).
In all these events, emergency agencies and responders lacked vital information
about who should be contacted and how. In some instances, voice communications
did not work; in others, voice was inadequate. Data communications facilitated by
a routing service could have helped to deliver the right information to those who
needed it.
Of course, many agencies do exchange data electronically. Unfortunately, they either
do so within their own professional silos (law enforcement to law enforcement, fire
to fire, etc.) or they use the same software products. If agencies want to share data
but do not want to procure the same software, they must absorb the costs of custom-
izing an interface and developing their own routing directories so that they know to
whom, to where, and how information should flow. As many agencies have learned,
this customization scenario only leads to software incompatibility and obsolescence
when one or both agencies buy new software or install upgrades. In addition, the
cost of keeping a routing directory up-to-date can sky rocket.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

It is impractical to expect all of the 100,000+ emergency agencies in the United

States to use the same software or to spend limited funds developing customized
interfaces and proprietary data routing utilities. However, many agencies have done
just that. The result has been multiple specialized directories of varying levels of
completeness and accuracy; all failing to provide a practical and reliable basis for
routing emergency information immediately to all of the right agencies, particularly
when multiple professions or jurisdictions are involved.
Another facet of the problem is that few, if any, of the existing directories can be
queried in geospatial terms. Moreover, these directories generally function in a
“top-down” sense, operating through telephone trees and “blast-fax” systems, or at
best, mass e-mail alerts, and are often effectively useless for intercommunication
between parties at “ground level” or for timely relay of information back to deci-
sion-makers at the top.
The lack of a shared, comprehensive routing directory is not only a problem during
major emergencies and disasters; it is also increasingly evident that it would have
significant value in day-to-day emergency operations. For example, the consumer
and commercial telematics providers, such as OnStar2 and Qualcomm,3 need to
deliver vehicle incident data, immediately after a theft or accident, to a number of
agencies such as law enforcement, fire, emergency medical services, hospitals, and
transportation, but either do not know who those agencies are, or their computer
addresses, or, usually, both. In another incident, an urgent alert or public warning
may need to be delivered to recipients in a geographic area, which may or may not
correspond to any predetermined jurisdictional or physical boundary. In addition,
in a large-scale disaster there is an immediate and massive requirement to marshal
state and local first responder resources. Each of these scenarios requires a directory
to route information based on the geography of the emergency event. COMCARE’s
E-Safety Network enabled by the Emergency Provider Access Directory4 (EPAD)
can provide one solution.

Architecture. Overview

Architectures can be broken down into macro and micro levels. At the macro level,
enterprise architectures provide a big picture view and encourage holistic under-
standing of various parts that make up the architecture. They manage the complexity
of the enterprise and align business strategies with implementations. At the micro
level, system architectures define the form and function of a system by defining
discrete, non-overlapping parts or components. The E-Safety Network provides
the enterprise architecture for interoperable emergency data communications that
includes the EPAD system.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

Figure 1. E-Safety Network: Five architectural layers

E-Safety.Network

The E-Safety Network is a unified emergency Web services information architecture

that ties together the various application systems used by law enforcement, fire,
emergency medical, public health, transportation, emergency management, and
homeland security agencies, among others (COMCARE, October, 2004). It treats
the entire emergency response community as one extended enterprise, providing a
framework within which each entity can operate. It is open, platform-neutral, and
standards-based to enable flexible interoperability across organizations, technology
platforms, and implementation methods.
There are five architectural layers (Figure 1) that must be in place to achieve effective
data interoperability. These layers include data transport, data standards, shared
facilitation services, individual agency applications, and the policies and protocols
that govern the use of the system when data interoperability is achieved.
The transport layer represents the networks used for communications. This layer
manages the end-to-end delivery of messages and determines how data are transferred
between network devices. It manages user sessions and dialogues and controls the
establishment and termination of logic links between users. The E-Safety Network
requires reliable and secure broadband data connections using Internet protocols.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

Data standards create a common language that enables data sharing among indi-
vidual agency application systems. The Department of Homeland Security Disaster
Management eGov Initiative has launched a program to facilitate standards develop-
ment by bringing together leaders from all professions needing to share data during
emergency response operations. These practitioners develop and field test common
sets of emergency message standards. Over the last year, this and related projects
have resulted in many eXtensible Markup Language (XML) standards such as the
common alerting protocol5 (CAP), the Vehicular Emergency Data Set6 (VEDS), and
the Emergency Data Exchange Language7 (EDXL) suite of standards (XML, 2005).
These non-proprietary XML message formats use simple object access protocol
(SOAP) “packaging” for both specialized and general emergency data exchange.
For content, they draw on the dictionary and data model work of other efforts such
as the Global Justice XML Data Model (OJP, 2005) and the new DHS National
Information Exchange Model (NIEM) (NIEM, 2005).
Facilitation services are common shared tools, services, and resources offered
through a collective effort of the emergency response community. They enable
interoperability and are available for use by authorized emergency entities. These
services include, but are not limited to, security, diagnostics, routing directory, iden-
tity management, access control, digital rights management, and authentication. By
using these shared facilitation services, agencies do not have to spend their limited
funds creating and maintaining these functions on their own. EPAD as a routing
directory is one of these services.
The fourth layer represents the wide array of agency applications used for emergency
preparedness, response, and recovery. These systems include complex computer-
aided dispatch systems (CAD), Web-based emergency management tools, local and
statewide geographic information systems (GIS), hospital capacity reporting systems,
and other innovative applications for data collection, analysis, and presentation.
Agencies are encouraged to purchase systems that best meet their needs. However,
in order to operate within the E-Safety Network, these applications must be able
to send and receive XML messages to and from other applications in standardized
formats. It should not matter to a 9-1-1 CAD system that it is receiving data from
an emergency management tool about a flood, a telematics message from OnStar,
a bio-terrorism alert from the Center for Disease Control (CDC), or data from a
wireless or Voice over Internet protocol (VoIP) call. The same standardized data
interface should be used.
Lastly, the E-Safety Network is not complete without the policies and protocols that
determine rules for operating within it. Does a hospital have the same privileges
as the county Department of Transportation (DOT), the 9-1-1 center, the police,
or the towing company? Who has access to what data and who is allowed to send
what messages? Some of these policies are already in place today. For instance, a
9-1-1 center is required to alert certain agencies when there is a hazardous materials
(HAZMAT). A fire chief becomes the incident commander if there is an explosion,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

and then sets up an Incident Command System (ICS) with required sections and
reports. Emergency Medical Services (EMS) may only administer certain kinds of
treatment to patients in ambulances. Only the secretary of DHS or his designee can
change the alert level, or communicate it. Only the president or his designee can
activate the Emergency Alerting System. These are only a few examples of the types
of policies and protocols needed for data interoperability. Many more still need to
be developed before this type of architecture is deployed.

Emergency.Provider.Access.Directory

The Emergency Provider Access Directory is a facilitation service as described

above. As a routing directory, it is the cornerstone of the E-Safety Network because
without it, the routing of geographically-targeted emergency messages could not
occur. The idea is simple: Create a directory that allows authorized emergency
response organizations (not individuals) to register so they can receive informa-
tion about emergency events for which they have a need or interest. Registration
enables agencies to receive the right information when they need it; it prevents the
receipt of unwanted or irrelevant information. This simple act of EPAD registration
supports the automatic routing of vital information about a mass emergency or a
single event quickly and securely.
During registration, agencies and other organizations involved in emergency response
efforts choose what information (emergency event types8) they want to receive
for which geographic areas. They indicate when and how they want to receive it,
and they classify their information so that only those with the proper authority can
access it. Once entities are registered, messaging systems (or application systems
with messaging capabilities) can query EPAD to determine to which agencies they
need to send information. EPAD returns a list of agencies along with their computer
addresses and prioritized contact information so that messages can be automatically
routed to responsible agencies based on their geographic areas of responsibility and
their incident information preferences.

EPAD. System.Architecture

The EPAD system follows an N-tier architectural design (Figure 2) that separates
software into different functional areas: presentation layer, business logic layer,
common business layer, infrastructure service layer, and data access layer. The Web
services integration layer within the EPAD system allows for a standardized inter-
face to be exposed and provides data for applications that require information from

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Woodhall

Figure 2. EPAD N-tier system architecture

Figure 3. EPAD system view

the EPAD. The components that make up the N-tiered system, including the data
sources, will not expose any remote interface to external applications. All external
applications will communicate to the EPAD through the Web services interface
(Figure 3) (COMCARE, June, 2005).
The functional architecture of the EPAD identifies six distinct Web service compo-
nents as well as three standard XML interfaces that interact with the EPAD system.
The EPAD Web service components include:

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

•. Administration.services: for agency and user set-up

•. Directory.services: for housing agency information
•. Geographic.information.system.(GIS).services: for establishing jurisdictional
or interest area boundaries
•. Identity.rights.management.services:.for authentication and authorization
•. Import.services: for the automatic uploading of agency and GIS information
into the EPAD when agencies already have this information electronically
from other in-house systems
•. Directory.management.services: for ensuring data integrity and reporting
on directory activity

The three standard XML interfaces include:

•. Incident.sending.and.routing.query: where the EPAD is queried to deter-

mine if the agency is authorized to send an incident message and who within
a given area should receive the message
•. Authentication/authorization.query: to determine the authenticity of and
privileges granted to a given agency/user
•. GIS.query: to establish, view, and name specific geographic areas

While all of these components are needed, the geographic information system (GIS)
services module is the most critical for the agency notification capability enabled
by the EPAD. It allows agencies to specify jurisdictions and geographic areas of
interest pertinent to their responsibility. Using the GIS services module, agencies
attribute these areas of interest with event type(s), responsibility type (primary,
support, advisory, or informational), and the timeframes in which their interests are
active. For example, a fire department may need to respond to fire emergencies for
multiple counties, but only has EMS responsibility within one county. It may also
provide backup services for a third county during certain timeframes. All of these
conditions must be identified during the registration process so that the system can
accurately identify who needs to be notified when.

EPAD.Technology.Components

COMCARE’s goal is to design EPAD using open source technology whenever pos-
sible. The use of open source technologies was driven by COMCARE to achieve
interoperability with other systems, scalability in terms of performance and cost,
reusability, and minimal vendor lock-in to a particular platform. Table 1 lists the

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

Table 1. EPAD solution set; open source solutions are in italics

Logical.Component Product Version

Operating.System Red Hat Linux Enterprise AS Enterprise AS 4.0
Web.Server Apache Web Server Apache 1.3 & 2.x
Web.Container Tomcat Tomcat 4.x & 5.x
Application.Server JBoss Application Server JBoss AS 4.x
PostgreSQL (including Spatial
Database PostgreSQL) PostgreSQL 8.x

LDAP.Server Sun ONE Directory Server* Sun ONE DS 5.2+

Access.Manager.and.Identity. Oracle/Oblix COREid, COREid: 7.0.3+, SHAREid:
Manager SHAREid, and COREsv 2.5+, COREsv 4.+
PostGIS PostGIS 1.0
MapServer MapServer 4.4.2
GeoServer GeoServer 1.2.4
GIS.Components
IMF Enterprise Edition (for
Moxi Media IMF
OpenGIS)

TopoZone Web Mapping Service GeoZone OGC WMS

Note: * Sun Directory Server was used instead of Open LDAP because the solution selected for access control
and identity management did not support Open LDAP.

open and proprietary products used in the EPAD system. These products reside on
different servers and are used to support all EPAD functions. Figure 4 shows the
interactions among the components and the systems on which they reside.
Tomcat is an open source product that is commonly referred to as a “Web” or
“servlet” container and has the ability to run on the chosen Linux operating system.
It implements the servlet and the Java Server Pages (JSP) specifications from Sun
Microsystems. Since Tomcat does not fully support the Java 2 Enterprise Edition
(J2EE) and is unable to execute Enterprise Java Beans (EJBs), the JBoss application
server is used as the EJB container.
PostgreSQL is one of the recommended open source database solutions for Linux.
It is an object-relational database management system (ORDBMS) based on
POSTGRES, Version 4.2, developed at the University of California at Berkeley
Computer Science Department. It is highly scalable and is optimized to store and
process geospatial data.
Public, free spatial data is being used to define agency boundaries nationwide.
The data, known as topologically integrated geographic encoding and referencing
(TIGER) data, is provided by the United States Census Bureau and can be used to
derive layers for state, county, local government, and tribal jurisdictions. TIGER
boundary data for the EPAD are extracted and loaded using open source PostGIS9
which spatially enables the PostgreSQL object relational database.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatially Enabled Directory for Emergency Response Interoperability 73

Figure 4. EPAD system layout

EPAD System

Apache Tomcat JBoss J2EE

Browser Web Server J2EE Web EJB Container PostgreSQL
Requests Container Database
PostGis Server
Struts
MapServer
Web
Service Axis GeoServer SunONE
Requests LDAP Server

Common Services
Logging
Security
Error Handling
Performance Monitoring

MapServer, an open source map generation component, is used to create maps for
interactive EPAD applications. MapServer is accessed using the Open Geospatial
Consortium (OGC) Web Map Service (WMS) protocol (2004). Moxi Media™ is an
OGC compliant, non-open source rapid development environment written primar-
ily in JSP. Moxi Media™ is used in the EPAD to support the interactive mapping
portion of the EPAD and to generate map views and map thumbnails.
Base mapping is being provided via a remote WMS service provided by TopoZone.
When an agency map thumbnail is created in the EPAD for an interest area, separate
WMS calls are made, one local to display the jurisdiction boundaries, one remote
to TopoZone to generate a basemap. Then the two images are overlaid to create the
map thumbnail in the EPAD. In the interactive mapping portion of the EPAD appli-
cation, the boundary and base map images are retrieved and overlaid as transparent
graphic image files to create the map view.
Table 2 summarizes the national geospatial standards used in the EPAD GIS module
(COMCARE, June, 2005).

EPAD GIS Services Module

There are two main components in the EPAD GIS services module: an administra-
tion interface to support visual interaction with geospatial information, and a query
interface to support the Web services queries to the GIS database.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

Table 2. National geospatial standards used in EPAD

National.Geospatial.Standards
GML 3.1.1 https://portal.opengeospatial.org/files/?artifact_id=7174 OGC’s XML-based
Geography Markup Language to describe geospatial language used as primary
information (v2.x) with enhanced features (v3.0 – 3D, definition of geospatial
temporal, etc) information stored in
PostGIS server and
interchanged with third
parties
WFS 1.0 & 1.1 http://geoserver.sourceforge.net/html/index.php Used as communication
http://www.opengeospatial.org/specs/?page=specs media with GeoServer
Web Feature Service interface specification to define
communication interface with WFS-enabled servers to
perform transactions on GML-based data (get, describe,
lock, transaction) over HTTP
WMS 1.3 http://mapserver.gis.umn.edu/doc.html, http://www. Used as communication
opengeospatial.org/specs/?page=specs media with MapServer
OGC’s web map service interface specification to define to retrieve map layer
requests for mapping information retrieved over HTTP. information or map
Capable of retrieving multiple map layers. snapshots based on geo
point definitions (lat/long,
projection, output type
— JPEG, GIF, etc)
SFS 1.1 http://www.postgis.org/ Used for storing of
Simple features specification for SQL is OGC’s geospatial information in
specification to represent data model for storing of PostGIS Server
geospatial information using SQL92 with geometry types.

The administration interface presents a series of pull-down menus that allow an

agency to select jurisdictional polygons and associate them with information about
the agency and its emergency response capabilities. The combination of these
polygons with information about an agency’s capabilities forms the spatial EPAD
database. The second component is the query support for the Web service interface.
The EPAD Web service allows organizations that have defined an incident location
to pass location information and event type information to EPAD, causing EPAD to
spatially query the EPAD database and develop lists of agencies for alert purposes.
The message sent to EPAD from these agencies contains spatial object(s) in OGC
well known text (WKT) format (see Figure 5 for examples) representing the loca-
tions of incidents as geographic primitives (points, lines, and polygons). WKT is
then used in the OGC simple feature specification for SQL format (1999) that is
used to query the database.
When an emergency event occurs and a system needs to send an incident message,
that system sends a Web service query (Figure 6) to the EPAD and its respective

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatially Enabled Directory for Emergency Response Interoperability 75

Figure 5. WKT examples

Figure 6. Incident routing query

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
76 Woodhall

GIS services module giving the event type, the time of the event, and the location
of the event. This module matches the location information in the request to agen-
cies registered in the corresponding area to determine which agencies should be
receiving this information. When matching requests to registrants, the GIS module
must accurately translate and transform the location information of the request to
locations on file. Since many agencies use different GIS systems, the EPAD GIS
services module must be able to accurately “translate” the location information it
receives so that the proper agencies are notified.
Figure 7 shows how a Web services query is satisfied using the EPAD architecture.
A first responder sends a message containing a location in Well Known Text to the
EPAD. The EPAD parses the message and generates a spatial SQL query to send
to the database. The EPAD system adds a buffer to the query to handle horizontal
error in the TIGER boundary data as well as error (in as far as it can be determined)
in the incoming message. The query is then run against all data layers simultane-
ously, and returns matching agencies that desire notification. This list is passed to
a messaging system10 that sends out alerts according to the agencies’ instructions
received from the EPAD (COMCARE, October, 2004).

Accuracy of GIS Data

TIGER data is developed by the United States Census Bureau to support the map-
ping and related geographic activities required by the decennial census. The census
requires a high level of relative accuracy, but does not place a premium on absolute

Figure 7. Translating a Web service query

First responder describes and
locates incident in messaging
system Query sent to EPAD, “drills down” through boundary layers, returns list of
agencies for that location.

System sends query contain-

ing Location and incident
information to EPAD

Location buffered and used

in spatial query Agency Codes
Retruned by GIS
EPAD identi- E PA D r e -
State
State FIPS Code fies agencies turns agency
County within the list with no- Messag-
County FIPS Code given area tification in- ing system
that want no- structions to sends out
Tribal tification for messaging
Census Code alert
this incident systems
Muni Place or Subdivision
FIPS

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

horizontal accuracy.11 When messages containing spatial coordinates are passed

to the EPAD’s Web services interface, the coordinates are used in a spatial query
against the EPAD database. Because of known horizontal error in the TIGER data
and because of much less predictable error in the incoming coordinate data, there
is a chance that all of the desired agencies will not be returned by the EPAD system
without the addition of an error factor. This factor makes the spatial object of the
incoming message larger before passing it to the EPAD.
The key question is how much larger to make the spatial query to achieve the desired
result of returning all appropriate agencies all of the time. The GIS term for mak-
ing the spatial object larger is “buffering.” If the buffer added to the object is too
small, the system risks not returning an agency that should have been returned and
subsequently notified. Obviously, depending on the type of event, a missed agency
notification could have serious consequences. If the buffer is too large, a large number
of agencies nowhere near the incident are returned. For the EPAD, it is probably
more important to capture all agencies registered for notification in a larger area and
take a chance that extra agencies are notified rather than miss an agency.
Nevertheless, a reasonable estimate of the maximum horizontal error for both the
TIGER and incoming coordinates must be developed. The National Map Accuracy
Standard (NMAS) developed by the United States Geological Survey may be used
as a starting point to define horizontal error for the two sources (TIGER and the
location information in the request). NMAS defines maximum acceptable horizontal
error for datasets at various scales such that 90% of well-defined points fall within
the acceptable error level for data of a given accuracy level. The challenge is to ar-
rive at a buffer value that delivers the correct set of agencies (from a spatial query
perspective) close to 100% of the time.
According to the U.S. Census Bureau, TIGER, which was developed by hand digi-
tizing 1:100,000 scale paper survey sheets, generally meets national map accuracy
standards for 1:100,000 scale data. This means that 90% of well-defined points
should fall within 166.7 feet of the feature they are describing. If one were to use
166.7 feet as a buffer, 10% of the time one could expect a spatial query to miss one
or more agencies. In a study that compared global positioning system (GPS) gathered
point locations corresponding to features in TIGER such as road intersections, the
displacement between the original map and the GPS-enhanced road positions was
often as large as 210 feet (Liadis, 2000). Clearly the buffer distance for TIGER will
need to be greater than 166.7 feet.
More problematic is the coordinate information that comes to the EPAD as part of
Web service queries. By requiring additional information in the Web service mes-
sage, it may be possible to develop a rule-based approach to improving accuracy.
For example, information in the message for “technique used to derive coordinates”
could be used to estimate the size of the buffer to be applied. If data describing the
accuracy of the coordinate being sent can be included in the Web services message,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

it would be possible to develop custom buffers on-the-fly based on the accuracy

defined in the message.
Assuming the NMAS standard represents one standard deviation, it is also possible
to calculate buffers to fit any desired confidence level. If the accuracy of the incom-
ing coordinates is unknown, this approach can be applied to incoming coordinate
values. The two error levels expressed in feet may then be added together to obtain
a total buffer distance to apply. This would result in a one-size-fits-all approach for
handling the total expected horizontal error. Both buffer calculation techniques are
planned for the EPAD.
Another type of error inherent in TIGER data is attribute error. Attribute error
results when a polygon does not have the correct descriptive database information
linked to it. In the case of this project, key attribute fields are federal information
processing standards (FIPS) codes and name information. For the EPAD, attribute
error is estimated using a sampling approach, which randomly selects records, then
compares the results to some known source for boundary information.

EPAD. Next. Generation

While the system design and components identified here are adequate for current
EPAD processing, changes will need to be made as the EPAD is extended for in-
formation sharing during international emergency events. Immediate consideration
needs to be given for the inclusion of boundary data for Mexico and Canada so
that incidents traversing borders can be handled in the same manner as any other
incidents. Boundary data sources for both countries need to be identified. While
census organizations exist in Mexico and Canada, it is not known exactly how the
data provided by those organizations compare to the U.S. Census TIGER dataset.
Some questions that need to be considered when incorporating data from Mexico
and Canada include the spatial accuracy of each country’s census data, their use
of FIPS or equivalent codes, the data formats used, licensing restrictions, and each
country’s update cycles.
Political boundary data for both Canada and Mexico are currently available in some
form, and data samples appear to match the United States fairly closely in terms of
jurisdiction hierarchy. While the spatial fit appears close when viewing all of North
America, there are gaps and overlaps between country datasets. A brief examination
of the border data match up between the U.S. and Mexico found gaps as large as 5
kilometers between country border lines, as well as large areas of overlap. An examina-
tion of the match between the U.S. and Canada found significant gaps and overlaps,
although not quite as large as those found between the U.S. and Mexico.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

In most cases, it will not be possible to determine which country’s boundary is

correct for a given location along a boundary line. A relatively simple rectification
process would involve choosing one boundary and correcting the other boundary
to match the first. In order to determine the best approach for addressing border
error between the U.S., Mexico, and Canada, careful consideration will be given
to the ways in which the EPAD will be used for border-related events and the con-
sequences of incorrect results being returned by queries. This consideration will
include an assessment of the legal consequences of incorrect queries. (Of course,
an understanding of the legal repercussions relating to incorrect query results is
important to consider in relation to all areas of the EPAD system.)
Another consideration for the project will be deciding the best way to provide a
basemap service. The first option, currently in use, is to purchase an existing basemap
service such as TopoZone. The second option is to build a basemap service in-house.
The two approaches have their advantages and disadvantages in different areas.
A commercial solution provides access to commercially-provided databases that
are more complete than TIGER in terms of road coverage for fast developing areas
of the country. However, TopoZone, the product offering selected for the EPAD,
does not provide basemapping services for Mexico or Canada. In addition, a remote
service such as TopoZone is accessed across the Internet and therefore could affect
system performance as Internet traffic patterns vary.
By building an in-house basemap service, a reasonably consistent basemap across
the U.S., Mexico, and Canada can be created by using a combination of data sources.
Also, an in-house service will, assuming loads are well balanced, provide much
faster response times than an Internet-based service. It will also give more control
when it comes to system stability and scalability issues. Unless another commercial
product offering can provide the same benefits, it is likely that a custom-developed
basemap service will be deployed.

Open. Source.Advantage

COMCARE’s goal in requesting the use of open source technologies for the EPAD
was the result of two major factors. First and foremost, it was to reduce its de-
pendence on particular software vendors and their respective upgrade schedules.
COMCARE, along with EPAD users, wanted the freedom to do what it wanted to
do when it wanted to do it. Secondly, while open source software is not “free”, it
does lower the total cost of ownership (TCO) particularly when the EPAD has to
scale quickly during national and international roll-outs. As a nonprofit wishing
to operate the EPAD as a common utility, a lower TCO is important. COMCARE
knows it has to pay for application system administration and support costs, the

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Woodhall

fact that it did not have the initial software license costs and will not have the an-
nual maintenance fees weighed heavily in the decision. Of course, only mature
and established open source offerings were chosen with hopes that the value of the
decision will be preserved long term.
When COMCARE first suggested the open source option in early 2004, many felt
this approach might be difficult to achieve. That was then. In the May, 2005, issue
of Government Technology, open source offerings are becoming mainstream even
in the public sector. Mississippi is using open source software for a new mobile data
infrastructure called the Mississippi Automated System Project (MASP). Chicago
is using it for a vehicle registration system that will issue over 1 million vehicle
stickers per year (Peterson, 2005). Only time will tell if these open source decisions
were good ones.

Conclusion

While the EPAD is not a production system as yet, many scenario-based demonstra-
tions of the E-Safety Network and EPAD have been conducted. The results are always
the same. Data messaging is an important and necessary component of emergency
response efforts. If an EPAD-enabled E-Safety Network was available during the
events sited in this chapter, many of the communications problems experienced in
Toronto, Oklahoma City, and Graniteville, SC, may have been solved.
During the SARS outbreak, public health officials could have easily routed mes-
sages to the right agency destinations, without needing to know their electronic
addresses. Medical and emergency professionals throughout the region, if they
were registered in EPAD, would have had the information they needed to recognize
SARS and understand treatment protocols so they could have provided faster and
better response (COMCARE, December, 2004).
In Oklahoma City, requests for blueprints, personnel listings, and bomb effects
could have been issued as soon as dispatch was notified of the bombing. Responders
would have had the information they needed when they arrived at the scene, the 9-
1-1 center could have communicated with EMS, and the hospitals would have been
kept informed about injuries, casualties, and requests for resources (COMCARE,
November, 2004).
The South Carolina Emergency Management Division, after running a plume model
based on wind direction for the chlorine gas spill, could have easily targeted and
routed messages to the appropriate emergency personnel, radio and television sta-
tions, and other disparate public alerting mechanisms without making telephone
calls. The public would have had the information they needed in time to take ap-
propriate action (COMCARE, March, 2005).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

When it comes to emergency response, location does matter. By using the data-sharing
framework of the E-Safety Network and registering in the EPAD, agencies can be
assured that the right information will reach the right people at the right time. The
technical complexities of the EPAD were discussed in this chapter. Agencies that
register in the directory probably do not realize what it takes to make this happen.
Luckily, they do not need to know.

References

Center for Disease Control (CDC). (2005, January). Public health consequences from
hazardous substances acutely released during rail transit — South Carolina,
2005; Selected states, 1999-2004. Morbidity and Mortality Weekly Report,
54(03), 64-67.
COMCARE. (2004, October). The E-Safety Network and EPAD architecture:
Data system interoperability for homeland security and emergency response.
Washington, DC: COMCARE.
COMCARE. (2004, November). EPAD in action — emergency management (Okla-
homa City Bombing). EPAD Case Study 2.0. Washington, DC: COMCARE.
COMCARE. (2004, December). EPAD in action — public health (SARS). EPAD
Case Study 3.0. Washington, DC: COMCARE.
COMCARE. (2005, March). EPAD in action — public warning (SC Train Derail-
ment). EPAD Case Study 4.0. Washington, DC: COMCARE.
COMCARE. (2005, June). EmergencyProvider Access Directory (EPAD): Define
and design phase technical design document, Version 1.1. Washington, DC:
COMCARE.
Cooper, J. M., Lewis, J. C., Hajek, F. P. & Shapiro, R. N., (2005, Winter). South
Carolina chlorine gas disaster shows risks to public. Back on Track, Virginia
Beach, VA: Hajek, Shapiro, Cooper, & Lewis.
Federal Communications Commission (FCC). (2004, August). Review of the Emer-
gency Alert System, notice of proposed rulemaking (EB Docket No. 04-296,
FCC 04-189, para. 12).
Lipowicz, A. (2005, January). South Carolina chlorine accident showed officials
a better Emergency Alert System is needed. CQ homeland security local re-
sponse. Retrieved February 11, 2005, from http://www.cq.com
Liadis, J. S. (2000). GPS TIGER accuracy analysis tools (GTAAT) evaluation and
test results. TIGER operation branch, geography division in the U.S. Census
Bureau. Retrieved from http://www.census.gov/geo/www/tiger/gtaat2000.
pdf
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

Manzi, C., Powers, M. J., & Zetterland, K. (2002). Critical information flows in the
Alfred P. Murrah building bombing: A case study (Special Report 3). Wash-
ington, DC: Chemical and Biological Arms Control Institute.
National Information Exchange Model (NIEM). (2005). National information ex-
change model. Retrieved from http://niem.gov
O’Connor, J. (2005, January). Graniteville toll rises to 9. The State, Columbia, SC,
A1.
Office of Justice Programs (OJP). (2005). Information technology initiatives:
Global Justice XML Data Model. Retrieved from http://www.it.ojp.gov/topic.
jsp?topic_id.43
Open Geospatial Consortium (OGC), Inc. (2004, August). Web map service, Version
1.3 (Project Document OGC 04-024). Wayland, MA: OGC
Open Geospatial Consortium (OGC), Inc. (1999, May). OpenGIS simple feature
specification for SQL, Revision 1.1 (Project Document 99-049). Wayland,
MA: OGC.
Peterson, S. (2005, May). Proving its mettle. Government Technology, 18(5), 18-
24, 64.
SARS Commission. (2004, April). SARS and public health in Ontario. The SARS
Commission Interim Report. Toronto, Canada: SARS Commission.
XML Cover Pages. (2005, December). XML and emergency management. Retrieved
from http://XML.coverpages.org/emergencyManagement.html

Endnotes
1
COMCARE’s definition of “emergency agency” is broader than the traditional
first responder community. For the E-Safety Network and EPAD, the term
includes any government, private or non-governmental organization that par-
ticipates in emergency preparedness and/or response activities. This includes
fire, law enforcement and emergency medical services, as well as utilities,
other private infrastructure companies, schools, and tow truck drivers. The
total universe described by this term can exceed 250,000.
2
OnStar uses automatic crash notification (ACN) technology, bringing together
emergency service providers, wireless telephone, and satellite technologies,
to help protect owners and their automobiles.
3
Qualcomm supplies a product called OmniTRACS to trucking and construction
equipment fleets. OmniTRACS includes tamper-alert systems, panic alarms,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geospatally Enabled Drectory for Emergency Response Interoperablty 

and satellite-tracking capabilities to help minimize vehicle tampering and theft

and to help facilitate quick emergency response when an accident occurs.
4
A prototype of the EPAD (www.epad.us) was developed by DICE Corpora-
tion in 2003. This prototype enabled COMCARE to demonstrate the data
interoperability concept. However, the design of the EPAD described in this
document was funded by a grant from the Department of Justice. As COM-
CARE contractors, Proxicom, Inc., of Reston, VA, and Advanced Technology
Solutions, Inc., of Lancaster, PA, developed the technical design.
5
The common alerting protocol (CAP) standard is an open, non-proprietary
Organization for the Advancement of Structured Information Standards (OA-
SIS) standard for the exchange of emergency alerts and public warnings over
data networks and computer-controlled warning systems.
6
The Vehicular Emergency Data Set (VEDS) is an XML standard for the
transmission of telematics data to emergency agencies. Initially designed
to transmit ACN crash data to an emergency agency, VEDS also serves as a
data receptacle, collecting important bits of information as the response effort
unfolds. The data set can contain data transmitted directly from the vehicle
like vehicle speed, airbag deployment, direction of force and rollover, as well
as information from the telematics provider about the vehicle and its owner.
Questions asked by a 9-1-1 operator about the age and gender of the occupants
and data from responders and witnesses at the scene can also be added.
7
The Emergency Data Exchange Language (EDXL) represents a suite of
emergency messaging standards. The first EDXL standard developed is the
“Distribution Element” or DE. While different from the other EDXL messages
sets, this standard acts as a container that facilitates the routing of message sets
(any properly formatted XML emergency message) to recipients. The DE car-
ries and routes “payload” message sets by specifying key routing information
such as event type and affected geography as well as basic information about
an emergency incident. In June, 2006, the OASIS international standards con-
sortium announced the approval of the Emergency Data Exchange Language
Distribution Element (EDXL-DE) version 1.0 as an OASIS Standard.
8
Event types represent a variety of emergency incidents. Values for this data
element are specified as part of the EDXL DE standard. Currently, there are
over 100 event types specified for use.
9
PostGIS has been developed by Refractions Research as a research project in
open source spatial database technology. PostGIS is released under the GNU
General Public License.
10
A messaging system or message broker accepts signals from an application
system that wishes to send an emergency message and routes the message to the

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Woodhall

appropriate parties based on information received from the EPAD. It functions

as a signal clearinghouse and notification service, using the incoming signal
as the driver and the EPAD database as the source data from which routing is
performed. It can be an independent service used by many application systems
or the functionality can be embedded into individual applications systems.
11
Absolute horizontal accuracy represents the difference between the represented
coordinates and their true position. It is expressed as a circular error at 90
percent probability.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

Chapter.IV

The.Weather.Tool:.
An.Agent-Based.Approach.to.
Information.Integration
Gunjan Kalra, Quantum Leap Innovatons, USA

Abstract

This chapter discusses the process of providing information in its most accurate,
complete form to its users and the difficulties faced by the users of the current
information systems. The chapter describes the impact of prevalent technologies
such as the multi-agent systems and the Semantic Web in the area of information
supply via an example implementation and a model use case. The chapter offers
a potentially more efficient and robust approach to information integration and
supply process. The chapter intends to highlight the complexities inherent in the
process of information supply and the role of emerging information technologies
in solving these challenges.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kalra

Introduction

The progress in distributed computing has played a significant role in facilitating

replacement of large, monolithic databases with smaller and conceptually self-
contained databases. This has resulted in efficient yet “ever-expanding” clusters
of heterogeneous sources of data (data sources) such as the traditional databases,
Web sites, e-mails, and operational applications. The need to view and analyze data
from various data sources unified under one application context has become more
prevalent and has gained prominence among data analysts, both in the information
technology industry and academia. For example, large-scale enterprises have a grow-
ing demand for enterprise integration products, while data mining researchers need
a unified view of data from different sources to select and test their algorithms. The
task of providing integrated data management within one umbrella is particularly
challenging. In this chapter, we describe the need and evolution of the Weather Tool
system, a weather data provisioning and integration system accessible via a standard
application programming interface (API). The chapter begins with an introduction
to the evolution of sources of data and to the problem of weather data integration.
Some of the previous and existing data integration technologies and products are
then summarized. The chapter will also address some of the new technologies that
the data/information integration systems benefit greatly from, including the agent
technology and the Semantic Web. We then provide an overview of the Weather
Tool architecture (and an example application) that enables integration of several
data sources to provide a unified data environment. The architecture discussion
describes how data processing services are provisioned in a dynamically-changing
environment. The advantages and pitfalls of such a system are elaborated, followed
by a discussion of potential improvements in the architecture. The chapter briefly
discusses the core ideas of Quantum Leap Innovation, Inc..(QLI) intelligent data
management (IDM) framework, the generalized, enhanced version of the Weather
Tool effort. This framework is aimed at performing beyond the Weather Tool’s single
repository approach to data integration by incorporating several strategies for on-
demand, data access and analysis remotely, thereby avoiding creation of redundant
intermediary repositories. Finally, the chapter highlights the need, evolution, and
importance of data/information integration efforts.

Evolution. from. Large. Databases. to................

Large. Data. Environment.

Historically, due to high set-up and maintenance costs for databases, extremely
large volumes of data were stored in one database; this strategy was considered ef-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

ficient and had the advantage of easy access to all data. In reaction to the scale and
complexity of such large databases, data analysts have often created and exploited
selective views of subsets of the large databases. Since the advent of cheaper, flex-
ible, and more efficient data-storage solutions, the large databases have made way
for individual databases that store smaller sets of relevant data. These databases
have varying data access mechanisms and protocols that are usually dependent
upon non-standard, product specific technology. Some examples of commonly-used
relational databases are MySQL, Oracle, Postgres, and DB2.
Data-access mechanisms for the databases vary in complexity of implementation
and user-friendliness. The most common, structurally straightforward but tricky to
use data-access mechanism is Structured Query Language (SQL) that allows forma-
tion of data queries via a set of special commands and keywords. Relatively more
expressive and easy-to-use approaches involve the graphical user interface (GUI)
applications developed using technologies such as Visual Basic, Java, and so forth.
However, these involve long implementation times. More recent tools that have
been developed for easier implementation, use, and data access are the Web sites
that enable access to remote databases via Web pages. The Web sites/portals that are
easily customized to create data-access forms have become an important source of
information in today’s world. There are examples of Web sites functioning not only
as data-access mechanisms but also as sources of intermediate transient data that
can become crucial input to relevant applications. Examples include unprocessed
weather data, hourly stock market data, and event announcements. Transient data
inherently do not require elaborate storage but often becomes invaluable for certain
categories of consumers. For instance, unprocessed weather data can help in as-
sessing weather conditions and road conditions in the absence of validated weather
data. Event announcements, though not useful once the event is over, can be of great
use in estimating population densities and movements in cities and can also help in
highlighting the factors that affect the cities’ economies.
The popular adoption of Web sites as front-ends to relational databases has resulted
in the emergence of many heterogeneous data sources, a large data environment
with easy accessibility via the Internet. Other than the inherent efficiency and easy
maintainability aspects of having relatively small, self contained data sources,
the industry has recognized the advantages of dynamically composing new data
structures from the data in these data sources. There is high potential that analyzing
previously unknown combinations of data will result in useful data that can even be
life-saving in certain situations. These combinations are not known in advance and
the data sources have complex and often unique access mechanisms, as they were
not designed to be used in conjunction with each other. This makes it inefficient to
use traditional IT development mechanisms to integrate such sources statically, es-
pecially in view of the computational costs and the configuration overhead involved
in run-time addition of new data sources. In the following subsection, we describe

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kalra

one such data integration problem in the weather domain, where all components of
weather data are not available at one data source.

The.Weather.Problem

The data retrieval process assumes knowledge of at least the structure of the data
and at times of other constraints pertinent to the data source. The retrieval process
entails the basic data processing steps starting with data extraction, followed by data
validation, completion, and optional transformation and storage. In case of storage,
one must also worry about storage and consolidation of the corresponding meta-
data with the existing data. Increasing the number of data sources only escalates the
complexity in providing these data processing services. Adding to the complexity of
data retrieval, there are types of data, such as news reports, demographic data, and
weather data that change over time and impose extra processing to account for the
constant updates. Furthermore, the data may need to be extracted from more than
one relevant source, each with their own formats. Data from different sources vary
not just by data quality and content, but also by data format and update frequencies

Figure 1. National Climatic Data Center’s (NCDC) text-based validated historical

data

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

Figure 2. National Oceanic and Atmospheric Administration’s (NOAA) textual

forecast data

(how often data is made available). For example, news articles differ in content and
quality over a period of time, across different newspapers and Web sites. Moreover,
several news providers may write articles on different aspects of the same incident.
News is validated typically anywhere from a few hours to days.
Weather data is a good example of time-varying data. The National Climatic Data
Center’s (NCDC) Web site allows its users to view validated historical data in
textual form (Figure 1); the National Oceanic and Atmospheric Administration’s
(NOAA) Web site (Figure 2) shows forecast data for the next ten days in textual
form, and National Weather Service’s (NWS) web-site has non-validated forecast
data in graphical form (Figure 3). These sources are all showing different aspects
of weather, potentially for the same geographical location, and two of them have

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kalra

Figure 3. National Weather Service HTML-based forecast data

non-validated data that has the potential to be inconsistent with the third source,
while claiming higher quality in terms of completeness, accuracy, and so forth.
A weather expert is usually responsible for collecting current weather data from a
set of forecast data sources. The weather expert may also be required to retrieve
historical weather data from another source. The often unique formats of these data
require the expert to go through specific data extraction procedures for all the re-
quired locations. The locations that can be designated by zip codes can quickly rise
in number, with 99 being the number for the state of Delaware, one of the smaller
states in the U.S. Following the collection of data, the expert may be required to
validate and convert this data into standard format. Sources with transient data entail
provisioning of efficient storage in the form of caches or extra data repositories.
This also helps to avoid duplicating efforts for future access by applications that
require weather data.
Furthermore, as forecast weather data is validated (typically within two or three
days), the expert must update the repository accordingly. The expert is also respon-
sible for providing weather data to the relevant applications in the required formats.
If the applications require advanced meteorological data such as radiosonde (up-
per air) data, the expert must integrate weather data from the different sources per
location and date.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

Static and consistent data can be directly retrieved from the data sources. However,
in the case of weather, where tools and applications (consumers) require time-vari-
ant data, extra care must be taken while automating the retrieval process to account
for a series of data processing steps needed to make the data useful for analysis.
Consumers must maintain their own enormous databases in cases where the provid-
ers only store the data temporarily. Hence there is a need for a Weather Tool that
enables a logical value chain with individual system components for carrying out
the various services required for handling time-sensitive information. Although the
industry is still looking for better solutions for realizing dynamic, distributed, real-
time clusters of data sources, a lot of foundational work has already been explored
by the research community and identified by the industry.

Information. Integration

In this section we discuss some of the data integration technologies and solutions
that have been prevalent in the past. The essence of the prior approaches can also
be seen in the newer technologies. In addition to touching on the progress of data
and information integration in academia, we discuss the direction that industry has
been taking to handle the data integration problem.

Research.Efforts

The two main aspects of databases that have formed the basis for new information
integration technologies and products are the storage mechanism and the structure
and semantics of data. Relational databases and its derivatives, including federated
databases (Sheth & Larson, 1991) have been the most commonly-used test bed for
such technologies. Relatively recent and widely-adopted information integration
systems are the data warehouses that facilitate data acquisition, storage, and distri-
bution. However, these systems are used primarily for historical, non-operational
data via a central repository to aid the management’s decision-making process. Data
warehouses also deal with data consolidation issues even though the consolidation
structure and rules for preset collections of databases are determined initially at the
system set-up time and are difficult to change. In addition, adding a new source
not only requires bringing down the system temporarily, but also necessitates in-
corporating consolidation and verification rules for the new source. Widom (1995)
outlines the various research issues related to the design and implementation of
data warehouses. Pure data warehouse technology has further been combined with
materialized WHIPS (Labio et al., 1997) and virtual (Zhou, Hull, King, & Franchitti,
1995) views for defining and maintaining data warehouses. There have also been

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kalra

mediator-based architectures (Wiederhold, 1992) where the users and the applications
have been layered with custom software to enable integration via standard APIs.
This is an “on-demand” approach since a data source is integrated into the system
only when there is a query for information that it provides. In the database realm,
there are always two main disadvantages associated with channeling data into the
central repository structure: static, diverse, and domain-specific (non standard) data
structure, and a predetermined set of data sources.
Two assumptions made by the above database approaches for integrating heteroge-
neous data sources are that all the information sources to be integrated are known
a priori and that there is a uniform notion of semantics across the sources. There
is a separate community of researchers in source integration that focuses on the
uniform semantics aspect. Many wrapper-mediator style systems adopt either a
predefined global schema (Cohen, 1998; Genesereth, Keller, & Duschka, 1997) or
construct one based upon the source schemas (Bergamaschi, Castano, & Vincini,
1999), for describing the information and their sources. Wrappers interpret and
maintain local information about their information sources in terms of the global
schema. A mediator handles the user queries, sends them to the appropriate wrap-
per, and consolidates the individual results. Several methodologies applied in such
architectures include a common thesaurus in the MOMIS (Bergamaschi, Castano,
& Vincini, 1999) project, description logic in the Information Manifold (Levy,
1998), knowledge representation techniques in the InfoMaster (Genesereth, Keller,
& Duschka, 1997) efforts, AI planning techniques in the OCCAM (Kwok & Weld,
1996) project, and highly structured relational models in the WHIRL (Labio et al.,
1997) effort. For Web information sources in particular, learning by induction has
been applied to create wrappers, including the work by Kushmerick (Kushmerick,
Weld, & Doorenbos, 1997) and Muslea (Stalker) (Muslea, Minton, & Knoblock,
1999). These wrapper induction systems apply machine learning algorithms to learn
Web resource extractors by using manually-labeled training data sets. In our system,
we focus more on the overall optimized efficiency of the system rather than on that
of the individual agent.
As we progress towards distributed, mobile, and unstable yet flexible systems, where
new providers of information come in and go out of existence unpredictably, we
need to take dynamism and adaptability features into account as these information
providers make significant contributions to the data in demand. To address the sec-
ond assumption of fixed sources, we draw upon the concept of intelligent software
agents (Etzioni & Weld, 1995; Wooldridge & Jennings, 1995). Researchers have
implemented several architectures for data and information integration by using the
agent paradigm. Decker, Sycara, and Williamson (1997) “agentify” the wrapper-
style architecture as “middle agents” that provide the service of mediating between
information producers and consumers. Decker et al. describe nine categories of
middle agents based upon the knowledge of consumers’ preferences and providers’
capabilities. Of particular interest with regard to unstable data environments is the

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

category of “broker-agents,” where the consumers and the providers do not know
each other’s capabilities and preferences, and the brokers are responsible for solving
the connection problem between them. However, the condition where a consumer’s
preferences are not met completely by one producer and the broker must arrange
for a combination of producer services is not adequately addressed. For example,
if the only available information producer does not provide the data in the desired
format, the consumer may end up looking for an appropriate translation service as a
separate transaction. The consumer has to specify the relevant preference to be able
to find the translation service required for the desired format. Furthermore, in case
the producers have partial information, the consumer has to either know or learn
information-specific rules for merging the components or look for an information
consolidation service.
RETSINA (Sycara, Decker, Pannu, Williamson, & Zeng, 1996), MIKS (Bergamas-
chi et al., 2001), KRAFT (Preece, Gray, & Hui, 1999) and InfoSleuth (Bayardo
et al., 1997) are representative examples of agent-based information integration
efforts. There are architectural similarities between InfoSleuth and the Weather
Tool, specifically in terms of the type and roles of the agents. They do not men-
tion incorporation of actively changing Web sources (in content and layout). In
addition, they do not provide for caching frequently-queried information; this may
result in unnecessary communication and overloading of their resource agents. In
the Weather Tool, the weather data repository reduces most response times to a
single lookup in the repository. IDM goes further to incorporate data caching at
the various system components. InfoSleuth describes a few fixed, yet complicated,
interaction protocols for communication of data across the various agents. Our
system is able to dynamically adopt more efficient protocols, according to where
the data is actually stored. Finally, InfoSleuth provides a Web interface for human
users, whereas our system provides an agent-based API that can be easily used by
other software applications.
Although the agent paradigm is not very evident in the commercial data integration
technologies in the industry, there is presence and proven successful use of enterprise
integration products at two levels of enterprise assets, namely the data and the ap-
plication. The following subsection provides an overview of such technologies.

Technologies.in.the.Industry.—.ETL,.EAI.and.EII

The business community has adopted products geared towards solving the various
integration issues within an enterprise. Efforts have focused on integration at the
data and application level. Enterprises typically deal with data, software tools that
analyze this data, and the information that is generated by these tools. Integration
at the data and tools level is an important element in providing a global view of an
enterprise’s asset (customers, products, sales, etc), market status (stock prices, sales),

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kalra

and growth profile (predictive analysis). Data integration facilitates easy, uniform
access to operational data existing in applications and databases and to archival
databases for efficient statistical and predictive analysis. Data integration is also a
step towards achieving a standard notion of semantics for the enterprise in terms
of enterprise metadata. Application integration enables data exchange and change
propagation among a set of applications, thus providing the ability to influence an
application’s analysis with external information.
Besides being a solution to data integration, well-adopted complex data architectures,
including data warehouses, operational data stores, data marts, and multidimensional
analytical databases, provide efficient information services to the enterprise applica-
tions. Such architectures provide metadata standards for fixed applications, which
have traditionally been scoped to individual departments. Integrating new enterprise
systems into these architectures and enabling data exchange across departments has
proven to be expensive, both cost- and effort-wise.
The extract, transform, and load technology (ETL) has dealt with the challenges
involved in accessing data across applications. Although the ETL tools are popular,
they function in batch mode, extracting, transforming, and loading batches of data,
typically once in 24 hours to one month (depending upon the type of data). In the
present highly competitive markets, an enterprise’s survival and success is highly
dependent upon real-time information.
Newer generations of enterprise applications have been geared towards enabling
real-time distributed computing. As these applications have become more efficient,
they behave more specifically in terms of their metadata and repositories, data
translation tools, and interaction protocols, thus locking critical information in
proprietary systems. This has led to severe impediments to the flexibility, integrity,
accessibility, and maintainability of overall enterprise information and metadata.
Enterprise application integration (EAI) technology has focused on methods and
tools for real-time integration and coordination of data exchange and information
communication between legacy applications and databases, and systems based on
new technologies. EAI solutions consist of a messaging system, a middleware for
message routing and data transformation, and a collection of application specific
adapters. Separation of brokering functionality into a middleware has proven to
be flexible, but EAI solutions still involve static configurations. In other words,
integrating new systems have still required bringing down the system temporarily
for modifications to the middle layer’s interface and to the standard data model
being used.
EAI is crucial to enterprises for maintaining integrity in enterprise data, keeping
widespread heterogeneous systems in synch, but EAI solutions are specific to a
given set of applications. It would be a great advantage if new system components
(applications or databases) and/or new data types could be dynamically linked to
the set of existing components without having to deal with any static configurations
or even having to reboot the system. Intelligent data management (IDM) intends to
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

provide such a dynamic and extensible framework for the creation, extension, and
efficient utilization of an enterprise’s data stores and data.
The characteristic functionalities and goals of the enterprise information integra-
tion (EII) technology are very similar to those of intelligent data management.
EII stemmed out of the persistent need to seamlessly access data and information
residing in systems across the enterprise as one unit instead of having to deal with
the intricacies of the individual systems. EII converges on real-time integration of
(components of) data from several data sources primarily by labeling data with a
unified, consistent, and semantically-rich metadata. EII is well suited for data sources
that are not stable and whose content gets updated often. Some of the common tasks
within EII and IDM are data extraction, cleansing, standardizing, validating, and
optionally labeling and formatting. Some of the current technologies in the research
community have high potential of enabling flexible, robust, and efficient EII solu-
tions. The following section delves into the details of two such technologies.

The. Prevailing. Technologies

Realizing the data environment, as discussed in the previous sections, will rely
on sophisticated frameworks supporting distributed computing with certain “in-
telligence” features. The word “intelligence” here signifies special features of a
technology that distinguishes it from other technologies in terms of performance
and operability. Examples of such features include the ability of systems compo-
nents to interact with their environment, and the ability of system components to
understand (process) system data. As the research community continues to build
innovative intelligence technologies, the task of putting together these technologies
for realizing vital market solutions becomes even more interesting and viable. In the
following sub-sections, we discuss some of the technologies that have the potential
and inherent robustness for supporting near real-time information integration efforts
and how Quantum Leap Innovations, Inc. (QLI) is using them to provide one such
solution. In particular, we will discuss the Semantic Web and agent technology that
have already impacted a host of industries such as the Web economies, the mobile
industry, and the IT industry.

The.Semantic.Web

The Semantic Web technology constitutes a new way of representing Web content
that makes it processable by software tools with the capability of processing and
otherwise supporting the semantic Web content. The Semantic Web is an effort of
the World Wide Web Consortium (W3C) to enhance and standardize the content

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kalra

being displayed on Web sites by marking it with meaningful tags, making it easy
for the reasoner software to manipulate it. The Semantic Web boasts of ubiquity as
it is accessible via standard network protocols by any application or software. The
Semantic Web languages are based on the XML technology that originally gave
arbitrary, user-defined, structure to the Web content. The representation scheme is
based on a special syntax, resource description framework (RDF), that uses unique
resource identifiers (URI) (strings starting with “http:” or “ftp :”) to represent data.
The RDF is a syntax language that allows data to be represented in the form of
triples, binding three data items in a “subject — object — predicate” relationship. A
collection of RDF triples defines resources and relations among them, thus mapping
the data to a model that is inherently decentralized and universally accessible. This
syntax is used to create languages/schemas/metadata for representing data. These
schemas in the Semantic Web terminology are referred to as ontologies. Since these
ontologies are easily accessibly via the Internet, applications can efficiently share
and access this metadata to understand the data being represented by them. The
basic schema created using the RDF syntax is the RDF Schema that gives types/
categories to data and thus imparts a hierarchy structure to data. RDFS allows tags
to define classes and properties and add properties to classes of data.
Higher levels of Semantic Web languages impart more logic and reasonability
features to data by adding further vocabulary to the existing semantics. One of the
most important of these languages in this context is the Web Ontology Language
(OWL) that has been formally described by the W3C in terms of a set of documents
(OWL). The OWL vocabulary allows the users to impart more structure to their data
by adding more features to the properties and classes. These features range from
special relations among classes such as union and disjoint-ness, to creating hierarchy
of properties and assigning of attributes to properties, quantification, and equality of
classes. The Semantic Web languages tend to get very verbose and should be dealt
with preferably via special editors such as SWOOP and Protégé for generating and
editing ontologies. There are other tools for supporting these languages such as Jena
and Pellet ontology reasoners and Schemagen and Kazuki that provide Java support
in terms of creating Java classes for OWL, DAML, or RDFS vocabularies.
In the Weather Tool, we demonstrate how data schema can be used to create a very
flexible yet stable network of data sources. The section on IDM further describes
the use of ontologies in the IDM framework.

Multi-Agent.Systems.(MAS)

The agent paradigm involves realizing the functional components in a system as

software agents (Etzioni & Weld, 1995; Wooldridge & Jennings, 1995) and having
them interoperate in an agent-based environment. A software agent is a software
program that can control its execution in terms of carrying out specific tasks, per-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

ceiving, and interacting with its execution environment and other software agents.
An agent is also a piece of software with rules for achieving its functionality and
for finding and responding to the input that it receives from other agents and its
environment with minimal (run time) participation of the human user. Agents exhibit
goal-directed behavior and can cooperate in teams, usually via asynchronous mes-
sage passing. Agent communication is typically realized via messages. Agents can
be made “more intelligent” by augmenting them with the capabilities of learning
from their past executions and of adapting to perform better in similar situations.
A multi-agent system (MAS) consists of an execution environment, agents that are
executed in the environment, and resources that are shared and consumed by the
agents. A multi-agent system has decentralized and asynchronous information, com-
munication, and control architecture and provides industry standard mechanisms for
agent discovery, monitoring, communication, and cooperation between agents. The
agent systems provide directory services where agents register the services that they
provide to allow other agents to find them. They also provide means for appropriate
communication to invoke the registered services. Certain implementations of agent
systems also provide for agents to migrate between machines by virtue of being
complete units of execution that allow them to make their own decisions.
Agents perform tasks in the form of executing services. In an agent-based approach
to data integration, the data sources are represented as data provisioning services
with specifications for how to discover and invoke these services in industry standard
formats. Other required functionalities are also realized by service provisioning.
This service-oriented architecture enables a dynamic environment where service
providing agents can enter and leave the system without disrupting any functionality
of the operational system. When new data sources to be integrated into the system
are identified, representative service providing agents can be implemented and added
to the system seamlessly. The relevant data services for the new data source can be
easily built into representative agents. The agents can then advertise the new data
source and the relevant data processing services, enabling the data from the new
source to be integrated into the system.

Weather. Tool

Weather Tool is an agent-based system for provisioning weather data from a dy-
namic collection of data sources via one standard interface. Weather Tool makes use
of the above mentioned state of the art AI technologies to enable an agent-based,
service-oriented framework for bringing many data sources together in a virtually
integrated environment. The goals that the weather tool has achieved are the ability
to access data from a given data source and to integrate a new data source into the
system. Data access involves setting up a standard data querying mechanism. The

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kalra

Semantic Web technology is not fully utilized in the Weather Tool, but the system
clearly highlights its advantages. All the functionalities in the Weather Tool are
realized as services. The following subsection elaborates on these services. The
services are provided by software agents.

Weather.Tool.Data.Services

The Weather Tool system components are implemented as services in an agent

system. Details of the agent system that we use are mentioned in the next subsec-
tion. In the context of data retrieval from weather Web sites and databases, the data
must go through a series of processing steps before it can be efficiently utilized by
the concerned applications. The data (pre) processing techniques that we include
in the Weather Tool are described below along with other agent services that make
up the Weather Tool:

•. Data.provisioning.(retrieval).service: Providers of this service are respon-

sible for allowing access to a data source. The service provider connects to
the data source and captures the data as it exists at the source including the
source-specific formatting. For example, this service would post a query at
the Weather Channel’s (WC) Web site and return the resultant Web page with
HTML formatting. In the Weather Tool the data provisioning service provid-
ers are implemented as wrapper-style agents that have a thin logic component
dealing mainly with the process of retrieving data from the source as requested
by the user.
•. Standardizing.and.completion.services: The standardizing service involves
extracting the essential parts of data by removing the source formatting from
the originally-retrieved data and converting it to a standard storage-efficient
format for potential data requirements. For example, this service would parse
the weather data fields’ values into java objects that are easily updated and
maintained. Typically, standardization also includes completion service that
is responsible for filling in any missing values in the data.
•. Validation.service: Data validation involves applying domain specific rules
to ensure data correctness. As per Lab Compliance’s Glossary (LabCompli-
ance), data validation is “A process used to determine if data are inaccurate,
incomplete, or unreasonable. The process may include format checks, com-
pleteness, checks, check key tests, reasonableness checks, and limit checks.”
Some examples of this service are checking if -1000 degrees Fahrenheit is
a valid value for temperature, updating today’s weather (available from the
forecast service) with historical data when it is available and replacing any
missing value in the temperature field by the standard value (-273 degrees
Fahrenheit).
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 

•. Updating.service: The data updating service comes in handy where data is

pushed from data source to data consumers. In this case, the consumers can
subscribe to a set amount of data and then the data source (or the data provi-
sioning service) can initiate a data updating service to ensure that data is sent
to the consumers periodically.
•. Consolidation.service: The data consolidation service is responsible for ap-
propriately merging data from many sources into one repository. This service
is highly dependant on domain specific rules to carry out its functionality. It
is essential for the designer/implementer of this service to encode the domain
rules for correct fusion of data. As per British Columbia Government’s Informa-
tion Resource Management’s glossary (BCG), data integration is the process
of “blending data items from various distinct sources to create a larger and
more comprehensive body of knowledge.” For example, historical and forecast
weather data as retrieved from their sources is stored in one repository.
•. Storage.service: As evident from the name, the data storage service is respon-
sible for providing efficient storage mechanisms for data.
•. Formatting.and.translation.services: This service includes converting data
into application-specific formats, for example, geographical information sys-
tems (GIS) applications may require data in the form of GIS layers, whereas
traffic monitoring tools may require the same data in the text form.

These services can be combined in numerous ways resulting in different architectures

that can be utilized as per the system’s requirements.

System.Architecture

Our agent-based data integration system consists of software agents providing com-
binations of data processing services in a heterogeneous, distributed environment,
as best suited for the problem of weather data provisioning (Figure 4). In Figure
4, we depict agents as special three-part images that are explained in our previous
paper (Kalra & Steiner, 2005). The Weather Tool’s agent system is implemented and
deployed using our in-house agent development and deployment framework, multi-
agent development environment (MADE). MADE provides essential enhancements
to JADE (Java agent development environment) (Bellifemine, Poggi, & Rimassi,
1999) with simplified interfaces for agent development and capabilities such as
subscription services. JADE is a Java-based middle-layer software framework for
agent development and deployment supporting the FIPA standards (FIPA). JADE
provides the Agent Communication Language (ACL) message performatives and
protocols for agent communication. At this point, we use a pre-determined metadata
for describing the services in the absence of Semantic Web technology. Agents register

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
00 Kalra

Figure 4. System architecture for the weather tool

Web
Pages
Traffic
Retrieval Agent
Monitoring
Email Application Agent
Email

Standardized data formats

Retrieval Agent Dispersion
Modeling
Application Agent
Structured
Structured Application-specific
Text
Text data formats

Retrieval Agent Domain-Specific

Consolidation & GIS
Validation Awareness
Rules Application Agent

Web
Service
Retrieval Agent Data
Repository

Database Repository Agent Translation Agent

Retrieval Agent

and advertise their services, namely the “retrieval,” “repository,” and “translation”
services (as described as follows) using the JADE directory facilitator (DF).
We use service descriptions to describe the type and content of the data provided
by a source for the agents to publish their services. These services descriptions
can then be discovered by other agents in the system. The system has four types
of agents:. retrieval agents, repository agents, translation agents and application
agents. Once registered, the retrieval agents either start retrieving data from their
sources or wait for a request for data. The repository agent, after registering its
repository service with DF, looks up the retrieval services and subscribes to them.
The repository agent periodically checks for and subscribes to new retrieval agents
that join the system. The application agents look up the weather repository service
and subscribe to them for their data requirements. The repository agent looks up
and invokes translation services when data is requested in formats other than its
standard, data storage format.

•. Retrieval.agents: The Weather Tool retrieval agents provide the data extrac-
tion, parsing, and standardizing services. These lightweight, wrapper-style
agents extract data as it exists at the data sources, use custom weather data
format (“QLI Format”) to parse the data into, and pass the weather data to their
subscribers. The retrieval agents are implemented specific to the data source
they are responsible for retrieving data from and are hence lightweight in terms

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 0

of process logic and concentrate only on the process of retrieval. These agents
do not maintain any repositories of the data they collect.
•. Repository.agents: The repository agents provide the data completion, con-
solidation, storage, and updating services. Repository agents either subscribe
to or send individual requests to the available retrieval agents as per the user
input. In our current implementation, the repository agent maintains a reposi-
tory of weather data in QLI format using the MySQL database. The repository
agent also uses the discovery and the integration of agent services to segregate
the original data request into individual service-specific requests.
•. Translation.agents: The translation agents provide format conversion services
between the QLI Format and other source-specific formats. The translation
agents apply information-specific conversion rules and are very specialized
in particular format conversions. The translation agents publish the formats of
their input and output data. For example, GIS Data Converters translate data
from weather repository specific “QLI format” to GIS application specific
format. For example, agents that wrap any GIS applications extract a set of
weather data attributes from the repository storage format.
•. Application.agents: Application agents are wrapper-style representatives of
various applications in the Weather Tool and act as information consumers.
Application agents look up and either request for or subscribe to repository
agents that provide services for the type of data that they require.

The.Weather.Tool.Test.Case

Beyond the need to directly inform human users, weather data is often an input for
many software applications including emergency response tools that support scenario
planning and building foresight for emergency planners. The applications require
different components of weather data as input, for example, viewing daily-summa-
rized weather data on a map of a geographical area helps provide a situational view
of the world. In our test case, maps/GIS are one of the weather data consumers.
The other application in the test case is a traffic monitoring and forecasting service.
Besides analyzing, predicting, and planning for the current traffic purely based upon
the time of the day and the road conditions, a traffic monitoring system should take
into account current and forecast weather conditions. Warnings such as snow, ice,
and high winds can help traffic experts to anticipate traffic problems, slowdowns,
and accidents, due to poor road conditions. For such applications, the weather re-
quirements can range from historical to forecast weather data.
Weather data comprises of readings on weather elements such as temperature, pres-
sure, wind speed, wind direction, dew point, precipitation, humidity, and so forth.
The weather experts collect data by using sensors and other advanced equipment, for

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kalra

example, the surface weather data (weather at sea-level) for a given time and loca-
tion is measured and can be accessed at the National Weather Service (NWS) Web
site. Weather data is validated within two to three days and can be accessed at the
National Climatic Data Center (NCDC) Web site (NCDC) for a few pre-determined
weather stations only. National Oceanic and Atmospheric Administration (NOAA)
has non-validated current forecast data. The Forecasts Systems Laboratory’s (FSL)
Web site allows requests for weather data at several altitudes in the atmosphere,
the radiosonde data (A radiosonde is an instrument for sending sensors up in the
layers of the atmosphere for measuring weather data). In the test case, there is one
retrieval agent each for the NCDC and the NOAA Web site. The NOAA-agent
retrieves data every hour as the NOAA Web site has transient data. The NCDC and
FSL agents retrieve data on demand as the corresponding sites maintain their own
repositories and provide URL-based query support. In the test case, we have the
following agents:

• The retrieval agents:

 NOAA.forecast.agent: Retrieves forecast weather data from the NOAA
Web site
 NOAA.historical.agent: Extracts validated historical weather data from
the NCDC Web site
• The application agents:
 Map. agent: Represents a map component that displays the map of
Delaware with various layers of geographical data including location of
schools, fire departments, police stations, and hospitals
 Traffic agent: Represents a traffic monitoring and analysis application
• The repository agents:
 Weather.repository.agent: Stores data and provides complete weather
data for a given date and location to application agents
• The translation agents:
 Map.translation.agent: Provides QLI Format to GIS layer format con-
version
 Traffic translation agent: Provides QLI Format to traffic specific format
conversions.
 NOAA.forecast.agent: Retrieves and parses data from the NOAA fore-
cast service Web site into standard JAVA objects. The NOAA Web site
is shown in Figure 2. Detailed forecast data is available for the next 60
hours and summarized data exists for the next ten days. NOAA Historical
Agent retrieves and parses data from the NCDC Web site on demand.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 0

Historical data is available for every hour in the format shown in Figure
1. Historical data for a day only becomes available after two to three days.
Historical data is validated by the experts and is more detailed than the
forecast data in terms of available weather attributes.

The map agent represents a GIS application that allows users to view data from the
geographical perspective, based upon data such as latitude and longitude. Weather
data for such applications is typically measured for an area or a point. GIS applica-
tions can provide the geographical perspective for weather data by showing their
coverage area on the map of a region. The traffic agent represents a traffic monitoring
and analysis application that utilizes the wind speed and other daily warnings fields
such as heavy rain, snow, sleet, icy road conditions, and so forth, of the weather
data. As weather data exists in the repository for a region and a time, the traffic
agent requests and subscribes to current weather data for the geographical regions
to which the various roadways belong.
The map translation agent extracts certain features from the standard weather data,
thereby reducing the set of attributes to those required by the map agent. The traffic
translation agent extracts wind speed and other warnings from the standard format
weather data. The weather repository agent stores forecast and historical data into
one weather repository as and when it gets updates and new data from the scrapers.
The weather repository agent also performs validation, completion, and consolida-
tion of data. It successfully updates the previously-existing forecast data when the
corresponding historical validated data becomes available.
The NOAA forecast agent and the NOAA historical agent register with the JADE
directory facilitator service as “weather retrieval” service providers. These agents
retrieve data and send update messages with the current results either to their
subscribers or to the requesting agent. The weather repository agent registers its
“weather repository” service and performs a DF look-up of weather retrieval service
providers. The weather repository agent subscribes to all the resultant agents using
the subscribe messages. The map agent and the traffic agent, as per their require-
ments, perform look-up for weather repository services and either subscribe to or
send queries to the resultant agent(s) using the ACL subscribe or query messages.
The map translation agent and the traffic translation agent register their “map con-
version” and “traffic conversion” services with the DF, respectively. The weather
repository agent checks its repository for requested weather data and accordingly
sends requests to appropriate retrieval agents. The weather data requests are chan-
neled to appropriate translator agents in case of a specific format request. The results
received from the translation agents are sent to the requesting application agents in
appropriate ACL inform or query reply messages.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kalra

Example.Enhancement

The Weather Tool test case worked well, with agents registering, finding, and invok-
ing the various data services. At this point, we identified a need to introduce a new
application to the Weather Tool. This application is an air quality modeling system
that models the transport of pollutants in the air following an accidental or intentional
release of a biological or chemical substance. The specific application is CALPUFF
(CALPUFF) that is the default choice of the U.S. Environmental Protection Agency
(EPA) for modeling long-range transport of the pollutants. CALPUFF computes an
approximate plume of the pollutant over a period of time based upon the weather
data for that time period. The assessment of the pollutants is done based upon the
weather data in a geographical area. The details of CALPUFF and the weather data
used by CALPUFF can be found in our previous paper (Kalra & Steiner, 2005).
To integrate the air quality modeling system, we needed a translation agent for
converting standard data to the formats that CALPUFF required, the “CALPUFF
translation” agent. We also built the logic for merging the radiosonde observa-
tions with the existing surface weather data into a service providing agent — the
“radiosonde consolidation” agent. As and when the radiosonde data for a time and
location becomes available, it is added to the existing surface data for the same time
and location. We also implemented a new retrieval agent; the “radiosonde agent”
that periodically retrieved and parsed data from the FSL Web site and passes on the
results to its subscribers. The weather repository agent employed the CALPUFF
translation agent to combine the radiosonde data for a particular time and region
with the existing weather record for that time and region.
Once the required agents were implemented, there was no need to bring down the
system to incorporate this new weather data source. Rather, upon launching, the new
retrieval, translation, and the consolidation agents dynamically joined the system by
registering their retrieval, translation, and consolidation services respectively with
the DF. During its periodic check for new weather retrieval service providers, the
weather repository agent subscribed to the new retrieval agent and started receiv-
ing the new weather data. Finally, to get the newly available data from the weather
repository agent to CALPUFF programmatically, we wrapped the CALPUFF mod-
eling system with an application agent, the plume agent.

Conclusion

“The Goods.” The example enhancement mentioned in the previous section demon-
strates how flexible agent architectures can be used for solving a variety of problems
associated with collecting data from a number of disparate, independently-managed
information sources, where the quality, format, and sources of the data may change
over time. It allows for dynamically integrating new agents corresponding to new
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 0

sources and data formats without hampering the operational system. It also enables
the sharing of data in different formats and different subscription models over a
wide range of applications.
The Weather Tool’s agent-based approach proves to have many advantages over
the current approaches. The Weather Tool seamlessly handles new data sources as
opposed to the statically-configured data warehouses. The Weather Tool architecture
has middle layer repository agents acting as data provisioning, data consolidation,
and validation service providers. The other approaches handle mostly validated
data, as adding a validation logic layer to the system components can slow down
information flow considerably.
“The Bads.” Our current architecture strictly associates one data source with each
retrieval agent. This can be extended to have one retrieval agent scrape data from
many sources. This allows the retrieval agents to continue to scrape data from other
sources when it fails to scrape data from a particular source due to changes such as
format change. The data retrieval agents can be enhanced to include a thin verifica-
tion and consolidation layer to support the above functionality. Another important
direction for improvement is to have discovery-style retrieval agents that go out and
search for the relevant sources given concise information descriptions. This will also
take care of the source failure case, which is not addressed in the current system.
The Weather Tool stores all its standardized, integrated, and validated data in a single
central repository, thus giving rise to a single point of failure. The best solution for
this would be to implement a federated network of replicated and/or distributed
repositories as an extension to this architecture.
The Weather Tool in its current form assumes that the data sources (new and existing)
and the external applications understand and share a common metadata that is used
for representing the services and the service descriptions. These restrictions make
the Weather Tool rely upon certain constants that are not industry standards by any
measure. The obvious solution to this problem is to use Semantic Web technology.
Another problem is associated with the absence of a required service, where a simi-
lar service exists in the system but remains invisible to the system. The extensions
of the semantic technology, specifically the concept of metadata/schema/ontology
mapping is key to solving this problem. The ontology mapping tools derive map-
pings between two data schemas using concepts such as word processing, informa-
tion flow, and so forth. In the next section, we talk about the next generation of the
Weather Tool that takes advantage of the Semantic Web technology for enhanced
performance and flexibility.
From the overall architecture point of view, the vision is to have a completely-
autonomous adaptive network that adapts to the data traffic and determines when
and where data should be cached or stored while at the same time optimizing the
request-response time. Such a vision is also planned for in the next generation of
the Weather Tool.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kalra

Intelligent. Data. Management:........................

The. Second. Generation

The Weather Tool has performed very well within the context of the weather-re-
lated applications. The agent-based, service-oriented architecture of the Weather
Tool has brought known and unknown data sources within one data environment,
where their data is accessible at a central repository. The idea of intelligent data
management (IDM) has evolved to overcome the universal schema and the central
repository aspects. The IDM framework enables connectivity to a set of disparate
data sources (including applications) QLI’s IDM framework provides dynamic
integration and (re-) configuration of several diverse, geographically-apart data
sources, thereby offering a virtual data environment. No data is collected centrally
but the data sources are polled for data on demand. IDM draws upon the service
oriented architecture (SOA), Semantic Web (SW), and multi-agent system (MAS)
technologies for representing, deploying, combining, and invoking various data-
processing services (such as data gathering, validation, completion, translation, and
formatting). These services are provided by agents supported by MAS mechanisms
for agent discovery, communication, cooperation, and easy system configuration for
service deployment and usage. The agents perform semantic similarity matching
between the data requests and available data provisioning services and invoke them
to fulfill the requests. The IDM architecture also provides for a simple, machine-
processable, highly-structured query-based data access interface to its users. The
query interface allows for formulation, decomposition, and solving of queries based
upon representative keywords. The keyword queries range from keywords-based
search for unstructured plain text data to highly-structured schema templates across
a large number of heterogeneous data sources. Scalability in terms of functional-
ity and types of queries as well as the number of data sources has been one of the
guiding factors of the architecture of the query interface.

Conclusion

Effective operation within any enterprise relies heavily upon persistent, ad hoc, and
intelligent collection of data from geographically-distributed heterogeneous sources.
The advantage of information as derived from efficient and effective data retrieval is
crucial to effective enterprise management. As the number of relevant data sources
increases, the complexity of integrating them into the existing system renders the
current information technology systems incapable of providing information to the
analysts. The demand for innovative technology supporting dynamic data integra-
tion in a highly distributed environment is unprecedented, both in the industry and

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 0

academia. We discuss our efforts on the Weather Tool, a service-oriented agent-

based approach to data integration that facilitates data collection by means of data
provisioning and processing services. We further describe our ongoing work on the
intelligent data management framework that is an enhanced, domain-independent
version of the Weather Tool.

References

AccuWeather, Inc. (n.d.). World weather — local weather forecast. Retrieved from
http://www.accuweather.com
AFEI, The Association for Enterprise Integration (AFEI). (n.d.). Retrieved from
http://www.afei.org
Bayardo, R. J., Bohrer, W., Brice, R., Cichoki, A., Fowler, J., Helal, A., et al. (1997).
Infosleuth: Agent-based semantic integration of information in open and dy-
namic environments. ACM SIGMOD, 195-206.
Bellifemine, F., Poggi, A., & Rimassi, G. (1999). Jade: A fipa-compliant agent
framework. Practical Applications of Intelligent Agents and Multi-Agents,
97-108.
Bergamaschi, S., Cabri, G., Guerra, F., Leonardi, L., Vincini, M., & Zambonelli,
F. (2001). Supporting information integration with autonomous agents. In
Proceedings of the Fifth International Workshop CIA-2001 on Cooperative
Information Agents, Modena, Italy.
Bergamaschi, S., Castano, S., & Vincini, M. (1999). Semantic integration of
semistructured and structured data sources. SIGMOD Record Special Issue
on Semantic Interoperability in Global Information, 28(1).
CALPUFF, The Atmospheric Studies Group. (n.d.). ASG at Earth Tech: Official
CALPUFF. Retrieved from http://www.src.com/calpuff/calpuff1.htm
Cohen, W. W. (1998). The WHIRL approach to data integration. IEEE Intelligent
Systems, 13, 20-24.
Decker, K., Sycara, K., & Williamson, M. (1997). Middle-agents for the Internet.
In Proceedings of the 15th International Joint Conference on Artificial Intel-
ligence, Nagoya, Japan.
Etzioni, O., & Weld, D. S. (1995). Intelligent agents on the Internet: Fact, fiction,
and forecast. IEEE Expert, 10(4), 44-49.
FIPA. (n.d.). Welcome to the foundation for intelligent physical agents. Retrieved
from http://www.fipa.org/

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kalra

FSL, Forecasts Systems Laboratory. (n.d.). FSL radiosonde database information

page. Retrieved from http://raob.fsl.noaa.gov/Raob_Software.html
Genesereth, M. R., Keller, A. M., & Duschka, O. M. (1997). Infomaster: An infor-
mation integration system. SIGMOD Record (ACM Special Interest Group on
Management of Data), 26(2).
Government of British Columbia. (n.d.). Information resource management glossary.
Retrieved from http://www.cio.gov.bc.ca/other/daf/IRM_Glossary.htm
Kalra, G., & Steiner, D. (2005). Weather data warehouse: An agent-based data
warehousing system. In Proceedings of the Hawaii International Conference
on System Sciences-38.
Kushmerick, N., Weld D., & Doorenbos, R. (1997). Wrapper induction for informa-
tion extraction. In Proceedings of the 15th International Joint Conference on
Artificial Intelligence (IJCAI), Japan (pp. 729-737).
Kwok, C. T., & Weld, D. S. (1996). Planning to gather information (Tech. Rep. No.
UW-CSE-96-01-04). University of Washington.
LabCompliance. (n.d.). Glossary related to compliance for analytical laboratories:
c-d. LabCompliance Glossary. Retrieved from http://www.labcompliance.
com/glossary/c-d-glossary.htm
Labio, W., Zhuge, Y., Wiener, J. L., Gupta, H., Garcia-Molina, H., & Widom, J.
(1997). The WHIPS prototype for data warehouse creation and maintenance.
In Proceedings of the 1997 ACM SIGMOD International Conference on Man-
agement of Data, Tucson, AZ.
Levy, A. (1998). The information manifold approach to data integration. IEEE
Intelligent Systems, 1312-1316.
Muslea, I., Minton, S., & Knoblock, C. (1999). A hierarchical approach to wrapper
induction. In Proceedings of the 3rd International Conference on Autonomous
Agents, Seattle, WA (pp. 190-197).
NCDC. (n.d.). National Climatic Data Center. Retrieved from http://www.ncdc.
noaa.gov/oa/ncdc.html
NOAA. (n.d.). National Oceanographic and Atmospheric Administration. Retrieved
from http://www.noaa.gov
NWS. (n.d.). National Weather Service. Retrieved from http://www.nws.noaa.gov
OWL. (n.d.). Web Ontology Language Overview. Retrieved from http://www.
w3.org/TR/owl-features
Preece, A., Gray, P. M. D., & Hui, K. (1999). Supporting virtual organizations through
knowledge fusion. In Artificial Intelligence for Electronic Commerce: Papers
from the AAAI-99 Workshop. Menlo Park, CA: AAAI Press.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 The Weather Tool 0

QLI. (n.d.). Quantum Leap Innovations. Retrieved from http://www.quantumleap.

us
Sheth, A., & Larson, J. (1991). Federated database systems for managing distributed,
heterogeneous, and autonomous databases. ACM Computing Surveys, 22(3).
Sycara, K., Decker, K. S., Pannu, A., Williamson, M., & Zeng, D. (1996). Distributed
intelligent agents. IEEE Expert, 24-31.
WC. (n.d.). The Weather Channel. Retrieved from http://www.weather.com
Widom, J. (1995). Research problems in data warehousing. In Proceedings of
the International Conference on Information and Knowledge Management
(CIKM’95), Baltimore (pp. 25-30). ACM Press.
Wiederhold, G. (1992). Mediators in the architecture of future information systems.
IEEE Computer, 25(3), 38-49.
Wooldridge, M., & Jennings, N. R. (1995). Intelligent agents: Theory and practice.
The Knowledge Engineering Review, 10(2), 115-152.
Zhou, G., Hull, R., King, R., & Franchitti, J.-C. (1995). Data integration and ware-
housing using H2O. Data Engineering Bulletin, 18, 29-40.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ku & Zmmermann

Chapter.V

GIME:.
A.Geotechnical.Information.
Exchange.Architecture.
Using.Web.Services
We-Shnn Ku, Unversty of Southern Calforna, USA
Roger Zmmermann, Unversty of Southern Calforna, USA

Abstract

We present an information architecture using Web services for exchanging and

utilizing geotechnical information, which is of critical interest to a large number of
municipal, state, and federal agencies as well as private enterprises involved with
civil infrastructures. For example, in the case of soil liquefaction hazard assessment,
insurance companies rely on the availability of geotechnical data for evaluating
potential earthquake risks and consequent insurance premiums. The exchange of
geotechnical information is currently hampered by a lack of a common data format
and service infrastructure. We propose an infrastructure of Web services, which
handles geotechnical data via an XML format. Hereafter we report on the design
and some initial experiences.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.

TLFeBOOK
 GIME 

Introduction

Geotechnical information on soil deposits is critical for our civil infrastructure.

Local, state, and federal agencies, universities, and companies need this informa-
tion for a variety of civil engineering and urban policy applications, including land
usage and development, and mapping of natural hazards such as soil liquefaction
and earthquake ground motions. Foremost examples of geotechnical information,
geotechnical boreholes are vertical holes drilled in the ground for the purpose of
obtaining samples of soil and rock materials and determining the stratigraphy,
groundwater conditions, and/or engineering soil properties (Hunt, 1984). In spite of
rather costly drilling operations, boreholes remain the most popular and economi-
cal means to obtain subsurface information. These types of data range from basic
borehole logs containing a visual inspection report of soil cuttings to sophisticated
composite boreholes combining visual inspection and in-situ, laboratory geotechni-
cal and geophysical tests. Figure 1a shows an example transcript of the Standard
Penetration Test (SPT), a particular type of geotechnical borehole test. Significant
amounts of geotechnical borehole data are generated in the field from engineering
projects each year. The data and results of boreholes are usually published and
released as hardcopy reports, without the digital data from the field test. Naturally,
sharing data via the traditional hardcopy reports is slow and results in errors when
the data is converted back to digital form for further processing. With the recent
ubiquity of communication networks, particularly the Internet, the trend toward
electronic storage and exchange of geotechnical borehole data has accelerated. So
far these efforts have often been uncoordinated and ad hoc.
Geotechnical borehole data is complex and sophisticated in that it contains both
well-structured and semi-structured elements. In Figure 1a, for example, the Sum-
mary field contains free-form text, while some of the other columns are well defined.
Therefore, an efficient data format for storage and exchange is required that is suit-
able for the diversity of geotechnical borehole data. Currently, there is no accepted
common format for exchanging these data among researchers and practitioners. To
date, the most commonly-used formats for representing geotechnical borehole data
include the Association of Geo-technical and Geoenvironmental Specialists (AGS,
1999), the Log ASCII Standard (Heslop Karst, Prensky, & Schmitt, 2000) for well
logging in petroleum engineering, and the National Geotechnical Experimental
Site (Benoit & Lutenegger, 2000) format that adopts the AGS data dictionary and
expands it to cover more research-oriented geotechnical tests. Another research
project was initiated at the U.S. Army Engineer Waterways Experiment Station
(U.S. Army, 1998) to establish a standard electronic data format for geotechnical and
geological exploration in order to automate data interchange. In addition, the Pacific
Earthquake Engineering Research Center at Berkeley (PEER) and the Consortium
of Organizations Strong Motion Observation Systems (COSMOS) commence a

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
112 Ku & Zimmermann

Figure 1. The photographs illustrate the geotechnical boring activities from drilling
until the soil samples are examined; the result is a boring log showing stratigraphy,
physical sampling record, and SPT blow count: (a) example of boring log; (b) drill-
ing and sampling activities

(a) (b)

project to create a geotechnical data dictionary and exchange format specifically

for the digital dissemination of geotechnical data (COSMOS/PEER, 2002).
Recently, we have proposed the use of XML as the preferred format for storage and
exchange of borehole data (Bardet, Hu, & Swift, 2003; Zimmermann, Bardet, Ku,
Hu, & Swift, 2003). Similar efforts have been carried out by the W3G (The World
Wide Web of Geotechnical Engineers), which proposed the Geotechnical Markup
Language (Geotech-XML), a geotechnical engineering version of the eXtensible
Markup Language.1 The eXploration and Mining Markup Language (XMML) project2
of the SEEGrid Community (Solid Earth and Environment Grid) also proposed an
XML-based encoding for geoscience and exploration information. XMML is intended
to support exchange of exploration information in a wide variety of contexts. In
addition, the Open Geospatial Consortium3 (OGC) suggested the specification of
Geography Markup Language (GML). GML is an XML encoding for the transport
and storage of geographic information, which includes features, coordinate refer-
ence systems, geometry, topology, time, units of measure, and generalized values.
Geographic features in GML include coverage and observations as subtypes.
XML offers many advantages over other data formats for borehole data. Its tree-
structure and flexible syntax are ideally suited for describing constantly evolving
and annotated borehole data. However, in addition to defining new data formats

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

for borehole data, there is much research to be done to improve the exchange and
utilization of geotechnical information over the Internet. There is a need for an
infrastructure for sharing and manipulating this valuable information.
As the data collection and digital technologies improve, more and more geotechni-
cal borehole data from the field, laboratory, or office are directly produced in, or
converted to, a digital format. This makes it possible to use, share, and disseminate
geotechnical data nationwide by coordinating geotechnical data standard and In-
ternet dissemination activities within large organizations or government agencies.
With the recent ubiquity of communication networks, particularly the Internet, the
trend towards electronic storage and exchange of geotechnical borehole data has
accelerated. The ROSRINE (Resolution of Site Response Issues from the North-
ridge Earthquake) project has produced an integrated system based on a relational
database management system (RDBMS), geographic information system (GIS),
and Internet map server (IMS) to disseminate geotechnical data via the Internet
(Swift, Bardet, Hu, & Nigbor, 2002). The U.S. Geological Survey (USGS) is con-
tinually publishing seismic cone penetration test (CPT) data, which is collected
with USGS CPT trucks and then distributed through a Web-based system managed
by the USGS Earthquake Hazard Program in Northern California (Holtzer, 2001).
All these efforts make it easier, faster, and less expensive for all levels of public
and private sectors to access geotechnical information. Hence, these systems try to
address the challenges and needs for timely, high-quality data as required in many
geotechnical projects.
The architecture of most current geotechnical data archive and distribution systems
follows the schematic in Figure 2. The end user utilizes a Web browser to gather
the information of interest via the Internet. The data providers host large informa-
tion sets in their database systems. Using standard Web server technology, a query
interface at a URL address is provided to the end user for querying and retrieving
data via the Web browser. Since geotechnical data is spatial in nature, a GIS map
server is often integrated to provide a more graphical and user-friendly data query
ability. This type of interactive data exchange system is designed for human partici-

Figure 2. Traditional client-server architecture with a Web server and a browser- based
client; the client is used mostly for information presentation and visualization
Web.Browser
Applications
HTTP/HTML

Data Internet
Source
Local
Memory
Web.Server
e.g.,.ROSRINE

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

pants since the data flow occurs between the end user (most likely engineers) and
the system servers. However, this architecture lacks the flexibility to handle more
complex tasks. For instance, how can we streamline the upload of data gathered in
the field via digital equipment such as CPT trucks, which not only involve human
interaction but also machine-to-machine communication? Furthermore, many times
a geotechnical data set is not the end product, but the input to an analysis, simula-
tion, or other computation. With a Web server/browser architecture, no simple,
direct programmatic access to remote data is possible. Because many of the data
repositories are under different administrative control, it is important that the data
exchange format and the communication protocol follow a universally-acceptable
standard that is both simple and extensible.
We believe that the emerging standard of Web services (WS) is well suited to al-
leviate some of these problems.4 Web services build upon an old idea of accessing
resources (storage space, compute cycles, etc.) from a local machine on a powerful
remote computer. Unlike earlier attempts to provide this functionality, WS are a
broadly-accepted and open standard that is supported by all major industry vendors.
The communication protocol to access services is based on the simple object access
protocol (SOAP),5 which uses XML syntax. A Web service interface is formally
described with a Web Service Description Language (WSDL) file. A new service
can be registered with a universal description discovery and integration (UDDI)
server, and it can then be located, explored, and accessed from computers around
the Internet. Figure 3 illustrates the Web services paradigm with a geotechnical
data repository. Multiple applications, such as soil liquefaction analysis or borehole
data visualization can be built in a modular fashion. The next section describes two
applications in detail. A traditional, browser-based interface can also be provided
if desired.
This chapter describes our efforts to design and implement such an infrastructure
based on an extensible set of Web services. The services that are initially required
are: (1) file storage and (2) query and exchange. With these facilities, practitioners
in the field can directly store newly-acquired data in a repository while the data
customers are able to access these data sets in a uniform way. Beyond these two
basic services, we have also designed a (3) visualization capability that allows an
automated conversion of XML geotechnical data into a graphical view similar to
the traditional hardcopy format. The output is presented as Scalable Vector Graph-
ics (SVG).6
Later in this chapter we introduce the proposed WS architecture for Geotechni-
cal Information Management and Exchange, hereafter referred to as GIME. We
continue in next with a detailed description of the GIME Web services. Then we
illustrate the usability of the GIME approach with two client applications. Related
work is discussed in next. Conclusions and future research directions are contained
are discussed last.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

Figure 3. A Web service architecture enables client applications to access remote in-
formation through a machine-level interface for further processing and analysis

Service Discovery
WSDL.Binding UDDI
Information Registry
Get.Location.and
Bind.Request
Application.1:
Soil.Liquefaction.Analysis

Data
Source Web.Service.1 Internet
Web.Service.2
Application.2:
Data Geotechnical.Data.Visualizati on

Source
Other.Applications

Geotechnical. Information............................
Management. and. Exchange. (GIME)

Our proposed architecture for the Geotechnical Information Management and

Exchange system makes extensive use of Web services to provide access to these
valuable data sets for the expert practitioners as well as the general public. We will
now describe each component of the system in detail.

GIME.Overview

Figure 4 illustrates the architecture of the GIME system. Multiple, distributed

geotechnical data archives are accessible via the Internet. These repositories are
under different administrative control, for example, the U.S. Geological Survey
provides some of their information to the public. We distinguish two types of ar-
chives on the basis of what data access they allow: read-write (RW) or read-only
(RO). RW archives host three geotechnical Web services, and additionally an XSLT
(XSL Transformations) processor for transforming XML files to SVG files, and a
database (with XML and spatial data query support). The three Web services that
are implemented within the application server provide the interface for distributed
applications to store (File Storage Web Service, FSWS), query and retrieve (Query
& Exchange Web Service, QEWS), and visualize (Visualization Web Service, VWS)
the geotechnical information. RO archives implement only the QEWS Web service
to access their database.
The services use the simple object access protocol (SOAP) as the communication
protocol between the server and clients. A complete Web Service Description Lan-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

guage (WSDL) file of the system provides two pieces of information: an application-
level service description (abstract interface), and the specific protocol-dependent
details that users must follow to access the service at a specified concrete service
endpoint. These services will be registered to a Universal Description Discovery
and Integration (UDDI) server, and users can retrieve the information about these
services from the Internet. The client program can use its service proxy (static bind-
ing) to request these services from the GIME system.

GIME.Web.Services.Functionality.

Geotechnical data may be added to an archive in two different ways. First, an on-
site database administrator may insert the data directly into the local database. In

Figure 4. The proposed geotechnical information management and exchange archi-

tecture composed of multiple, distributed data archives; some archives are read-only
while others allow reading and writing; Web services provide the programmatic
interface to the data

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

this case, insert operations are not available via Web services, and we consider this
archive read-only within the GIME architecture. Second, geotechnical data may be
uploaded into a database via the file storage Web service. Hence the repository is
considered read-write. Any uploaded XML file is first placed in a temporary space
where it is validated against the document type definition (DTD) or XML schema7
for geotechnical data sets. If the file is accepted, the data is then stored in the main
database.
The stored data can be queried and downloaded via the QEWS interface. Cur-
rently GIME supports a number of predefined queries that are further described in
earlier. Note that each repository may contain a unique set of geotechnical data. To
avoid that a user application must contact each repository directly, we implement
a distributed query mechanism that automatically forwards queries to other known
archives and collects the results before returning the data to the application. Such
forwarding mechanisms can be effective as demonstrated earlier by the SkyQuery
project (Malik, Szalay, Budavari, & Thakar, 2003) and our own distributed query
routing techniques (Zimmermann, Ku, & Chu, 2004). At the current stage the query
results are complete borehole data sets. In our local repository, we have added the
PostGIS package to support spatial queries on the geotechnical data. For example,
users can query boreholes based on their geographical location.
Finally, the visualization Web service, VWS, translates the original geotechnical
data files into graphical data images (in SVG format) via an XSLT processor. We
use SVG because it is based on XML and is hence easily compatible with the overall
architecture.

GIME.Access.Authentication

The main purpose of geotechnical Web services is to assist in the utilization of

geotechnical information by local, state, and federal agencies mandated to develop
and sustain our civil infrastructures. For some particular applications, a service
may need to control the authentication of the users who want to access geotechnical
Web services. Some entry of information into the system database might have to be
restricted and controlled to expert members of key organizations. Some other Web
services such as data exchange and visualization may not require such stringent
security levels and may be anonymously accessed through read-only queries. Some
tools (e.g., the Microsoft .NET framework, IBM WebSphere, etc.) may be useful
to construct several different levels of a security model for certificate users. Our
current GIME prototype does not yet implement fine-grained access control.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

Geotechnical. Web. Service.........................

Functionality. and. Implementation

The three main geotechnical Web services of the system provide a number of
specialized methods for programmatic access to the data, and the parameters and
variables are described in the GIME WSDL file.8 These methods follow the W3C

Figure 5. The logical organization of the GIME server components (left) and the
architecture for the borehole query (BQ) client application (right); the communica-
tion between the client and the server is based on the SOAP and HTTP protocols.

GIME Web Services

GIME Borehole Query.(BQ)
Apache.Tomcat (Web server)
Apache Axis (Appl. Server)
Java
GIME Services.Code Xalan

JDBC Axis.1.0 Batik.1.5

HTTP./.SOAP
PostgreSQL+
PostGIS

Table 1. Software components and toolkits that are leveraged to build GIME
Name.of.Component Description
Apache Axis Axis is a Web service middleware that acts as SOAP server and client. Axis
handles the task of converting Java objects to SOAP requests when they are sent
and received over the network.
Apache Tomcat Tomcat is the official reference implementation for the Java servlet and JavaServer
pages technologies. It serves as a servlet engine (mainly for Axis) and a Web
server for GIME.
Apache Xalan Xalan is an XSL Transformation (XSLT) processor for converting XML docu-
ments into HTML, SVG or other XML document types. It generates on the fly
SVG files from XML borehole files.
Apache Batik Batik is a toolkit based on Java. It can be used in Java applications or applets for
viewing, generating, and manipulating scalable vector graphics (SVG) format
images. Batik library is used in one of our Java client applications for viewing
SVG files produced by GIME.
PostgreSQL PostgreSQL is an object-relational database and can be spatially enabled by
the PostGIS package. PostgreSQL and PostGIS provide the backend database
and spatial functionalities of GIME.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

standardized data-types and structures of the SOAP9 specification to provide an

easily-usable interface for client program construction. Web services usually oper-
ate from a combination of a Web server and application server; they can be imple-
mented using many available tools and libraries. GIME utilizes the open source
software components listed in Table 1. The GIME Web server is Apache Tomcat10

Table 2. Description of the GIME application programming interface (API)

Name.of.Function Description.of.Function. Restriction Related..

(client.perspective) Web.Service

GeoPutDTD() Upload a new DTD file and update the current DTD Super user FSWS
on server. Input parameters: a DTD identification
number and a DTD file

GeoGetDTD() Send a unique identification number of a DTD file All users FSWS
and the server returns a string- type object to the
client program. Input parameter: a DTD identifica-
tion number; return value: a string-type object

GeoPutXMLFile() Upload a borehole XML file and store it in the Only in write en- FSWS
server. Input parameters: a borehole XML file abled databases
name and a borehole XML file

GeoGetXMLFile() Send a unique identification number of a borehole All users QEWS

XML file, and the server returns a string-type
object to the client program. Input parameter: a
XML file identification number; return value: a
string-type object

GoeQuery() Execute the input query statement which is parsed All users QEWS
by this function to select tuples inside tables. All
the result tuples are stored in a result structure
and returns to the client. Input parameters: query
parameters; return value: a result data object

GeoVisualization() Send unique identification number of a borehole All users VWS

file and display its SVG picture inside a pop-
up window. The transformation from XML to
SVG format is executed on the fly and the SVG
file is rendered in the client. Input parameter: a
borehole file identification number; return value:
a SVG file

GeoGetSVGFile() Send unique identification number of a borehole All users VWS

file and returns its SVG file as string type. Input
parameter: a borehole file identification number;
return value: a SVG file

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ku & Zmmermann

and its application server is Apache Axis.11 The application code specific to GIME
is embedded within Axis, which is convenient to use; when Tomcat starts, Axis
automatically compiles the application codes located in its working directory and
generates the necessary object class files. Figure 5 shows the components of the
GIME Web services and all the application programming interface (API) methods
are listed in Table 2. We now describe the overall functionality and implementation
of each GIME Web service.

Geotechnical File Storage Web Service

The geotechnical file storage service provides client programs with an interface to
upload their geotechnical XML files into the main database. All files are stored in
XML format, and at the same time, meta-data is extracted and saved. The meta-data
includes specific elements of the imported files to facilitate querying (described
earlier). This service is subject to user authentication because only registered expert
users are allowed to make changes to the database. We currently utilize the open
source PostgreSQL object-relational database with the PostGIS package. To manage
the data we implemented two different paradigms to store XML data. The first one
extracts elements of interest from an XML file and stores them in table columns
as meta-data. The XML file itself is also stored in its entirety in a column of type
XML in the main table. Hence, the meta-data columns can be rapidly accessed when
users want to query and retrieve XML information in the database. This second ap-
proach to store XML data is to completely decompose the elements of an XML file
into columns of multiple tables. The second process is bi-directional, which means
that the XML files can be regenerated after individual elements have been updated.
Furthermore, new XML files can be generated from multiple existing tables.

Verification of the Geotechnical XML Files

Two levels of verification are performed on uploaded files. The first level is executed
automatically by checking the XML data against a borehole DTD to ensure that
the uploaded file is well formed. The DTD defines geotechnical tags that allow the
full description of data provided by our partner organizations (California Depart-
ment of Transportation, California Geological Survey, Los Angeles Department of
Water and Power, etc.). If any file cannot pass the first level of verification, it will
be rejected and an error condition will be raised. Otherwise the file will be stored
in a temporary directory and a notification e-mail will be sent to an expert of that
category of geotechnical files. The expert verifies the content of the uploaded file
and decides if it can be stored into the main database or should be returned to the
author for clarification. After passing the two levels of verification, the file will be
imported into the database and its meta-data will be extracted. Users can down-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

load the DTD files from the project Web site for guidance when constructing their
geotechnical XML files. They can also use the verification interface on our Web
site for remote checks.

Geotechnical.Query.&.Exchange.Web.Service

The main purpose of the query and exchange Web service is to facilitate the dis-
semination of the valuable geotechnical data sets and encourage their use in broad
and novel applications. Once XML files are uploaded into the main database as
described in the previous paragraphs, they are then available for dissemination.
A data user can query the geotechnical files in the main database on meta-data
attributes such as the name of the data provider, the geographical location of the
borehole, and so forth, to find the files of interest. The files can then be downloaded
with the GeoGetXMLFile() method. A graphical version of the data in SVG format
can also be obtained.

Predefined Queries of the Data Exchange Web Service

In the current GIME version, we have identified six parameters for extracting
particular sets of geotechnical files from the main database; these parameters are
also indexed as meta-data. Client programs can use these parameters in the query
interface to find the geotechnical files of interest. At this point we restrict the query
interface to these six parameters to gain some experience with our architecture
and to simplify the distributed processing of queries. Spatial query constructs are
supported within the database management system via a PostGIS package. The six
parameters are:

1.. Project. Name. [Type:. string]: The individual name of geotechnical proj-
ects
2.. Site.[Type:.string]: The field site name where borehole testing has been car-
ried on
3.. Location.[Type:.polygon]: The geographical region within which boreholes
have been drilled
4.. Testing.type.[Type:.string]: Field investigation type (e.g., SPT, CPT, etc.)
5.. Project.Date.[Type:.date]: The date of the beginning of the project
6.. Data.Provider.[Type:.string]:.The organization or agency who produced the
data

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

Recall that the GIME system is an evolving federation of autonomous and geographi-
cally-distributed geotechnical databases on the Internet. To facilitate the querying
process, we require a client application program to only contact one of the GIME
nodes and then have this node forward the query to and collect results from other
GIME databases that are part of the federation. Each node maintains a list of other
known GIME nodes. A node recognizes when it receives a query from another
GIME node and will not forward the query any further so as to avoid loops in query
processing. The result of a query is a list of the borehole data sets that satisfy the
query predicate. Each data set is identified with a link descriptor similar to a URL.
All the qualifying links are collected from the distributed GIME nodes and returned
to the user application to form the result of the query. If the user application decides
to perform any processing on the data sets, then it can download the data directly
from each individual node via the link descriptor. The details of this query routing
mechanism are contained in (Zimmermann et al., 2004).

Using.the.Web.Service.of.TerraServer..

Our query interface allows boreholes to be searched by geographical location. A

background map with superimposed borehole locations is a compelling way to
present the search results, as illustrated in Figure 6. However, rather than storing
aerial maps in the system, the functionality of the Microsoft TerraServer is enlisted

Figure 6. Geotechnical borehole locations superimposed onto an aerial map

image (San Fernando Valley, Los Angeles)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

Figure 7. Borehole SVG file rendering (the right window) that was automatically
generated from a selected XML data file

(Barclay, Gray, Strand, Ekblad, & Richter, 2002). The TerraServer provides maps
(aerial imagery and topographical maps) covering the United States with up to one-
meter resolution. Furthermore, it provides Web service interface methods to obtain
these background maps. Our system then plots the geotechnical borehole locations
on top of the map according to their longitude and latitude.

Geotechnical.Visualization.Web.Service

XML borehole files, although easily readable by computers, become meaningful

to geologists and civil engineers only after they are rendered into SVG images, for
example, stratigraphy of soil deposits and geotechnical test results (e.g., CPT and
SPT). We use the Apache Xalan parser12 in combination with Extensible Stylesheet
Language (XSL) files for this purpose. The parser translates XML files into SVG
descriptions according to a set of XSL rules. After the translation, the SVG data is
transmitted to the client, and the user can display and analyze them (Figure 7). Data
may not only be visualized from the main database, but also from temporary data
provided by users who do not have database write privileges. Again, the user must
construct their geotechnical XML files in accordance with the DTD files that are
published by our system. They may then upload the data and use the visualization
method to translate the file into the SVG format.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

GIME. Client.Application. Examples

The feasibility of the GIME concepts are illustrated using two examples of client
applications: (1) borehole query and (2) soil classification. The borehole query
client is a Java application, which demonstrates all of the GIME functionalities
through a graphical user interface (GUI). The soil classification client, which uses
the Microsoft Visual Studio .NET framework, demonstrates how to modify existing
geotechnical programs to directly import borehole data from GIME.

Borehole.Query.Client.Application

Shown in Figure 8, the Borehole Query (BQ) is a Java client application which
demonstrates all the features of GIME, that is, query and visualization of borehole
data, and upload/download of XML and DTD files. BQ was implemented using
the Java Software Development Toolkit (JDK, version 1.4.1).13 BQ accesses GIME
through the Axis library, and visualizes SVG files using Xerces14 and Batik15 (Table
1). BQ can be downloaded from the GIME Web site and installed as a standalone
Java program along with the Batik Squiggle package (Table 1) which displays SVG
files. As shown in Figure 8, five query parameters can be entered into the BQ dialog
box with optional AND/OR relations. The other query parameters, that is, the loca-

Figure 8. The sample Borehole Query (BQ) client application is programmed in

Java; the dialog box allows the user to input query parameters while the small pop-
up display (the left window) illustrates the metadata of a borehole when the user
clicks on one of the dots (indicating a borehole location) on the map

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

Figure 9. Sample Java code to invoke the “GeoQuery” Web service of the GIME
server; this code fragment would typically be executed by GIME clients
Import org.apache.axs.clent.Call; // The class files for SOAP communication
mport org.apache.axs.clent.Servce;
:
:
Strng endpont = "http://datalab.usc.edu:00/axs/jdbc_dbs.jws?wsdl"; // The GIME server URL
Servce servce = new Servce (); // Create connections to the
call = (Call) servce.createCall (); // GIME server
call.setTargetEndpontAddress( new java.net.URL(endpont) );
:
:
call.setOperatonName("GeoQuery"); // Invoke the GeoQuery function and preparing the query string
Strng[] queryStrng = { select
BOREHOLE_ID,PROJ_NAME,SITE,LONGITUDE,LATITUDE,TEST_TYPE,PROJ_DATE,DATA_PRO from dbnst.IDENTIFICATION_TAB
where . };
Vector result = (Vector) call.nvoke(queryStrng);

tion parameters, are defined using an interactive rectangular box displayed over a
background map. The background map is assembled from aerial maps downloaded
from the TerraServer Web service16 (Barclay et al., 2002). Figure 9 displays excerpts
from the BQ Java code for processing queries through the GeoQuery function of
GIME (Table 2). BQ sends the search parameters to GIME, retrieves the metadata
of matching borehole files, and displays the search results (i.e., borehole locations)
as dots over the background map (Figure 8). Borehole metadata can be viewed by
clicking on the dots. BQ allows users to invoke all the functions of GIME (Table 2)
through menus and buttons. Borehole XML files can be uploaded to the remote GIME
archive using the “Upload Geo” button. BQ prompts users to select the XML file on
the local disk before uploading it. After GIME has received a XML file, it compares
it to a specific DTD, and if that file conforms, inserts it into the GIME database, and
generates and stores the corresponding metadata. Borehole XML files are visualized
using the visualization menu, which downloads SVG files from GIME and displays
them using the Squiggle package (Figure 7). More details about the functionalities
of the Borehole Query Client are available from the GIME Web site.17

Figure 10. The block diagram of the visual basic (VBA) client application that
implements a core penetration test (CPT) soil classification (CSC)

Query & Retrieval

Initialization
SOAP. Display & Conversion
GIME
Server HTTP. Data Data
queryng retrevng View
XML

Data Checkng / Soil classification

convertng Excel Vsualzng
Additional
data
Save Soil
classification

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

Soil Classification Client Application

GIME can exchange borehole data not only with newly-created Java applications, but
also with many existing engineering applications. With minor modifications, stand-
alone engineering programs can import information from GIME. This integration
of GIME with engineering programs is illustrated with a CPT Soil Classification
program, hereafter referred to as CSC. CSC classifies soils based on cone penetration
tests (CPT) data and the soil classification method described in Robertson, Woeller,
and Finn (1992). CSC was developed starting from an existing Excel Visual Basic
Application (VBA), which required importing data manually into spreadsheets.
As shown in Figure 10, CSC has three main functions: (1) query and retrieval of
borehole data via GIME; (2) display and conversion of borehole XML data; and
(3) soil classification. CSC added to the existing VBA application the capabilities
of querying and retrieving XML borehole data from GIME and pasting those into
spreadsheets. The first modification to the existing VBA application was to import
the Web service References Tool 2.0, which is a resource toolkit adding the Web
service functionalities to the Excel VBA environment. With the Web service Refer-
ences Tools, VBA can call functions for discovering Web services, select particular
Web services, and create proxy classes of Web service objects. With these refer-
ences, VBA can access the WSDL files of GIME, and the GIME objects defined

Figure 11. The query box (left) and the query results (right) of the CSC client ap-
plication; the borehole XML data is decomposed in a tree-like structure

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

Figure 12. Excerpt from the VBA subroutine DisplayXMLNode that converts the
content of an XML file into the Excel tree view shown in Figure 11

Sub DsplayXMLNode(ByVal xmlnode As IXMLDOMNode, ByVal root As Strng,  As Integer)

' Add a TreeView node for this XMLNode. (Using the node's Name is OK for most XMLNode types)
Dm tvnode As Node
Set tvnode = tvXML.nodes.Add(root, tvwChld, xmlnode.nodeName & , xmlnode.nodeName)
tvnode.Expanded = False
tvnode.EnsureVsble
root = xmlnode.nodeName & 
 =  + 
Select Case xmlnode.nodeType
Case NODE_ELEMENT ' This is an element: Check whether there are attributes .
If xmlnode.Attrbutes.Length > 0 Then ' Create an ATTRIBUTES node.
Dm attrNode As Node
Set tvnode = tvXML.nodes.Add(root, tvwChld, root & "attr", "(ATTRIBUTES)")
Dm xmlAttr As IXMLDOMAttrbute ' Add all the attributes as children of the new node .
For Each xmlAttr In xmlnode.Attrbutes ' Each node shows name and value.
Set tvnode = tvXML.nodes.Add(root & "attr", tvwChld, "", xmlAttr.Name & " = '" & xmlAttr.Value & "'")
Next
End If
Case NODE_TEXT, NODE_CDATA_SECTION ' For these node types we display the value
tvnode.Text = xmlnode.Text
Case NODE_COMMENT
tvnode.Text = "<!--" & xmlnode.Text & "-->"
Case NODE_PROCESSING_INSTRUCTION, NODE_NOTATION
tvnode.Text = "<?" & xmlnode.nodeName & " " & xmlnode.Text & "?>"
Case Else ' Ignore other node types.
End Select
Dm xmlChld As IXMLDOMNode ' Call this routine recursively for each child node.
Set xmlChld = xmlnode.FrstChld
Do Untl xmlChld Is Nothng
Call DsplayXMLNode(xmlChld, root, )
Set xmlChld = xmlChld.nextSblng
Loop
End Sub

in the proxy class. As shown in Figure 10, CSC checks at first for the availability
of GIME services. If the service is available, CSC retrieves metadata about the
locations of borehole sites and the number of data sets at each site, and adds this
metadata to a list of available sites (site list box in Figure 11). CSC forwards the
selection parameters input into its query dialog box to GIME, which replies with
a list of selected boreholes (Figure 11). XML borehole files are downloaded using
the GeoGetXMLFile function of GIME (Table 2). They are loaded into memory
using the LoadXML function of the XML Document Object Model (DOM), which
is detailed in Barwell, Blair, Crossland, Case, Forgey, Hankison, Hollis, Lhotka,
McCarthy, and Roth (2002) and Balena (2002). Figure 12 shows excerpts from the
subroutine DisplayXMLNode which converts XML files into XML document objects
and display them as a tree structure (Figure 11). That display is useful to inspect
the structure and contents of XML files before pasting them into spreadsheets. The
VBA code also uses a DOM function to paste onto spreadsheets the contents of the
“CPT” elements of XML borehole files. Each “CPT” element, which contains the
values for depth, CPT tip resistance qc and sleeve friction Sf, is parsed and written
sequentially onto spreadsheets (Figure 13). Figure 13 illustrates the final result of
the CPT soil classification obtained by CSC. The techniques that CSC uses for

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

Figure 13. The Excel application display showing the borehole data and the CPT
analysis results processed with the CSC client application

reading, inspecting, and pasting the contents of XML files can be adapted to many
other client applications.

Related.Works

During recent years, several geotechnical Web services have been proposed and
implemented. The majority of this work focuses on the design of geotechnical in-
formation exchange platforms and integration of geotechnical Web services from
different providers. Certain design goals of these prototype systems overlap with
the GIME system, and we summarize them as follows.
Badard and Braun (2004) proposed a platform named OXYGENE for the develop-
ment of interoperable geographic applications and Web services. OXYGENE aims
at providing users with an open framework for the development of geotechnical
research applications and allows for the centralization of code, documentation, and
easy maintenance. There are three main characteristics of the OXYGENE platform.
First, the architecture and the object-oriented schema of OXYGENE are compliant
with OGC/ISO specifications. Second, OXYGENE adopts modularized design and
all the components (e.g., DBMS, map server, etc.) can be easily updated. Third, a
generic graphical object browser has been embedded in clients of OXYGENE, and
most geographical data objects can be visualized by users.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

A concrete design and implementation of an integration and interoperation platform

that aims to solve heterogeneity and online service for geographical information is
presented in Luo, Liu, Wang, Wang, and Xu (2004). This platform aims at providing
service integration which takes layers as the basic geographical data organization
units, accomplishing data retrieval from multiple GIS data sources, and visualizing
digital map data. There are four main Web services provided for achieving the design
goals: (1) application integration services, which is the core service of business process
scheduling to implement interoperation, (2) metadata service, which maintains the
service information, (3) transform service, which can be implemented into platforms
of different GIS providers to accomplish data format transformation into Geography
Markup Language (GML), and (4) spatial analysis service, which receives GML
format data and provides GML-based common spatial analysis functionalities as
data clipping, projection, coordinate transformation, and overlapping.
In addition some researches have focused on automated service discovery, binding,
and composition of geotechnical Web services. An, Zhao, and Bian (2004) proposed
the utilization of patterns with geotechnical Web services based on ontology. Their
design employs ontology to advertise geotechnical Web services and uses them for
semantics-based discovery of relevant geotechnical Web services. Furthermore, in-
tegration of geotechnical Web services can also be implemented based on ontology
and system developers can integrate geotechnical functions into their applications
without having to host all the functionalities locally.

Conclusion. and. Future. Research. Directions

We have presented research results on the development of Web services for exchang-
ing and utilizing geotechnical information across a variety of applications critical
to our civil infrastructure systems. The proposed Web services are being presently
tested in realistic work environments in collaboration with municipal, state, and
federal agencies, as well as international partners. It is likely that our first prototype
of geotechnical Web services will evolve in different directions depending on the
needs of our research collaborators. It is anticipated that the concepts developed in
this research will contribute to pave the way for the creation of a virtual repository
of geotechnical information operated by a consortium of local, state, and federal
agencies.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ku & Zmmermann

Acknowledgments

We would like to thank Jianping Hu, Yu-Ling Hsueh, Chirag Trivedi, and Xiaohui
He for their help with the GIME system implementation. We also wish to thank our
collaborator on the COSMOS project, Jennifer Swift. The GIME design, implemen-
tation, and experiments were made possible by the NSF grants ITR CMS-0219463
and ERC Cooperative Agreement No. EEC-9529152. Any opinions, findings, and
conclusions or recommendations expressed in this material are those of the authors
and do not necessarily reflect those of the National Science Foundation.

References

AGS (1999). In Electronic transfer of geotechnical and geoenvironmental data (3rd

ed.). Kent, UK: Association of Geotechnical and Geoenvironmental Special-
ists.
An, Y., Zhao, B., & Bian, F. (2004). Geo Web services based on semantics. In ER
Workshops, 3289, 139-157. Shanghai, China: Springer.
Badard, T., & Braun, A. (2004). OXYGENE: A platform for the development of
interoperable geographic applications and Web services. In DEXA Workshops,
888-892, Zaragoza, Spain: IEEE Computer Society.
Balena, F. (Ed.). (2002). Programming Microsoft visual basic.net. Microsoft
Press.
Barclay, T., Gray, J., Strand, E., Ekblad, S., & Richter, J. (2002, June). TerraService.
NET: An introduction to Web services (Tech. Rep. No. MSR-TR-2002-53).
Microsoft Research.
Bardet, J. -P., Hu, J., & Swift, J. (2003). Data storage and exchange format of bore-
hole information. Submitted to Computers & Geosciences.
Barwell, F., Blair, R., Crossland, J., Case, R., Forgey, B., Hankison, W., Hollis, B.
S., Lhotka, R., McCarthy, T., & Roth, J. C. (2002). Professional VB .Net (2nd
ed.). Wrox Press Ltd.
Benoit, J., & Lutenegger, A. J. (2000, April). National geotechnical experimentation
sites (NGES). ASCE Geotechnical Special Publication, 408(93), 221-234.
COSMOS/PEER User Survey. (2002). Archiving and Web dissemination of geo-
technical data Website. COSMOS/PEER-LL 2L02, USC. Retrieved December
1, 2005, from http://geoinfo.usc.edu/gvdc/

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIME 

Heslop, K., Karst, J., Prensky, S., & Schmitt, D. (2000, June). Log ASCII standard
document #1 — file structures, Version 3. Canadian Well Logging Society.
Calgary, Canada, p. 44.
Holtzer, T. (2001, October). Distribution of USGS CPT data via the Web. In Proceed-
ings of the COSMOS/PEER-LL Workshop on Archiving and Web Dissemination
of Geotechnical Data. Richmond, CA: U.S. Geological Survey.
Hunt, R. R. (1984). Geotechnical engineering investigation manual. New York:
McGraw-Hill.
Luo, Y., Liu, X., Wang, W., Wang, X., & Xu, Z. (2004). Web service and geographi-
cal information integration. In Proceedings of the COMPSAC Workshops (pp.
130-133). Hong Kong, China: IEEE Computer Society.
Malik, T., Szalay, A. S., Budavari, T., & Thakar, A. R. (2003, January). SkyQuery:
A Web service approach to federate databases. In Proceedings of the First
Biennial Conference on Innovative Data Systems Research (pp. 188-196).
Asilomar, CA: VLDB Endowment.
Robertson, P. K., Woeller, D. J., & Finn, W. D. L. (1992). Seismic cone penetra-
tion test for evaluating liquification potential under cyclic loading. Canadian
Geotechnical Journal, 15(29), 686–695.
Swift, J., Bardet, J. -P., Hu, J., & Nigbor, R. (2002). An integrated RDBMS-GIS-IMS
system for dissemination of information in geotechnical earthquake engineer-
ing. Accepted by Computers & Geosciences.
U.S. Army Waterways Experiment Station (1998). Standard data format for geo-
technical/geological exploration (Project 98.005). Retrieved December 1,
2005, from http://tsc.army.mil/products/geotech/geocover.asp
Zimmermann, R., Bardet, J. -P., Ku, W. -S., Hu, J., & Swift, J. (2003, July). Design
of a Geotechnical Information Architecture Using Web Services. In Proceed-
ings of the Seventh World Multi-Conference on Systemics, Cybernetics, and
Informatics (pp. 394-398). Orlando, FL: IIIS Publication.
Zimmermann, R., Ku, W. -S., & Chu, W.-C. (2004, November). Efficient query
routing in distributed spatial databases. In Proceedings of the International
Symposium Advances in Geographic Information Systems (pp. 176-183).
Washington, DC: ACM Press.

Endnotes
1
http://www.ejge.com/GML/
2
https://www.seegrid.csiro.au/twiki/bin/view/Xmml/WebHome

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ku & Zmmermann

3
http://www.opengeospatial.org
4
http://www.w3.org/2002/ws/
5
http://www.w3.org/2000/xp/Group/
6
http://www.w3.org/Graphics/SVG
7
http://www.w3.org/XML/Schema
8
http://datalab.usc.edu/axis/CLS\_Services.jws?wsdl
9
http://www.w3.org/TR/soap12-part0/
10
http://jakarta.apache.org/tomcat/index.html
11
http://ws.apache.org/axis/index.html
12
http://xml.apache.org/xalan-j/
13
http://java.sun.com/j2se/index.jsp
14
http://xml.apache.org/xerces2-j/
15
http://xml.apache.org/batik/
16
http://terraserver.microsoft.com/
17
http://datalab.usc.edu/gime/

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Chapter.VI

ScienceMaps:
An.Online.Resource.Portal.for.
Standards-Based.Science.
Instruction.Using.Geographic.
Information.System.Technology
June K. Hlton, Claremont Hgh School, USA
Davd E. Drew, Claremont Graduate Unversty, USA

Abstract

At the beginning of the 21st century, technology implementation and its effect on
student achievement is a topic of much interest and debate. Science education in
particular has been criticized for not utilizing the skills of a “techno-savvy” genera-
tion and for not effectively integrating technology into the K-12 science classroom.
This chapter discusses ScienceMaps, an ongoing research and development effort
consisting of an online resource portal for standards-based science instruction us-
ing GIS technology. ScienceMaps is unique in that it concentrates on using GIS to
teach, not on teaching GIS. Using an Internet-based GIS, ScienceMaps provides
access to GIS technology and data to anyone, anywhere, with access to an Internet
browser. Assessment, evaluation, and future development directions for ScienceMaps
are also discussed.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

Introduction

The issue of the effect of technology on student achievement has recently been the
subject of much research and debate (J. Hilton, 2003; Hilton, 2005, 2006). Further-
more, as schools are being held more accountable for meeting state and national
standards through their performance on standardized tests, the focus on improving
student achievement through the use of technology is becoming an even greater is-
sue. As school funding becomes more closely linked to performance on standardized
tests, improving student achievement through innovative means, such as technology
integration and use, is critical (Hilton, 2006). ScienceMaps is an online resource
portal that provides standards-based instructional materials, designed and integrated
for use with an Internet-based geographic information system (GIS), that allows
for their immediate integration into science lessons. GIS technology is a powerful
analytical tool that can help students visualize spatial relationships and supports
the spatial analysis of data.
The goals of the ScienceMaps research and development effort are twofold: to sup-
port the ongoing development and field-testing of this innovative Internet-based
courseware and to promote the integration of technology into the science curriculum
so as to develop the problem-solving and critical-thinking skills of students while
simultaneously addressing the diverse learning styles of learners.

Background

Archer (1998) believes that computers can raise student achievement and even
improve a school’s climate. Eliot Levinson (2000) agrees but adds that many fac-
tors, such as staff development, infrastructure, and effective instructional materi-
als, influence the effectiveness of technology. Simply put, if schools are to realize
benefits from education technology, teachers and students must have adequate and
equitable access to hardware and network connections; states and districts must give
schools the capacity to use technology well by devising a thoughtful technology plan
and offering adequate teacher training and technical support; teachers and students
must use technology in effective ways (Jerald, 1998). Thus, not only must teachers
be adequately trained in the technology, but they must also understand how to use
the technology appropriately in support of students and the curriculum (Hanson
& Carlson, 2005). This effective use of technology can only help to “re-excite”
students about science. According to Drew (1996), “the number of students who
find science interesting seems to shrink as they progress through the school system.
Good teaching involves knowing what methods are effective” (p. 6).
Even though GIS is now available to K-12 education, modest budgets for technol-
ogy and staff development, and limited emphasis on students developing spatial

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

analytical skills, prevent schools from effectively integrating the technology into
the classroom curriculum (Joseph, 2004). Thus, if schools are to be effective in
utilizing technology, particularly GIS, to raise student achievement, accessibility,
professional development, and effective use must be addressed.

Accessibility

Access to computers at school has risen significantly. According to Skinner (2002),

“Nationally, in 2001, there were just over four students to every instructional school
computer, and the number of students per Internet-connected computer in schools
dropped from 7.9 in 2000 to 6.8 in 2001” (p. 53). This trend continues. In 2004,
Fox (2005) reports that nationally, there were 3.8 students to every instructional
computer and the number of students per Internet-connected computer in schools
dropped to 4.1. Rumberger states that, newer and more powerful desktop comput-
ers, along with the development of computer networks and the Internet, provide the
opportunity for educational technology to transform teaching and learning in more
fundamental ways across the curriculum and throughout the educational system
(Rumberger, 2000).
The digital divide continues to exist between high (75% or more of the students
eligible for the federal free and reduced lunch program) and low poverty schools.
While Internet access across all types of schools has shown steady improvement,
the actual use of technology in high poverty, high minority, and academically-fail-
ing schools lags behind technology use in more advantaged schools (Fox, 2005). In
multiple measures of access, schools with a large number of poor students, receiving
free or reduced price lunch, rated lower than schools with smaller numbers of poor
students (Du, Havard, Olinzock, & Yang, 2004). Students who do not have access
to high-quality computer experiences at home or school are not being provided with
the opportunities they need to be successful in society. Furthermore, increased ac-
cess to computers will only have positive results when the educator has a complete
grasp of the role and use of computers, and an understanding of the student’s home
environment and how their deficiencies must be met in order to realize their full
potential, thus enhancing society instead of reducing the average achievement (Du
et al., 2004).

Professional.Development

Another important factor for effective implementation is staff development. Provid-

ing sufficient development and training to give staff skills and confidence in the
use of technology is widely viewed as an ongoing challenge to schools (Schmitt,
2002). Trotter (1999) reports that nearly four out of every ten teachers who do not

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

use software for instruction say they do not have enough time to try out software,
and almost as many say they do not have enough training on instructional soft-
ware. Adelman et al. (2002) found that although most teachers had participated
in multiple professional development opportunities, more than 80% indicated a
need for training in how to integrate technology into the curriculum. Ironically, in
2003, 82% of public schools nationwide with Internet access indicated that their
school or school district had offered professional development to teachers in their
school on how to integrate the use of the Internet into the curriculum (Parsad &
Jones, 2005). Interestingly, students believe that professional development and
technical assistance for teachers are crucial for effective integration of the Internet
into curricula (Levin, Arafeh, Lenhart, & Rainie, 2002). Unfortunately, schools
and school districts spend most technology money on hardware and software, and
spend relatively little on training, which averages no more than 15% of all technol-
ogy funds (Hanson & Carlson, 2005). In the technology context, teachers reported
having limited time to practice technology skills, to develop new activities, and to
conduct activities during a tightly scheduled school day. These were reportedly the
most significant barriers to increased integration of educational technology in the
classroom (Adelman et al., 2002).

Effective.Use

The third important factor to consider is effective use. Technology, it has been
argued, helps change teacher-student relationships, encourages project-based learn-
ing styles, and supports the acquisition of skills such as “higher order thinking,”
analysis, and problem solving (Schmitt, 2002). Thus, its integration and effective
use is critical to increase student achievement. Although teachers now have the use
of an unprecedented amount of technology for their classrooms and schools, there
is little evidence to indicate that teachers systematically integrate instructional tech-
nology into their classroom curriculum (Wetzel, 1999). Hedges, Konstantopoulos,
and Thoreson (2000) found that even though teachers have had increasing access to
computers for instruction, very few actually use them. Although 84% of all public
school teachers said personal computers were available to them, approximately 60%
indicated that they actually used them (U.S. Bureau of the Census, 1998). While
75% of teachers say the Internet is an important tool for finding new resources to
meet standards, two-thirds agree that the Internet is not well integrated in their class-
rooms, and only 26% feel pressure to use it in learning activities (Ely, 2002). Thus
even though computer technology may be widely available, in general, it is poorly
integrated into the classroom curriculum and is under-used (Hedges et al., 2000).
Unfortunately, the manner in which technology is used in schools also depends on
socioeconomic status. Disadvantaged children, even with access to new technolo-
gies, are more likely to use them for rote learning activities rather than for intel-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

lectually-demanding inquiries (Du et al., 2004). Equally disturbing is the evidence

that teachers of students with different ability levels are also using the computer
differently. Research indicates that technology use with low-achieving students
is mostly skills-based, drill and practice, while more sophisticated programs are
used with advanced students (Becker, 2000; Manzo, 2001). While all teachers use
technology in multiple areas of their work and believe in technology’s potential to
positively impact teaching and learning, instructional use (e.g., student projects,
accommodations, lesson presentation) is one of the least frequent ways teachers
employ technology (Hanson & Carlson, 2005). However, researchers have concluded
that technical tools, when used appropriately as instructional supports, have the
potential to enhance student learning and teacher instructional success (Bednarz,
2004). Williams believes that education will be affected by how educators and stu-
dents use the technology to prepare for life-long learning in the face of unrelenting
change (Williams, 2002).
The CEO Forum 2001 Report (CEO Forum, 2001) found that while students fre-
quently use computers at school for research (96%) or to write papers (91%), their
actual use for learning new concepts or practicing new skills learned in class was
significantly lower (60% and 57% respectively). Quintana and Zhang (2004) report
that online inquiry activities are important for all K-12 learners to explore substan-
tive driving questions in different areas, especially science.

Geographic.Information.Systems

Geographic information science is an approach to measuring and understanding the

spatial context of the human and physical environment. By combining theories and
methods from many disciplines, and use of information technologies, its purpose is
to provide geographic representations of our world that allow for visualization and
spatial analysis through the use of geographic information system (GIS) technology.
For example, the earth’s climate, natural hazards, population, geology, vegetation,
soils, land use, and other characteristics can be analyzed in a GIS using maps,
aerial photographs, satellite images, databases, and graphs. GIS technology is used
in many areas such as agriculture, business, earth and space sciences, education,
energy, emergency management, health sciences, life sciences, logistics, physical
sciences, telecommunications, and transportation.
In education, using GIS to analyze phenomena about the Earth and its inhabitants can
help students to better understand patterns, linkages, and trends about our planet and
its people (Gutierrez, Coulter, & Goodwin, 2002). Integrating GIS into the school
curriculum answers the call for including critical thinking, integrated learning, and
multiple intelligences in curriculum design (Joseph, 2004).
Students who are visual learners particularly benefit from the mapping process, as
the presentation of the data is consistent with their cognitive strengths (Gutierrez

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

et al., 2002). Research has shown that educators must be attuned to the individual
differences in students’ strategical information processing styles, and an awareness
of the student’s style allows for the design of educational experiences that tap into
the student’s cognitive resources (Farrell & Kotrlik, 2003). Teachers realize that
technology enhances science, technology, engineering, and mathematics (STEM)
instruction by providing resources for individual students, as well as for the whole
class, and serves as a means to address individual learning needs (Hanson & Carlson,
2005). Thus, GIS should be emphasized throughout K-12 education where it can be
used to teach concepts and skills in earth science, geography, chemistry, biological
science, history, and mathematics.

ScienceMaps

ScienceMaps was created to promote the integration of GIS technology into the science
curriculum and to address the aforementioned issues of accessibility, professional
development, and effective use. As a result, the ScienceMaps project seeks to:

• Develop standards-based instructional materials for use with an Internet-based

GIS and allow for their immediate integration into science lessons (accessibil-
ity and effective use)
• Reduce the necessary technical skills needed to effectively integrate technology
into the science classroom and reduce the time needed to preview/evaluate
instructional materials (professional development and effective use)
• Provide access to these resources through an online resource portal and in so
doing reduce the need for schools to purchase expensive software/hardware
for students and teachers to access the instructional materials and applications
(accessibility and effective use)

Current science learning standards promote inquiry teaching as a means to help

students develop a deeper conceptual understanding of science (Borgman et al.,
2004). Drawing on the California State Science Standards, numerous science les-
sons and GIS applications have been developed. California was selected for several
reasons:

• California has the largest pre K-12 enrollment (6,356,348) as well the largest
number of public school teachers (307,672).
• California ranks fourth in the country in capacity to use technology, that is,
number of technology- related policies state has in use, but 30th in the country
in use of technology.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

• California ranks 47th in the country in access to technology and is one of only
five states that have a plan to regularly update technology, but no funding
mechanism to ensure that technology is updated (Fox, 2005).
• California was 1 of 7 states to receive an “A” rating on its science standards,
receiving top honors for its standards. According to a study by the Thomas B.
Fordham Institute, “California has produced an exemplary set of standards
for school science; there was no question among readers about the “A” grade”
(Gross, Goodenough, Haack, Lerner, Schwartz, & Schwartz, 2005, p. 30).

Until the emergence of GIS and related mapping technologies, constructing maps
by hand was an elaborate, time-consuming task that often required good drawing
skills. It is not necessary for users of ScienceMaps to “learn GIS” as the online
applications are developed in such a manner that the necessary level of technical
proficiency for both student and teacher is minimal. However, the spatial analysis
performed using the applications is sophisticated and helps to develop students’
critical thinking skills through the practice of spatial thinking (Bednarz, 2004). In
addition to supporting the visual display of spatial data, students can use the online
GIS applications to perform various quantitative analyses. Students can also gain
confidence in the use of advanced technologies, giving them additional technology
skills for future employment. It has been noted that there is a serious shortfall in
professionals and trained specialists who can utilize geospatial technologies in their
jobs, which indicates the need for the education, training, and development of these
individuals (Gaudet, Annulis, & Carr, 2001).

ScienceMaps.Online.Resource.Portal

Major science education standards call for K-12 students to engage in online, hands-on
science activities where they can investigate personally meaningful scientific driving
questions. When students learn science through inquiry, they are imitating practicing
scientists (Borgman et al., 2004 ). The use of the Internet has led to a paradigm shift
in education from traditional didacticism to technology-rich, constructivist learning
where the learner is more active and independent, and the processes of teaching and
learning are emphasized (Houtsonen, Kankaanrinta, & Rehunen, 2004).
One aspect of online inquiry involves information seeking. As a result, there is
an increased focus on information seeking support for K-12 learners, such as new
collections in larger projects like the National Science Digital Library (NSDL)
(Quintana & Zhang, 2004). As educators increasingly look to technology to meet
the challenges of teaching and learning in a rapidly-changing educational environ-
ment, the field of interactive visualization, illustrating educational concepts through
visual, interactive applications, and simulations, presents a promising and emerging
paradigm for learning and content delivery (Marsh et al., 2005). Research on the
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Hlton & Drew

use of digital resources among STEM teachers found that digital resources had the
added benefit of providing opportunities to supplement limited resources, such as
labs (Hanson & Carlson, 2005).
The ScienceMaps Online Resource Portal was developed using the Collection
Workflow Integration System (CWIS) a Web-based software application designed
to allow groups with collections of information to share that information with others
via the World Wide Web and integrate that information into the NSDL (University of
Wisconsin - Madison, 2004). CWIS is a software application to assemble, organize,
and share collections of data about resources, similar to the Yahoo and Google online
directories, but conforming to international and academic standards for metadata.
CWIS was funded by the National Science Foundation (NSF) and specifically cre-
ated to help build collections of STEM resources and connect them into the NSF’s
NSDL, but it can be (and is being) used for a wide variety of other purposes.
The ScienceMaps version of CWIS has all the features of the standard application
including:

• Resource annotations and ratings

• Keyword searching
• Field searching
• Recommender system (similar to the “Amazon” recommender system)
• OAI 2.0 export (a method to allow a “harvester” such as the NSDL, to collect
metadata regarding content from the resource portal)
• RSS feed support (a method to “push” a short description of resource portal
content and a link to the full version of the content to other Web sites)
• Customizable user interface themes (e.g., large-text version)

Figure 1 is a screenshot of the ScienceMaps online resource portal home page.

Users may perform a keyword search to find lessons that pertain to a particular
topic (e.g., earthquakes, water, hurricanes, invasive species, etc.) or by California
Earth science, biological/life science, or investigation/experimentation standard
(e.g., 9a, 6b, 1c, etc.). It is possible to view the entire lesson, including objectives,
questions, and so forth, by clicking on “Full Record”. It is not necessary to login
to access the lessons.
The ScienceMaps online resource portal contains two integrated resources, sci-
ence lessons and GIS applications. Each science lesson is aligned to California
science standards to promote inquiry-based learning as well as investigation and
experimentation. Each GIS application is specifically designed and developed to
meet the learning objectives contained in the science standards and lessons in such
a way as to enhance learning.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Figure 1. ScienceMaps online resource portal — home page

To more effectively meet the requirements for the distribution of ScienceMaps les-
sons and GIS applications, some of the features of the standard portal application
have been customized, and others added, specifically:

•. Description: background information regarding the lesson topic

•. Objectives:.specific objectives of the lesson
•. Lesson.outline:.specific procedures to follow to complete the lesson
• Classification: specific content area (e.g., Earth Sciences)
•. Resource.type: type of online resource (e.g., Interactive GIS)
•. Audience:.appropriate grade level (e.g., Grades 10 – 12)
•. Coverage:.geographic region that the lesson covers (e.g., California, United
States)
•. California.science.standards:.specific content standards that the lesson cov-
ers

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

•. California.investigation.and.experimentation.standards: specific standards

that the lesson covers

Figure 2 is a screenshot of “Seismic Hazard Map for the United States — Earth
Science Lesson 16” illustrating the features described above and the content for
this particular lesson. In this lesson, students are asked to analyze the relationship
between the occurrences of seismic hazards and topography in California and the
United States. This lesson begins with background on seismic hazards and how the
dataset used in this lesson was derived. Details about the origins of the map as well
as a glossary of terms are provided to the user via links. Information on the United
States Geological Survey is also available. The objectives that each student is ex-
pected to master upon completion of the lesson follow along with the procedures
and questions. It is possible for teachers to print the instructions and questions for
students if they prefer hard copies. Each lesson concludes with the coverage area,
grade level, and content and investigation standards assessed in the lesson.
Figure 3 is a screenshot of the corresponding GIS application “Seismic Hazard
Map for the United States — Earth Science Lesson 16” and illustrates basic user
interface features such as a legend and layer control and basic GIS operations such
as zoom in/out, pan, identify, and “zoom-to.” Advanced GIS operations such as
query and a spatial analysis function (proximity analysis) are also illustrated. In
this lesson, students begin by entering a state name, which allows them to focus on
one state. Using the legend, they are able to draw inferences regarding the seismic
hazard potential as measured by percent of gravity in the selected state. Students
are then able to perform queries by comparing other areas that have similar seismic
hazard characteristics in the United States by entering a specific percent of grav-
ity. They may also enter a specific address along with search radius, to determine
the type of seismic hazard in a particular area (e.g., their home, school, landmark,
etc.). In either of these queries, detailed seismic hazard information appears in the
box below the map. Many advanced spatial operations, such as routing, overlay
analysis, and buffering, can be added to meet the operational requirements of any
particular lesson.

ScienceMaps.Resource.Development.Methodology

The ScienceMaps Resource Development Methodology depicted in Figure 4 is being

utilized to support the development of the two ScienceMaps resources — science
lessons and GIS applications.
Starting with the Content Area Focus activity, the resource developer chooses one
of the four major California standards-based science content areas for instructional
materials development: biology/life sciences, chemistry, Earth sciences, and physics

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Figure 2. ScienceMaps online resource portal — Earth science lesson 8

(the investigation and experimentation standards are applicable to all standards-based

science content areas). This choice informs and influences both the science lesson
development and the GIS application development activities.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

Science lesson development involves the development of a lesson description includ-

ing background information regarding the lesson topic; the development of “lesson
objectives” including specific objectives of the lesson; and the development of a
lesson outline including specific procedures to follow to complete the lesson. It also
involves varying the level of questioning according to Bloom’s Taxonomy in order
to promote higher order thinking skills (analysis, application, and synthesis) as well
as providing all students with a level of success when utilizing the application.
GIS application development is further informed by the choice of the content area
focus during lesson development. In general, the GIS application development
activity involves four major tasks: data acquisition and manipulation, determina-
tion of GIS functionality required (basic and advanced operations as well as pos-
sibly advanced analysis functionality requiring special programming), application
implementation and testing, and data and application maintenance (based on user
requests and field-study feedback).

Figure 3. Earth science lesson 16 — GIS application

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Figure 4. ScienceMaps resource development methodology

Feedback

Science Lesson
Inform s Developm ent

Content Area Inform s Field-Studies

Focus

Inform s
GIS Application
Developm ent

Feedback

Assessment.and.Evaluation

There is a tremendous need for the development of a research-based, classroom-

tested set of instructional materials for teaching and learning with GIS in order
to plan software systems designed to better support student learning (Bednarz,
2004). In keeping with the goals of ScienceMaps, to support the ongoing develop-
ment and field-testing of this innovative courseware and to promote the integra-
tion of technology into the science curriculum, and to validate the efficacy of the
ScienceMaps research and development methodology, assessment and evaluation
will include quantitative and qualitative data collection and analysis. Field-studies
will provide feedback from the various ScienceMaps users (teachers and students)
to understand how well participant school districts are integrating GIS technology
into their classrooms using the resources available in the ScienceMaps portal. These
field-studies will include participant surveys and interviews to assess such system
usability issues as ease of navigation and search features of the online portal, as
well as how well students interact with the many GIS applications. This feedback
will not only inform future resource development, such as new functionality to be
added, but will also be used to update existing portal resources. Empirical analyses
of this data will provide findings regarding the effective use of technology in the
science classroom and will add to the knowledge base in this area.

Initial.Results

Data was collected from 17 undergraduate students enrolled in a teacher prepara-

tion program at a private university in Southern California. Students completed a
system usability scale. The system usability scale (SUS) is a simple, ten-item scale
giving a global view of subjective assessments of usability (Brooke, 1996). The SUS

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

was developed by the Digital Equipment Corporation and is a Likert scale, where
a statement is made and the respondent then indicates the degree of agreement or
disagreement with the statement on a 5-point scale. The selected statements address
a variety of aspects of system usability, such as the need for support, training, and
complexity, and thus have a high level of coverage for measuring the usability of
a system (B. Hilton, 2003). Once a respondent has used the system, they are asked
to immediately respond to each item prior to discussing their reactions with others.
The SUS provides a single number, between 0 and 100, which is a composite mea-
sure of the overall usability of the system being studied. SUS scores are calculated
as follows:

1. For questions 1, 3, 5, 7, and 9, the score contribution is the scale position minus
1.
2. For questions 2, 4, 6, 8 and 10, the contribution is 5 minus the scale posi-
tion.
3. Multiply the sum of these scores by 2.5 to obtain the overall value of System
Usability.

Additionally, respondents were asked three open-ended questions regarding their

impressions of ScienceMaps.
The mean overall SUS score for this group (N = 17) was 57.21. Table 1 illustrates
the descriptive statistics for the group. These statistics are comparable to other
studies for first-time use (American Institutes for Research, 2001; B. Hilton, 2003;
Musgrave & Ryssevik, 2000). These studies included N = 23 and Mean = 69.13
for Hilton; N = 72 and mean SUS score for the alpha version = 69 and N = 53 and
mean = 69.5 in the beta version for Musgrave & Ryssevik; and N = 22 with mean
SUS score for Office XP = 91.71 and 51.82 for Office 2000 for American Institutes
for Research.
Table 2 outlines the descriptive statistics for the individual SUS questions for the
group. The means of the positive questions (1, 3, 5, 7, and 9) ranged from 1.71 to
2.47 while the means of the negative questions (2, 4, 6, 8, and 10) ranged from 1.88
to 2.82. Standard deviations for all questions ranged from 0.81 to 1.17.

Table 1. Descriptive statistics for SUS score

Mean 57.21
Std. Deviation 15.68
Range 55.00
Minimum 35.00
Maximum 90.00

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Table 2. Descriptive statistics for individual SUS questions

SUS.Questions Mean Std..Dev

Q1: I think that I would like to use this system frequently. 1.71 1.05
Q2: I found the system unnecessarily complex. 2.41 1.06
Q3: I thought the system was easy to use. 2.06 1.09
Q4: I think that I would need the support of a technical person to be able to use the 2.53 1.07
system
Q5: I found the various functions in this system were well integrated. 2.29 0.85
Q6: I thought there was too much inconsistency in this system. 2.35 1.00
Q7: I would imagine that most pople would learn to use this system very quickly. 1.47 0.94
Q8: I found the system very cumbersome to use. 1.88 0.99
Q9: I felt confident using the system. 2.35 1.17
Q10: I needed to learn a lot of things before I could get going with this system. 2.82 0.81

Subjective responses indicated that generally users liked the visual representation of
the maps and thought it provided good background information. Many thought the
ability to enter an address, and to find specific information in relation (proximity)
to that address, was both useful and informative. Respondents believed that the site
addressed California science standards and would be a useful tool for instruction.
Suggestions for improvement included adding online help/training as well as mak-
ing minor improvements to the user interface. These suggestions will be addressed
and changes implemented in future versions of the system.

Conclusion

ScienceMaps will continue to utilize the latest cutting edge geospatial technologies
to enhance science instruction. The research findings of this effort will inform the
ongoing research on the effects of technology on science education. Given the focus
on classroom teachers, the resulting impact on K-12 students could be substantial.
ScienceMaps can help shape future resource portals, especially those using GIS so
that they become more effective instructional tools. The significance of Science-
Maps lies in its ability to address the issue of how to assist educators in developing
more in-depth science content knowledge while building a structured approach for
teaching science standards using GIS technology. Furthermore, the use of an Inter-
net-based GIS provides the ability to bring geospatial data to anyone with access
to an Internet browser. There is no need to run specific GIS software or download
data; the science lessons and data layers are already constructed and available for
immediate viewing and analysis through the various GIS applications.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

The future effort of ScienceMaps is to utilize both the knowledge, gained as well
as the lessons and applications developed, to provide the foundation for an initia-
tive that would be replicated nationwide. This would include customized lessons
and applications for individual states’ science standards, professional development
seminars, and evaluation of effectiveness. States with attributes similar to California,
large student and teacher populations, high capacity to use technology but low use
and access, would be selected first for this initiative.
There is much enthusiasm about ways that these technologies can help shrink
the digital divide. ScienceMaps focuses on the integration of technology into the
secondary science curriculum by emphasizing effective use while simultaneously
addressing accessibility issues and learning styles. Furthermore, it promotes the use
of innovative GIS technology in a discipline with which it is not usually associated.
While there are programs that exist to promote the use of GIS in education, they
require a high level of proficiency in GIS skills and knowledge, expensive software,
and a tremendous time commitment. ScienceMaps concentrates on using GIS to
teach, not on teaching GIS.

References

Adelman, N., Donnelly, M. B., Dove, T., Tiffany-Morales, J., Wayne, A., & Zucker,
A. (2002). The integrated studies of educational technology: Professional
development and teachers’ use of technology (Tech. Rep. No. SRI Project
P10474). Menlo Park, CA.
American Institutes for Research. (2001). Microsoft Office XP vs. Office 2000
comparison test public report (Tech. Rep. No. AIR Project No. 01674.001).
Washington, DC.
Archer, J. (1998). The link to higher scores. Education Week — Technology Counts,
1998 (p. 18).
Becker, H. J. (2000). Findings from the teaching, learning, and computing survey:
Is Larry Cuban right? Paper presented at the Council of Chief State School
Officers Annual Technology Leadership Conference, Washington, DC.
Bednarz, S. W. (2004). Geographic information systems: A tool to support geography
and environmental education? GeoJournal, 60(2), 191–199.
Borgman, C. L., Leazer, G. H., Gilliland-Swetland, A., Millwood, K., Champeny, L.,
Finley, J., et al. (2004). How geography professors select materials for class-
room lectures: Implications for the design of digital libraries. In Proceedings
of the 4th ACM/IEEE-CS Joint Conference on Digital Libraries (pp. 179-185).
Tuscon, AZ: ACM Press.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Brooke, J. (1996). SUS: A ‘quick and dirty’ usability scale. In I. McClelland (Ed.),
Usability evaluation in industry (pp. 189-194). London: Taylor & Francis
Ltd.
CEO Forum. (2001). Year 4 StaR Report, 2003. Retrieved from http://www.elec-
tronicschool.com/2001/09/0901ewire.html#forum
Drew, D. E. (1996). Aptitude revisited: Rethinking math and science education
for America’s next century. Baltimore, MD: The Johns Hopkins University
Press.
Du, D., Havard, B. C., Olinzock, A., & Yang, Y. (2004). The impact of technol-
ogy use on low-income and minority student academic achievements. Paper
presented at the American Educational Research Association 2004 Annual
Meeting, San Diego, CA.
Ely, D. P. (2002). Trends in educational technology (5th ed.). Syracuse, NY: ERIC
Clearinghouse on Education and Technology.
Farrell, B. A., & Kotrlik, J. W. (2003). Design and evaluation of a tool to assess
strategical information processing styles. Journal of Vocational Education
Research, 28(2), 141-160.
Fox, E. (2005). Technology counts, 2005: Tracking U.S. trends. Education Week,
24, 40-79.
Gaudet, C., Annulis, H., & Carr, J. (2001). Workforce development models for geo-
spatial technology. Hattiesburg, MS: The University of Southern Mississippi,
Geospatial Workforce Development Center.
Gross, P., Goodenough, U., Haack, S., Lerner, L., Schwartz, M., & Schwartz, R.
(2005). The state of state science standards. Washington, DC: Thomas B.
Fordham Foundation.
Gutierrez, M., Coulter, B., & Goodwin, D. (2002). Natural disasters workshop
integrating hands-on activities, Internet-based data, and GIS. Journal of Geo-
science Education, 50(4), 437-443.
Hanson, K., & Carlson, B. (2005). Effective access: Teachers’use of digital resources
in STEM teaching. Newton, MA: Gender, Diversities, and Technology Institute
at the Education Development Center.
Hedges, L. V., Konstantopoulos, S., & Thoreson, A. (2000). Designing studies to
measure the implementation and impact of technology in American schools.
Paper presented at the Conference on The Effectiveness of Educational Tech-
nology: Research Designs for the Next Decade, Menlo Park, CA.
Hilton, B. (2003). The impact of open source software and Web services on informa-
tion system development: An investigation of proximate computing. Claremont,
CA: School of Information Science, Claremont Graduate University.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Hlton & Drew

Hilton, J. (2003). The effects of technology on student science achievement. Clare-

mont, CA: School of Educational Studies, Claremont Graduate University.
Hilton, J. (2005). Narrowing the digital divide: Technology integration in a high-
poverty school. In D. Carbonara (Ed.), Technology literacy applications in
learning environments (pp. 385). Hershey, PA: Idea Group Publishing.
Hilton, J. (2006). The effect of technology on student science achievement. In E. M.
Alkhalifa (Ed.), Cognitively informed systems: Utilizing practical approaches
to enrich information presentation and transfer (pp. 346). Hershey, PA: Idea
Group Publishing.
Houtsonen, L., Kankaanrinta, I.-K., & Rehunen, A. (2004). Web use in geographi-
cal and environmental education: An international survey at the primary and
secondary level. GeoJournal, 60(2), 165 - 174.
Jerald, C. D. (1998). By the numbers. Education Week — Technology Counts (Vol.
18).
Joseph, E. (2004). Community GIS: University collaboration and outreach with
K-12 teachers. Paper presented at the ESRI 2004 Users Conference, San
Diego, CA.
Levin, D., Arafeh, S., Lenhart, A., & Rainie, L. (2002). The digital disconnect: The
widening gap between Internet savvy students and their schools. Washington,
DC: The Pew Internet & American Life Project.
Levinson, E. (2000). Technology and accountability: A chicken-and-egg question.
Converge, 3(11), 58-59.
Manzo, K. K. (2001). Academic record. Education Week — Technology Counts
(Vol. 20, pp. 22-23).
Marsh, T., Wong, W. L., Carriazo, E., Nocera, L., Yang, K., Varma, A., et al. (2005).
User experiences and lessons learned from developing and implementing
an immersive game for the science classroom. Paper presented at the HCI
International 2005 — 11th International Conference on Human-Computer
Interaction, Las Vegas, NV.
Musgrave, S., & Ryssevik, J. (2000). NESSTAR final report. Brussels: Networked
Social Science Tools and Resources.
Parsad, B., & Jones, J. (2005). Internet access in U.S. public schools and classrooms:
1994–2003 (Rep. No. NCES 2005015). Washington, DC: National Center for
Education Statistics.
Quintana, C., & Zhang, M. (2004). IdeaKeeper notepads: Scaffolding digital library
information analysis in online inquiry. Paper presented at the Conference on
Human Factors in Computing Systems (CHI 2004).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 ScenceMaps 

Rumberger, R. W. (2000). A multi-level, longitudinal approach to evaluating the

effectiveness of educational technology. Paper presented at the Design Meeting
on Effectiveness of Educational Technology, Menlo Park, CA.
Schmitt, C. (2002). Technology in schools: Suggestions, tools, and guidelines for
assessing technology in elementary and secondary education (Rep. No. NCES
2003313). Washington, DC: National Center for Education Statistics.
Skinner, R. A. (2002). Tracking tech trends. Education Week — Technology Counts
(Vol. 22, pp. 53-56).
Trotter, A. (1999). Preparing teachers for the digital age. Education Week —-Tech-
nology Counts (Vol. 19, pp. 37-43).
U.S. Bureau of the Census (1998). Statistical Abstract of the United States: 1998
(118th ed.). Washington, DC: U.S. Government Printing Office.
University of Wisconsin - Madison. (2004). The Internet Scout Project, 2005. Re-
trieved from http://scout.wisc.edu/Projects/CWIS/index.php
Wetzel, D. R. (1999). A model for the successful implementation of instructional
technology in science education. Paper presented at the Mid-South Education
Research Association Annual Conference, Point Clear, AL.
Williams, R. S. (2002). Future of education = Technology + teachers. Washington,
DC: U.S. Department of Commerce Technology Administration.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Hlton & Drew

Section III

Decision-Making
Environments

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 153

Chapter VII

Flexible Spatial
Decision-Making
and Support:
Processes and Systems
Shan Gao, University of Auckland, New Zealand
David Sundaram, University of Auckland, New Zealand

Abstract

Spatial decision-making is a key aspect of human behaviour. Spatial decision sup-

port systems support spatial decision-making processes by integrating required
information, tools, models, and technology in a user-friendly manner. While current
spatial decision support systems fulfill their specific objectives, they fail to address
many of the requirements for effective spatial problem solving, as they are inflexible,
complex to use, and often domain-specific. This research blends together several
relevant disciplines to overcome the problems identified in various areas of spatial
decision support. We proposed a generic spatial decision-making process and a
domain-independent spatial decision support system (SDSS) framework and archi-
tecture to support the process. We also developed a flexible SDSS to demonstrate
an environment in which decision-makers can utilize various tools and explore
different scenarios to derive a decision. The use of the system is demonstrated in a
number of real scenarios across location, allocation, routing, layout, and spatio-
temporal problems.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
154 Gao & Sundaram

Introduction

Decision-making is an essential element of our lives and critical for business suc-
cess, because many natural phenomena and socio-economic activities take place
in a spatial context. Spatial decision-making (SDM) becomes one of the important
aspects of human behaviour. SDM activities are either dependent or influenced by
geographical information. They are based on spatial problems that are normally
semi-structured or ill-defined. Spatial problems are also multi-dimensional as they
contain both spatial and non-spatial aspects. It is not easy to measure or model all
the aspects of a spatial problem in a single step. Therefore, a sophisticated model-
ling process is needed for solving these spatial problems.
Spatial decision-making usually involves a large number of alternative solutions, and
these alternatives need to be managed using an appropriate scenario management
facility. The multi-criteria decision-making (MCDM) method helps the decision-
maker to select a solution from the many competitive alternatives. It facilitates the
evaluation and ranking of the alternative solutions based on the decision- maker’s
knowledge or preference with respect to a set of evaluation criteria.
Decision support systems (DSS) have been proposed to support decision-making.
Silver (1991) broadly defines a decision support system as a computer-based infor-
mation system that affects or is intended to affect the way people make decisions. A
focused DSS definition given by Sprague and Carlson (1982) states that a decision
support system is an interactive computer-based system that helps decision-makers
to solve unstructured problems by utilising data and models. To support spatial deci-
sion-making, a variety of systems have been developed: These include geographic
information systems (GIS) and spatial decision support systems (SDSS). Our focus
is on SDSS in this research. Peterson (1998) defines a spatial decision support
system as an interactive and computer-based system designed to support a user or
a group of users in achieving higher effectiveness for solving semi-structured or
non-structured spatial decision problems.
As technology progresses, there is increasing opportunity to use SDSS in a variety
of domains. Flexible support of decision-making processes to solve complex, semi-
structured or unstructured spatial problems can offer advantages to individuals and
organisations. We synthesise ideas, frameworks, and architectures from GIS, DSS, and
SDSS. In addition, concepts from spatial modelling, model life cycle management,
scenario life cycle management, knowledge management, and MCDM methodol-
ogy are explored and leveraged in the implementation of a flexible spatial decision
support system using object-oriented methodology and technology.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 155

Background

Moloney, Lea, and Kowalchek (1993) observe that about ninety percent of business
information are geographically related and cover diverse domains, for example,
resource management, environmental modelling, transportation planning, and geo-
marketing. Spatial problems are normally categorised into allocation, location, routing,
and layout problems based on their geographical features. The primary goal of SDSS
is to support decision-making activities using its flexible modelling capabilities and
spatial data manipulation functions. SDSS encompass spatial analytical techniques
and enable system output in a variety of spatial forms. The characteristics of the
spatial data, models, and operations as well as the integration processes of non-spatial
systems with spatial systems make SDSS more complex. Currently- available SDSS
frameworks and architectures are suitable for their specific objectives. However,
they fail to properly address many of the requirements of a generic, flexible, and
easy-to-use SDSS. As we have noted earlier, incorporating a GIS with a DSS de-
velops some of SDSS frameworks and architectures. However, the existing SDSS
frameworks and architectures are neither comprised of all the DSS components, nor
are they generic. Fedra’s (1995) framework is good for analytical modelling, but
the solver is tightly integrated within the model, which does not provide flexibility
in using a solver with different models. It does not provide any mapping instrument
for flexible integration of model and data; rather it uses a pre-customised integration
system that is limited to a specific domain. The model management framework of
Yeh and Qiao (1999) supports the modelling life cycle but overlooks spatial pre-
sentation functionalities. The dynamic environmental effects model (DEEM) is a
software framework that provides optimal interoperability among environmental
models, supports flexible decision-making, but fails to address scenario development
and run-time model generation. The SDSS framework proposed by Armstrong and
Densham (1990) does not separate model and solver. The knowledge management
system framework of Mennecke (1997) does not pay attention to spatial visualisa-
tions, but rather focuses on the queries and reports. Some implemented SDSS are
good in modelling systems, for example, Illinois River Decision Support System
(ILRDSS) and Environmental Decision Support System (EDSS). Some are good
in spatial data analysis and visualisation, for example, GRASSLAND, but none of
these systems is generic and domain-independent.
The scenario-based SDSS architecture of Hall, Bowerman, and Feick (1997) is the
only one that supports scenario management and facilitates the multiple scenario
development. The system allows the decision-makers to generate, save, and recall
solutions and supports comparing different scenarios. It is well developed for solv-
ing domain specific spatial problems, but the scenario evaluation has not been fully
explored in this system as a non-spatial model; for example, the evaluation model
has been absent in this framework. None of these frameworks and architectures
addresses multi-criteria decision-making.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
156 Gao & Sundaram

Issues, Controversies, Problems

Though significant progress has been made in the context of decision-making and
decision support systems, there has not been sufficient emphasis on SDM nor on
SDSS. Densham (1991) argues that the main objective of a SDSS is to support deci-
sion-makers to make well-informed decisions based on complex spatial tasks. The
processes and the various mechanisms used for solving spatial problems are also
complex. The available technologies related to spatial decision-making are neither
quite sufficient nor easy-to-use for managing this complexity.
Decision-makers often perceive the decision-making process adopted to solve
complex multi-dimensional spatial problems as unsatisfactory. Decision-makers
have been using the decision-making frameworks and processes for many years, but
the general approaches proposed by Simon (1960) and others were not particularly
developed for solving spatial problems; rather they provide a guideline for devel-
opment of decision-making processes. Though Malczewski (1999) has proposed a
multi-criteria decision-making framework, the implementation of the process has
not been fully explored in the spatial context. A generic process to guide decision-
makers to solve spatial problems is lacking. Decision-makers have to rely on their
own processes and experience for spatial decision-making.
On the other hand, existing GIS, DSS, and SDSS that support decision-makers
have their limitations in solving spatial problems. GIS do well in managing spatial
data, but lack flexible modelling capacity. DSS are typically used in the non-spatial
domain. SDSS encompass analytical techniques inherited from DSS and spatial
modelling and various spatial input and output mechanisms provided by GIS to
support decision-makers to make well-informed decisions. Densham (1991) argues
that SDSS should facilitate a number of functions such as spatial data input, model-
based analyses, and powerful visual presentations. Furthermore, these systems
generally do not facilitate flexible spatial data manipulation mechanisms and lack
output facility for presentation of the result of the decision-making process. They
do not provide knowledge storage facility and lack a process-oriented user interface,
thus enlarging the complexity of using the systems. A review of SDSS frameworks
and architectures led us to conclude that current approaches fulfill their specific
objectives, but fail to address many of the requirements of a generic, flexible, and
easy-to-use SDSS.
In addition, model and scenario management processes have not been well developed
in SDSS. The modelling process is ad hoc rather than generic and does not address
the need of separation and integration of spatial and non-spatial models. Research
indicates that the difficulties in synthesising various decision alternatives are primary
obstacles to spatial problem-solving (Ascough II et al., 2002). Some SDSS support
the development of single spatial or non-spatial scenario at one time, but few sys-
tems support integration of spatial and non-spatial scenarios to develop numerous

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 157

multi-attribute spatial scenarios simultaneously. And also, spatial problems often

have numerous alternative solutions, and use multiple criteria upon which they are
evaluated. These complex spatial problems can be solved efficiently by incorporat-
ing the analytic modelling capabilities of the application specific spatial decision
support system. The multi-criteria decision-making helps in screening alternatives
and identifying the best solution. At this point many strategic requirements of SDSS
have not been implemented completely. Their capability to solve complex multi-
dimensional spatial problems is very limited.
Based on the problem areas identified in the field of spatial decision-making, we
believe that a generic process for solving complex spatial problems is needed for
achieving high effectiveness in spatial decision-making. The ideas from related
disciplines (i.e., spatial modelling, model and scenario life cycle management,
knowledge management and MCDM) need to be addressed in this process. Current
SDSS frameworks and architectures address or implement one or a combination
of requirements that are needed for complex spatial problem-solving, but they are
incapable of managing all the issues required in a single coherent way. Synthesis-
ing ideas, frameworks, and architectures from related disciplines helps to overcome
some of the problems identified in spatial decision support systems. A generic
spatial decision-making process and a domain-independent SDSS framework and
architecture will improve spatial decision quality and provide more opportunities
for the use of SDSS.

Solutions and Recommendations

To address the issues of spatial decision-making and spatial decision support systems,
we first propose a spatial decision-making process that allows the decision makers
to better use SDSS to solve spatial problems in a generic way, thus resulting in a
better decision. We then develop a flexible spatial decision support system (FSDSS)
framework and architecture to support this spatial decision-making process. The
FSDSS provides flexible spatial data manipulation facility and supports integra-
tion of spatial and non-spatial data; it also facilitates the entire model and scenario
management life cycle process, as well as support for multi-criteria spatial problem
solving. We further implement a prototypical FSDSS that acts as a proof-of-concept
for these proposals.

Spatial Decision-Making Process

Malczewski (1997) identifies complexity, alternatives, and multi-criteria charac-

teristics as key features of spatial problems. Spatial problems are complex because

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
158 Gao & Sundaram

they are semi-structured or ill-defined in the sense that the goals and objectives are
not completely defined. Spatial problems are multi-dimensional and often related to
non-spatial information. Each spatial problem can have a large number of decision

Figure 1. Spatial decision-making process

Identification
Problem Identification

Problem

Intelligence
Creation of Model
for spatial and non-spatial aspects

Modelling
Problem
Grouping non-spatial aspects Layering of spatial aspects

Instantiation of non-spatial Instantiation of spatial models

Instantiation
models for a single aspect in a single layer with
with appropriate data appropriate data/ layer/ theme

Model
Need Need
Transformation Transformation
? ?
No No
Yes Yes

Execution
Model
Execution of Execution of
non-spatial models spatial models
with appropriate solver with appropriate solver

Design
Instantiation Scenario Modelling
Model Integration

Integration of Integration of
non-spatial aspects spatial layers

Integration results of spatial and non-spatial aspects

(Modelling of Scenario)

Instantiation of scenarios (multiple instances)

Scenario

Need
Transformation
Evaluation Execution
Scenario

? No

Yes
Apply Decision
Makers Preference
Execution of scenarios Scenario
Scenario

Comparison/
Build MCDM
Choice

Scenario evaluation and

sensitivity analysis
Decision
Making

Decision-making

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 159

alternative solutions. These alternative solutions to the spatial decision problems

are normally characterised by multiple criteria upon which they are judged.
Model building is critical in decision-making as it clarifies thinking and improves
the decision-making quality (Georgantzas & Acar, 1995). Spatial modelling tech-
niques are used for finding relationships among geographic features and helps de-
cision-makers to address the spatial problem clearly and logically. Spatial aspects
and non-spatial aspects can be coexistent in a spatial problem, so that we need to
consider both aspects at the same time. It is difficult to model a complex spatial
problem in a single step, but it is possible to model one aspect of a complex prob-
lem at a time. For example, create a spatial model to deal with spatial aspects and
a non-spatial model that caters to the non-spatial aspects of the problem, and then
integrate them.
A spatial model contains spatial parameters that refer to the geographical features
of a spatial problem. Vector-based spatial data can be categorised into three major
groups, that is, spatial objects, spatial layers, and spatial themes. A spatial object
represents a single spatial item, for example, a point, a line, or a polygon. A spatial
layer contains a collection of spatial objects similar in nature, and every spatial
object belongs to a certain layer. A spatial theme comprises a number of spatial
objects and/or spatial layers that represent a particular meaning to a particular
spatial problem. Vector data is linked to non-spatial domain data through a spatial
reference system; a point could be associated with a residential location, and a line
may represent a running path. Each aspect of a spatial problem can be modelled
in one layer. These layers are then integrated into a complex model that represents
the many facets of the problem.
We propose a spatial decision-making process (Figure 1) by synthesising ideas of
decision-making processes (Simon, 1960) and the multi-criteria decision-making
process (Malczewski, 1999). It also integrates concepts from spatial modelling,
model, scenario, and knowledge management. The process contains nine specific
steps, namely, problem identification, problem modelling, model instantiation, model
execution, model integration, scenario instantiation, scenario execution, scenario
evaluation, and final decision-making.
The decision-making process begins with the recognition of a real-world problem
that involves searching the decision environment and identifying comprehensive
objectives that reflect all concerns relevant to a decision problem. The problem
is then put into a model by specifying the relevant attributes and behaviours. The
parameters in a model structure are instantiated with appropriate data. Decision-
makers select a solver for execution of a model instance and generate a result, that
is, the scenario. A scenario improves cognition by organising many different bits
of information (De Gues, 1997). Multiple scenarios are needed to explore different
ways of seeing problems and enhancing the decision-making quality (Forrester,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
160 Gao & Sundaram

1992). The scenario evaluation process evaluates many competitive alternatives

simultaneously and helps to identify the best solution.
The process is iterative in nature so that multiple scenarios instances can be gener-
ated using the same scenario structures. The scenario integration process enables the
decision-maker to combine both spatial and non-spatial scenarios to create a com-
plex multi-criteria spatial scenario that addresses all the requirements of a complex
spatial problem. When required, the instantiated scenarios are called for execution
using different solvers. The execution of the scenario allows the decision-maker to
further develop a more desirable solution to a particular problem. Scenario evalu-
ation ranks the many alternative scenarios based on decision-makers’ preferences.
Sensitivity analysis is employed as a means for achieving a deeper understanding
of the structure of the problem by changing the inputs, for example, data, solver,
or evaluation model. This helps decision-makers learn how the various decision
elements interact and allows them to determine the best solution. In completing the
above processes, the best-evaluated scenario is selected. As there is no restriction
on how the user chooses to solve a problem, decision-makers can select the phases
to follow based on the nature of the specific problem and their specific purposes.

The FSDSS Framework

We propose a flexible spatial decision support system (FSDSS) framework (Figure

2) to support the decision-making process and overcome the problems identified
earlier. The FSDSS framework is comprised of six major DSS objects or components
namely, data, models, solvers, visualisations, scenario, and knowledge. These objects
are stored in the object repository independently, and they communicate through
the kernel, the programmatic engine that makes the system run. The framework ac-
commodates spatial data (spatial objects, layers, and themes) and non-spatial data. It
contains both spatial and non-spatial models, solvers, scenarios, and visualisations.
The knowledge is the output of the decision-making process and can be stored in the
system for future reference. The decision-maker interacts with the system through
the user interface. Different data, model, and solver can be selected from the object
repository and mapped together to generate a scenario, or a specific decision support
system that is tailored for a particular problem domain.
This framework allows generating multiple scenarios at one time and stores them
in the scenario pool. The framework supports the integration of several simple
scenarios into a complex multi-attribute scenario that contains both spatial and
non-spatial aspects through the scenario integration process. The multiple scenarios
and the evaluated scenarios can be presented using the appropriate visualisation
component. The output of the decision-making process can be saved in the object
repository as knowledge.
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 161

Figure 2. The FSDSS framework

User Interface

Base Objects
Data/Layer Kernel Scenario
/Theme
Mapping
Visualisation
Model
Validation
Knowledge
Solver

Spatial and Non-Spatial Object Repository

Object, Layer Model Solver
&Theme Pool Pool Pool
Visualisation Scenario Knowledge
Pool Pool Pool

The FSDSS Architecture

We propose an architecture that implements the FSDSS framework and supports the
proposed decision-making process, as shown in Figure 3. The FSDSS architectural
components are organised into five distinct layers; these are persistence layer, object
services layer, DSS objects layer, integration layer, and presentation layer. These
layers and their components are briefly described as follows.
The persistence layer contains the object library used to store the system objects.
This includes the storage of non-spatial data and a variety of spatial data (objects,
layers, themes, and map). It is also responsible for the storage of models, solvers,
visualisations, scenarios, and knowledge, either spatial or non-spatial in nature,
using the object-oriented database management system.
The object services layer manages the system objects through the component control
that contains several parameters and methods to coordinate the component pool and
the application. It exports objects from the object library to the DSS objects layer,
as well as importing the resulting scenarios and knowledge from the DSS objects
layer back to the object library. It also facilitates dynamic creation, updating, and
deletion of the objects.
The DSS objects layer supports independent development and use of the decision-
support components including spatial and non-spatial data, models, solvers, and
visualisations, for generating basic spatial and non-spatial scenarios. It is responsible
for integrating scenarios to develop complex spatial scenarios. It supports the evalu-
ation and ranking of multiple scenario instances using the evaluation model. This

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
162 Gao & Sundaram

layer also facilitates the storage and reuse of the result from the decision-making
process. It also provides graphical and map-based presentation of data, scenarios,
and knowledge. The data component includes non-spatial and spatial data, that is,
spatial objects, layers, themes, and maps. The model can be of the primitive type or
the compound type (Geoffrion, 1987). Primitive type model parameters are directly
derived using base data type variables or executed model values of the base models.
The compound type parameters inherit and/or aggregate the base models as well as
adding some other user-defined parameters. The non-spatial model handles non-
spatial problems or non-spatial aspects of a spatial problem. Spatial models cater to
spatial problems. The evaluation model is made of different parameters as well as the
weights for each of these model parameters. The FSDSS architecture has a number
of spatial-oriented solvers and generalised solvers that can be used to solve/execute
both spatial and non-spatial models. The scenario combines data, model, solver, and
other relevant information. The scenario structure and its multiple instances can be
stored in the database. The FSDSS supports three types of visualisation: spatial,
non-spatial, and map-based. Spatial visualisation is used to represent spatial data,
scenarios, and knowledge. Non-spatial visualisation, for example, 3D graphs, are
used to present the output of analytical results. In addition to the general graphical
report functions, the FSDSS visualisation is particularly important when used with
maps. Different spatial objects, layers, or themes are overlaid to generate a new
map. The knowledge component contains the final results of the decision-making
process, including information about the decision-maker, the rules that were applied,
alternative scenarios, the final decision, as well as the system components used in
reaching a particular decision.
The integration layer contains the communication components, that is, kernel, map-
ping, and validation components. In addition to activating and using the component
functions, the kernel works as a user interface and is responsible for the communica-
tion and integration of system components. Mapping enables the model component
to communicate with data and solver components properly through model-data
and model-solver mapping processes. The model parameter or attributes are fixed;
the user selects the data attributes for model-data mapping and selects the solver
name and solver attributes for model-solver mapping. Validation enables proper
communication between system components. The validation module is responsible
for checking the input data type to the model and to the solver during the mapping
process. The model-data validation tests whether the data type of the model attri-
butes is similar or convertible to the data attributes, while model-solver validation
checks whether the data types of the attributes of the model instance are similar or
convertible to the data type of the solver attributes.
The presentation layer or user interface supports all the interactions between users
and the system. It provides a flexible environment where spatial and non-spatial
components are used together to solve complex spatial problems. The architecture

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 163

Figure 3. The FSDSS architecture

User Interface and Presentation Layer

Kernel Mapping Validation

Integration Layer

DSS Objects Layer

OBJECT SERVICES LAY

Scenario Manager Knowledge

Base Objects
Component Control

Data/Layer/Theme
Scenario
Knowledge Storage
Model t
Evaluation Knowledge Retrieval
Solver (MCDM)

Visualization
r

Spatial Objects Persistence Layer Non-Spatial Objects

Object Library
Spatial Data/Layer/Theme Spatial Model Non-Spatial Data Non-Spatial Solver
Scenario Knowledge
Spatial Visualization Spatial Solver Spatial and Non-Spatial Spatial and Non-Spatial Non-Spatial Model Non-Spatial Visualization

as a whole is technology independent so that it can be implemented using com-

monly-available platforms.
A simple decision-making flow in Figure 3 illustrates how the FSDSS architecture
supports the decision-making process. The decision-maker initiates the decision-
making process at the interface layer and interacts with the system through the ker-
nel. The component control picks up the relevant components from the persistence
layer. The selected data, models, and solvers are combined in the integration layer
to develop scenarios using the mapping component. The scenario manager manages
these scenarios, and the evaluated scenarios can be presented using the appropriate
visualisation component. The interaction between the DSS objects layer and the
persistence layer are bi-directional. On the one hand, the architecture allows flexible
selection of objects from the object library. On the other hand, the executed results
(e.g., scenarios generated) can be stored back to the object library.

The FSDSS Implementation

A prototypical FSDSS was implemented to prove the validity of the proposed spatial
decision-making process as well as the FSDSS framework and architecture. Object-
oriented concepts, object-oriented database management system, and the object-
oriented programming language were the tools and technologies used to develop
the FSDSS prototype. Jade (www.jadeworld.com), a fully integrated development
environment (Post, 2000) with its own object-oriented database and programming
language, was selected as the implementation platform. The complete prototype

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
164 Gao & Sundaram

Table 1. Spatial problems and implementation domains

Spatial Problem Application Domain Example Spatial Problems

Allocation Geo-Marketing Find geographical distributions

Layout Running Design and select best running path

Routing Delivery Identify the fastest route

Location Housing Search the most suitable house

Spatio-Temporal Health Trace the spread of a disease over space and time

was developed within Jade without recourse to any other tool. The proposed spatial
decision-making process and the implemented FSDSS were evaluated through five
scenarios across different spatial decision problem domains including location, al-
location, routing, and/or layout. Table 1 details of the type of spatial problems and
the specific domains where we tested the prototype. The same environment was
used in the testing, but with different data, model, and solver sets.

Sample Sessions with the FSDSS

In the following sections, we explore the interaction with the FSDSS to solve three
of the spatial problems mentioned above. We first use the problem of the design of
a running track to introduce some of the core spatial modelling concepts such as
objects, layers, and themes. The second example explores in detail the use of the
FSDSS, following the proposed spatial decision-making process, in the context of
the purchase of a house. To highlight the generic applicability of the system we show
its use in the context of solving a spatio-temporal problem in health care.

Design of a Running Track

The first scenario we explore is the design of a path for a particular running event
that takes into consideration the required amenities like drink points and toilets, as
well as the distance. This example introduces some of the basic spatial concepts
and functionality of the system. The decision-makers can plot a number of paths by
clicking points on the map. These paths are then saved in the database as running
path 1 and running path 2 (see Figure 4).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 165

Figure 4. The two alternative paths for running event

Figure 5. Spatial layer manager

In addition to these two paths, the decision-makers need to consider other facilities
that might be useful to this event, for example, drink points and toilets. These ob-
jects are stored in the database as individual objects or as a layer. The spatial layer
manager manages these layers. As we can see from Figure 5, there are four drink
points in the drink point layer.
Figure 6 presents the map layout of the two paths (running path 1, running path
2) and the relevant layers, that is, the four toilets in the toilet layer and four drink
points in the drink point layer.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
166 Gao & Sundaram

Figure 6. Spatial layers (toilet and drink point layer)

Four drink
points

Irrelevant to Four toilets

this event

It also shows that some of the facilities are not useful for the event, for example,
toilet 4, as it is far away from these paths. Further observation indicates that toilet
1, toilet 2, drink point 1, and drink point 2 are particularly useful for running path
1, while toilet 3, drink point 3, and drink point 4 are useful to running path 2. Based
on these observations, the system enables the decision-maker to further group these
objects or layers into spatial themes.

Figure 7. Implementation of spatial modelling

Running Theme 1 (Integrated

Map)

Toilet 1 and Toilet 2 in

Toilet Layer

Drink Point 1 and Drink Point 2

in
Drink Point Layer

Running Path 1

Area Map

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 167

A spatial theme contains a group of spatial objects and/or spatial layers and presents
a particular meaning to the decision-maker. In this illustration, running path 1, toilet
1, toilet 2, drink point 1, and drink point 2 are grouped together to form a theme,
that is, running theme 1. Similarly, running path 2 along with toilet 3, drink point
3, and drink point 4 forms running theme 2. The spatial objects and layers used for
creation of running theme 1 are presented in Figure 7. Integration of these spatial
items from different layers through map overlay allows viewing the whole picture
of a spatial problem from different perspectives.
The spatial theme manager provided by the FSDSS manages the spatial themes. It
allows the decision-makers to save themes to the database and retrieve themes from
the database. The interface for the spatial theme manager is given in Figure 8.
The spatial theme manager facilitates the creation and deletion of themes as required.
In addition, it facilitates the updating of themes, for example, add new objects or
layers to the theme, or delete spatial objects or spatial layers from the theme. Figure
9 shows the map presentation of running theme 1 and running theme 2.
Decision makers can also perform other tasks using appropriate models and solvers
to further investigate the problem, for example, calculate the distance of each of these
running paths using the path model and distance solver. The following illustration
shows two different scenarios for the above-mentioned themes. These scenarios are
compared with each other to decide the best path for this running event.

Figure 8. Spatial theme manager

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
168 Gao & Sundaram

In this case, running path 1 is better than running path 2 for two reasons. Firstly,
running path 1 has additional facilities (one more toilet) along its path compared
with running path 2. Secondly and most importantly, the geographical layout of
running path 1 is better than running path 2 since all available facilities are more
evenly distributed along the track.

Purchase of a House

This section illustrates the implemented FSDSS to solve a location problem using
the proposed spatial decision-making process. The application of each step of the
process shown in Figure 1 is described in detail.

Step 1: Problem Identification

The problem presented in this session is to identify the optimal location of a prop-
erty that maximises “return,” that is, the satisfaction level that is measured on the

Figure 9. Spatial themes (Running theme 1 and running theme 2)

Running
Theme 1

Running
Theme 2

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 169

Figure 10. Value tree of location problem

Number of Rooms

Functions Has Garage

Non-Spatial Price …
…
House Location Problem School Zone
Environment
…
Spatial Criteria
Distance to School

Distance to Main Facilities Distance to Hospital

Distance to Public
… Transportation

Problem Problem Simple Spatial and Complex Spatial and Non-

Identification Identification Non-spatial Scenario spatial Scenario Development
Development

basis of the following three criteria. The value tree of after-analysis of the problem
is presented in Figure 10.

• Quality criteria: for example, construction material, year built, size, function,
and number of rooms
• Economic criteria: such as market price or rental cost
• Location: for example, property accessibility, vicinity, and environmental
conditions

Some of these factors are difficult to evaluate or predict, as relative impacts for some
of these factors on return remain unknown. It is hard to structure the problem in its
entirety at one time, that is, precisely define and measure the objective for every
possible solution. In the next step, the decision-maker models this problem using
the proposed modelling approach by separating the spatial and non-spatial aspects
of a complex spatial problem.

Step 2: Problem Modelling

The problem modelling involves both spatial and non-spatial aspects. Quality and
economic factors are non-spatial in nature whereas accessibility criteria are of a spa-
tial nature. On the non-spatial side, cost and quality of the property can be analysed
using non-spatial models and solvers. The spatial aspect of the problem focuses on
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
170 Gao & Sundaram

the location of the property, as it is an important criterion when people rent or buy
a house. Location is a complex criterion that has multiple spatial dimensions, for
example, environment and distance to important facilities. These spatial dimensions
need to be analysed one by one in order to find a best location.
In this illustration, the decision-maker broadly selects a target area and then carries
out accessibility analysis. The analysis involves both the non-spatial and spatial
models, and it uses both non-spatial and spatial solvers. The problem is solved
iteratively by first, considering spatial and non-spatial data, models, solvers, and
scenarios; second, applying spatial and non-spatial criteria; and last, using goal-
seeking and sensitivity analysis.

Step 3 and 4: Scenario Development

The decision-maker now needs to load relevant decision-making components. These

include the property table and relevant map in which the properties are located,
the various models, solvers, and visualisations to be used for building the different
scenarios. A simple non-spatial scenario and a simple spatial scenario are developed
separately at first; they are then integrated into a combined scenario. These scenarios
are then transformed into a complex multi-criteria scenario through a structural
integration process. The scenario development process is illustrated as follows:

Simple Non-Spatial Scenario

The non-spatial scenario is created using the non-spatial filtering model and the
range solver. In this example, we have specified that we need a 3-bedroom flat with
a price range between $300,000 and $400,000. Several properties are identified
through this filtering process (see Figure 11). These are then stored in the database
as scenario 1 (4 instances).

Simple Spatial Scenario

The decision-maker has selected a buffer zone (a 500-meter radius circle) around a
particular location (e.g., x, y coordinates: 200,200). The filtering model is instantiated
with the property data and executed using the distance solver to find the properties
within the defined circle. This process develops many scenario instances as shown
in Figure 12. These scenario instances are then stored in the database as scenario
2 (14 instances).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 171

Figure 11. Simple non-spatial scenario creation

non-spatial filtering
criteria:
Price: 300k to 400k
House Style: flat
Room Numbers: 3

Filtered results:
4 properties

Saved as Scenario 1

Figure 12. Simple spatial scenario creation

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
172 Gao & Sundaram

Combined Scenario (Pipelining Integration)

Pipelining integration of spatial and non-spatial scenarios can be done in two ways.
The first way is to create non-spatial scenario 1 and then execute the geographical
filtering model using spatial solvers, for example, distance or point-in-polygon
solver (Figure 13).
During this integration process, the four non-spatial filtered scenario instances of
scenario1 as described earlier are supplied as input to the spatial filtering model.
The resulting scenario instances are stored as scenario 3 (3 instances).
The second way for integration of scenario 1 and scenario 2 is to supply the spatial
scenario 2 as input into the non-spatial filtering model and then apply the non-spatial
Range solver for execution, as illustrated in Figure 14. The process develops three
instances that are stored in the database as Scenario 3.
The scenario pipelining integration process can take place in many ways, either
from non-spatial to spatial or from spatial to non-spatial or from spatial to spatial
or from non-spatial to non-spatial. The flexible use and integration of spatial and
non-spatial models, solvers, and scenarios is one of the most important features of
the FSDSS. The above process helps the decision-maker to choose the properties

Figure 13. Integration of non-spatial with spatial scenarios

Non-Spatial Spatial Resulting

Scenario Model Scenario
Scenario 1 Scenario 3

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 173

Figure 14. Integration of spatial with non-spatial scenarios

Spatial Non-Spatial Resulting

Scenario Model Scenario
Scenario 2 Scenario 3

non-spatial filtering
criteria:
Price: 300k to 400k
House Style: flat
Number of Rooms: 3

load
Scenario 2

filtered using
non-spatial
criteria
3 Instances

output saved as
Scenario 3

that satisfy the non-spatial criteria, for example, quality, cost, and the basic location
requirements such as area. The following section illustrates another aspect of the
location problem, namely, accessibility analysis.

Complex Spatial Scenario (Structural Integration)

The complex spatial scenario is generated using the property data, distance model,
and distance solver as shown in Figure 15. The previously created scenario 3 and its
three instances are loaded from the scenario pool. Now, the decision-maker focuses
on distance to major facilities for accessibility analysis.
Distance has multiple dimensions. It includes the distance from a particular spatial
object (e.g., property 0014) to another object (e.g., hospital 2). The distance from
one object to a spatial layer (e.g., school layer) returns multiple values, in this case
the system returns the shortest distance from the target object to a single object
(e.g., school 1) in that layer.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
174 Gao & Sundaram

Figure 15. Multi-attribute spatial scenario creation

Step 5 and 6: Scenario Integration and Instantiation

The decision-maker integrates the simple combined scenario (scenario 3) structure

with these newly- developed distance parameters to develop a more complex sce-
nario that contains all the criteria for the problem.
The structural or permanent scenario integration takes place in two steps. First, a
bare scenario template is created as shown in Figure 16. Then multiple scenario
instances are created (Figure 17).
The scenario integration process is iterative in nature until all scenario instances
have been generated. The scenario template and its multiple instances are stored in
the database as a complex scenario, and they can be retrieved for further analysis or
evaluation. The distance to the schools and shops are calculated on a spatial layer,
rather than a single spatial object. The system picks up the distance to the closest
object in the layer for instantiation of the scenario parameter. The decision-maker

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 175

Figure 16. Scenario template for integration of spatial and non-spatial scenarios

Original New parameters

Figure 17. Multi-criteria spatial scenarios

can select any spatial object, layer, or theme for integration of scenarios using the
spatial manager as shown in Figure 18.

Step 7: Scenario Execution

Scenarios can be instantiated with the relevant data, model, and a number of solvers
can be applied for execution of the scenarios. The scenario can be executed in a simple
process or using multiple steps. The integration of executed models (scenarios) is
also the process of modelling the scenario. During the scenario execution process,
one scenario is instantiated and executed using different solvers.

Figure 18. Multiple spatial scenario generation

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
176 Gao & Sundaram

Step 8: Scenario Evaluation

The FSDSS supports the MCDM scenario evaluation process. The decision-maker
needs to build a MCDM evaluation model by specifying parameters and assign-
ing weights to each of these parameters. The evaluation model is instantiated with
alternative scenario instances. These scenarios are executed using the solver that is
tightly coupled within the evaluation model. The results are then ranked for selec-
tion. The sequence of the steps taken in this process is shown in Figure 19. The
decision-maker selects the scenarios for evaluation to the scenario table as indicated
in step 1. Then, an evaluation model is built by selecting the appropriate criteria
from the input scenario. In step 3, the decision-maker assigns a weight to each of
the criteria. Step 4 evaluates the scenarios using the model template created in step
2 and step 3. The built-in solver not only calculates values according to the formula
but also ranks these values. The highest value is given as 100%, and other scenarios
are calculated on a ratio basis by comparing the highest value.

Step 9: Decision-Making

As we can see from the results, property 0014 (Figure 19) is ranked highest. Fur-
thermore, the decision-maker can apply different evaluation models to explore
alternative scenarios by considering the uncertainty involved in the decision-mak-
ing process. Uncertainty may be caused by the error in available information to
the decision-maker, or improper judgment regarding the relative importance of
evaluation criteria.
Sensitivity analysis is employed as a means for achieving a deeper understanding
of the structure of the problem. Sensitivity analysis is done through changing data,
model, solver, scenario, and evaluation models. The decision-maker can change

Figure 19. Multi-criteria spatial scenarios

Step 1

Complex Scenario and its 3 instances

Step 2

Ranking results
Step 3

Step 4

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 177

any one of these aspects and then re-evaluate these scenarios. This process can be
repeated until all the scenarios relevant to the decision-maker are explored.

Trace the Spread of a Disease

The problem presented in this section is to identify the spread of the SARS epidemic
in Hong Kong. We try to understand, analyse, and manage geographically distributed
data using spatial functions. Overlaying quantitative graphics on a map enables the
viewer to realise potential information in an extremely clear manner so that spatial
patterns can be discovered.
We have aggregated a case database that includes the date, the number of cases
confirmed, and the places they were found. Through the geo-coding process, each
case was mapped to a geographical object, for example, a building or a hospital.
These spatial objects can then be referenced by the geo-coding locations on the map.
Figure 20 presents an overall picture of the geographical distribution of SARS cases
and the population density; this gives us information about the number of SARS
cases among different population groups at any given time. The visual representation
would be useful for developing hypotheses about relationship between the distribu-
tion of SARS and population density or groups (Chu, Gao, & Sundaram, 2005). It
could provide useful pointers for further analysis and investigations.

Figure 20. Case occurrences and population density (Source: Chu, Gao, & Sun-
daram, 2005)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
178 Gao & Sundaram

Figure 21. Time series presentation of SARS spread period 1

We explore the above scenario one step further by applying a time frame, to give
pictorial views of the rate of case increases in various hospitals over time. As we can
see from Figure 21, on the 13th of March, 2003, 33 SARS patients were admitted by
the Prince of Wales Hospital (PWH); among them, 18 were confirmed. There were
also two confirmed patients in Pamela Youde Easter Hospital (PYEH). The total
number of patients admitted in Hong Kong hospital was 35 (20 confirmed).
On the 20th of March, the total number of admissions increased to 99 (58 were
confirmed). The number of suspected patients in PWH jumped to 77, and PYEH
cases increased to ten. New suspected patients were also found in Princess Margaret
Hospital and Queen Elizabeth Hospital (Figure 22).

Figure 22. Time series presentation of SARS spread 2

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 179

Figure 23. Infected buildings, school, and others

Using the time series mapping, we were able to map the number of patients admit-
ted with suspected SARS symptoms and compare the number of these cases being
confirmed as SARS patients subsequently at later times. Similarly, we can use the
FSDSS to map the spread of SARS infections in the community. The system modules
are inter-connected, and new entry of data dynamically changes other modules. An
example of this is the dynamic refresh of collection of spatial objects where there
is an infection (as seen in the drop-down combo box that appeared in the left-hand
side in Figure 23). These spatial objects could include residential buildings, schools,
shops, and any other objects that have a spatial reference.
Visual overlay of spatial objects over time allows us to ask a question such as “why
has one block (in Lower Ngau Tau Kok Estate, quite far away from the Amoy
Gardens) a much higher rate of infection than any other building?” Demographic
analysis later reviewed that residents in this block comprised mostly senior citizens
and that they passed through the Amoy Garden shopping mall each day to do their
shopping.

Future Trends

Advances in MCDM, GIS, other relevant technologies, and their integration will
have major impacts on the future development of SDSS. Multi-criteria spatial de-
cision support is clearly the appropriate paradigm because it is such an adaptable
and comprehensive concept. Multi-criteria spatial analysis is moving to a more
exploratory, interactive emphasis with new decision analysis tools. There is also an

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
180 Gao & Sundaram

increasing emphasis on the study of complex systems through simulation (Ascough

II et al., 2002).
GIS itself is moving to a greater level of integration with other types of software.
Spatial data and tools will be ubiquitous, embedded in business systems, and trans-
parent to most users as shown in Figure 24. GIS can be an interoperable part of most
critical business information systems including ERP/enterprise systems (such as
SAP and PeopleSoft), environmental, asset, financial, human resource, emergency,
and customer relationship management systems, thus bringing the power of spatial
analysis and data mining to bear to measure results (business intelligence) and future
impacts (analytic intelligence). A major focus is the design and implementation of
true strategic intelligence frameworks that integrates spatial and non-spatial data
across different information systems and business units to create broader knowledge
and understanding within organisations at all levels of decision-making (Holland,
2005).
The design of an intelligent SDSS by coupling a GIS with an expert system is dis-
cussed by Zhu and Healey (1992). Spatial decision-making is based on two streams:
the quantitative, which includes data analysis and modelling, and qualitative, which
includes experience, intuition, judgment, and expertise. Current GIS systems focus

Figure 24. Historical and future trends of GIS (Source: Holland, 2005)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 181

on quantitative information and largely ignore the qualitative aspects of the process.
The inclusion of expert systems, which perform decision-making tasks by reasoning,
using rules defined by experts in the relevant domain, would greatly aid qualitative
spatial decision-making within GIS, and thus SDSS.

Conclusion

Decision-makers perceive the decision-making processes for solving complex

spatial problems as unsatisfactory and lacking in generality. Current SDSS fulfil
their specific objectives, but fail to address many of the requirements for effective
spatial problem-solving, as they are inflexible, complex to use, and often domain-
specific. This research blends together several relevant disciplines in a unique way
and attempts to overcome the problems identified in the fields of spatial decision-
making and SDSS.
We proposed a spatial decision-making process. Within the context of the spatial
decision-making process, we have proposed a modelling approach by addressing the
need of differentiating the spatial and non-spatial elements for multi-dimensional
complex problem modelling. We then developed a flexible spatial decision support
system framework and architecture to support this process. We also implemented a
prototypical FSDSS that acts as a proof-of-concept for the spatial decision-making
process, FSDSS framework, and architecture. The proposed spatial decision-making
process and the implementation of FSDSS have been evaluated through a number of
scenarios across diverse domains. The evaluation results indicate that the proposed
spatial decision-making process is generic, and it is effective in solving complex
spatial problems in different domains. Furthermore, the flexible use across domains
and different types of problems was due to the generic nature of the architecture and
design that leveraged spatial modelling and object-oriented concepts.

References

Armstrong, M. P., & Densham, P. J. (1990). Database organisation strategies for

spatial decision support systems. International Journal of Geographical In-
formation Systems, 4(1), 3-20.
Ascough II, J. C., Rector, H. D., Hoag, D. L., McMaster, G. S., Vandenberg, B. C.,
Shaffer, M. J., et al. (2002). Multi-criteria spatial decision support systems:
Overview, applications, and future research directions. In Online Proceed-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
182 Gao & Sundaram

ings of the iEMSs (Vol. 3, pp. 175-180). Retrieved from http://www.iemss.

org/iemss2002/proceedings/pdf/volume%20tre/290_ascough%202.pdf
Chu, S., Gao, S., & Sundaram, D. (2005). A flexible multi-criteria spatial decision
support architecture: Design and evaluation by healthcare test cases. The
Journal on Information Technology in Healthcare, 2005, 3(1), 5–20.
De Gues, A. (1997). The living company: Habits for survival in a turbulent business
environment. Boston, MA: Harvard Business School Press.
Densham, P. J. (1991). Spatial decision support systems. In Maguire, D. J., Goodchild,
M. F., & Rhind, D. W. (Eds.): Geographical information systems: Principles
and applications (pp. 403-412). Burnt Mill, UK: Longman.
Fedra, K. (1995). Decision support for natural resources management: Models, GIS
and expert systems. AI Applications, 9(3), 3-19.
Forrester, J. W. (1992). Policies decisions and information sources for modeling.
European Journal of Operational Research, 59(1), 42-63.
Geoffrion, A. M. (1987). An introduction to structured modeling. Management
Science, 33(5), 547-588.
Georgantzas, N. C., & Acar, W. (1995). Scenario-driven planning: Learning to
manage strategic uncertainty. Westport, CT: Greenwood Press.
Hall, G. B., Bowerman, R. L., & Feick, R. D. (1997). GIS-based decision support
architecture and applications for developing countries. South African Journal
of Geo-Information,17(3), 73-80.
Holland, W. S. (2005, April 10-13). Integration of SAS® data warehousing/mining
and enterprise collaboration and data acquisition portals with ESRI-based
geographic information systems and public sector business systems. In SUGI
30 Proceedings, Philadelphia (Paper 188-30).
Malczewski, J. (1997). Spatial decision support systems. NCGIA Core Curriculum
in Geographic Information Science. Retrieved from http:// www.ncgia.ucsb.
edu/giscc/units/u127/u127.html
Malczewski, J. (1999). Spatial multi-criteria decision analysis. In J.-C. Thill (Ed.),
Spatial multi-criteria decision-making and analysis: A geographic information
sciences approach (pp. 11-48). Brookfield; New York: Ashgate.
Mennecke, B.E. (1997). Understanding the role of geographic information technolo-
gies in business: Applications and research directions. Journal of Geographic
Information and Decision Analysis, 1(1), 44-68.
Moloney, T., Lea, A. C., & Kowalchek, C. (1993). Manufacturing and packaged
goods in profiting from a geographical information system. Fort Collins, CO:
GIS World Books Inc.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Flexible Spatial Decision-Making and Support 183

Peterson, K. (1998). Development of spatial decision support systems for residential

real estate. Journal of Housing Research, 9(1), 135-156.
Post, E. (2000). Jade for developers (2nd ed.). Auckland, New Zealand: Pearson
Education.
Silver, M. S. (1991). Systems that support decision makers: Description and analy-
sis. Chichester, UK: Wiley Series in Information Systems, John Wiley and
Sons.
Simon, H. A. (1960). The new science of management decision. New York: Harper
and Row.
Sprague, R. H. Jr., & Carlson, E. D. (1982). Building effective decision support
systems. Englewood Cliffs, NJ: Prentice-Hall.
Yeh, G. -O. A., & Qiao, J. (1999). An intelligent solution support system for spa-
tial modeling and decision support. Proceedings of the 32nd Annual Hawaii
International Conference on System Sciences (Vol. 6, CD-ROM). January
5-8, Maui, HI.
Zhu, X., & Healey, R. (1992). Toward intelligent spatial decision support: Integrat-
ing geographical information systems and expert systems. In Proceedings
GIS/LIS ‘92 Annual Conference (Vol. 2, pp. 877-886). San Jose, CA: ACSM-
ASPRS-URISA-AM/FM.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran, Ilavajhala, & Sugumaran

Chapter.VIII

Development.of.a.
Web-Based.Intelligent.
Spatial.Decision.
Support.System.
(WEBISDSS):
A.Case.Study.with.
Snow.Removal.Operations
Ramanathan Sugumaran, Unversty of Northern Iowa, USA
Shrram Ilavajhala, Unversty of Maryland, USA
Vjayan Sugumaran, Oakland Unversty, USA

Abstract

A SDSS combines database storage technologies, geographic information systems

(GIS), and decision modeling into tools which can be used to address a wide vari-
ety of decision support areas (Eklund, Kirkby, & Pollitt, 1996). Recently, various
emerging technologies in computer hardware and software such as speedy micro-
processors, gigabit network connections, fast Internet mapping servers along with
Web-based technologies like eXtensible Markup Language (XML), Web services,
and so forth, provide promising opportunities to take the traditional spatial decision
support systems one step further to provide easy-to-use, round-the-clock access to

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

spatial data and decision support over the Web. Traditional DSS and Web-based
spatial DSS can be further improved by integrating expert knowledge and utilizing
intelligent software components (such as expert systems and intelligent agents) to
emulate the human intelligence and decision-making. These kinds of decision sup-
port systems are classified as intelligent decision support systems. The objective
of this chapter is to discuss the development of an intelligent Web-based spatial
decision support system and demonstrate it with a case study for planning snow
removal operations.

Introduction

Spatial.Decision.Support.Systems

The past decade witnessed an explosive growth of spatial data and various appli-
cations that utilize spatial data. Geographic information systems (GIS) have been
developed to facilitate storing, retrieving, editing, analyzing, and displaying spatial
information. The increasing complexity of spatial data and a need for better model-
ing requires decision support systems that can handle spatial data. This led to the
idea of spatial decision support systems (SDSS). Since the early 1980s, SDSS have
been used in several applications that provide spatial functionalities such as routing,
allocation modeling, and so forth.
Most of the existing SDSS do not employ any intelligent software components to
enhance decision support. Only a very few researchers have explored the possibility
of integrating intelligent software components with an SDSS for applications like
multi-criteria decision analysis, routing, and weather-based decision-making. Most of
the literature reviewed for Intelligent GIS systems deals with architectural as well as
implementation issues of GIS-based decision support systems and integrating them
with agents. The use of software agents for GIS-based systems is well documented
(Odell, Parunak, Fleischer, & Brueckner, 2003; Sengupta, Bennett, & Armstrong,
2000; Shahriari & Tao, 2002; Tsou, 2002). Most of these systems are not Web-based,
and they lack the advantages of Web-based systems like ease-of-use, cross platform
functionality, low maintenance costs, centralized data storage, and so forth.
Also, recent advances in Web technologies like rich site summary (RSS), XML
feeds, and asynchronous JavaScript and XML (AJAX) can help us device a seamless
interface by providing real-time access to data over the World Wide Web. There-
fore, integrating the process of decision-making with an intelligent component and
Web-based technologies proves to be very beneficial. When integrated with encoded
human intelligence, the spatial decision support systems can rival a human expert
in a particular domain (e.g., snow removal, traffic management, logistics, etc.).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran, Ilavajhala, & Sugumaran

This chapter explores and discusses the development of a Web-based intelligent

spatial decision support system for planning snow removal operations. Specifically,
this chapter addresses the existing problems with snow removal decision-making
in the USA. The SDSS discussed here integrates knowledge from snow removal
experts and real-time weather information into a Web-based interface. The system
is intended to provide advised decision support for officials at various departments
of transportation across the country, and to serve as a guideline for development of
a snow removal DSS for the decision-makers and stake-holders around the world.

Background.on.Snow.Removal.Operations

Snow removal operations during the winter are of prime importance in avoiding
traffic accidents and providing safe travel conditions on the nation’s highways and
city streets. Quality snow and ice control service is critical for preserving traffic
safety, maintaining city commerce, and allowing residents access to schools and
medical facilities (Hintz, Kettlewell, Shambarger, & Sweeney, 2001). Department of
Transportation (DOT) of each state is responsible for snow removal on all interstates,
and primary highways like the U.S. Federal highways and state highways. The city
streets are snowplowed by the Street Department of that city, or sometimes by the
DOT itself (Iowa Department of Transportation (IDOT), 2005). Snowplowing is
done according to a set priority assigned to each road depending upon the annual
average daily traffic (AADT). Higher priority roads like the interstates are cleared
before the lower priority routes like city streets.
Managing snow removal operations necessitates activities ranging from the prepa-
ration of roads before the snowfall by pre-wetting by salt (e.g., most roads and
bridge decks are pre-wetted with salt before the snow to avoid bonding of ice to
the pavement in order to prevent slippery road conditions) to the timely clearing of
the snow off the roads after the snowfall. The personnel in charge of snow removal
operations keep tabs on weather forecasts to verify conditions that require snow-
plowing. These conditions include snow, flurries, freezing rain, and sleet. Once the
weather forecast predicts these conditions, routes that need snow removal or salt-
treatment are determined, and the snowplowing vehicles are loaded with material.
These vehicles are then sent out to the roads, both before and after the snowfall.
Considering the number of managerial aspects like monitoring weather informa-
tion, allocating resources like vehicles, drivers, and material, it is an overwhelming
task for the personnel in charge of snow removal operations to administer timely
deployment of snowplows and manage efficient resource allocation. Therefore,
maintaining snow removal operations efficiently and optimally during the winter
is a major challenge for many governmental as well as non-governmental agencies
and prove to be a big budgetary burden (Hintz, Kettlewell, Shambarger, & Sweeney,
2001; Salim, Timmerman, Strauss, & Emch, 2002).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

Existing.Methods.for.Snow.Removal.Operations...............
and.Their.Disadvantages

Most of the existing methods for snowplowing management such as resource al-
location, inventory management, and routing are performed manually by a person
or persons in charge of snow removal operations. These methods, in most cases, are
based on a pre-determined set of static rules that are inefficient for snow plowing
and resource allocation under constantly-changing weather conditions. They often
result in inefficient and non-optimal results and tend to increase the cost of snow
removal. Moreover, the reaction time and the margin of error during heavy storm
conditions can prove deadly, and human-based methods are always prone to such
errors.
In addition, our observations with various snowplowing stations showed that there
is no substantial use of analytical methods before planning snowplowing opera-
tions to reduce cost and maximize overall effectiveness. For example, the methods
used for most local government agencies (e.g., the DOT) do not use any special
measures for efficient routing and resource allocation. Many of the existing meth-
ods do not choose the shortest or the quickest path from the snowplow station to
reach the snowplow route. Neither do they optimize resource allocation by taking
into consideration various important factors like the cost of operation of a vehicle,
availability of drivers or vehicles, and so forth. Such methods have a severe limi-
tation of being inefficient; even a very slight change in weather conditions would
almost demand double the amount of resources and unplanned travel for a different,
non-predetermined route, and thus result in waste of time and money. Moreover,
the existing manual methods cannot deal efficiently with resource allocation for
newly-added routes and changing weather conditions .
Also, existing methods do not use automated inventory management and control.
Inventory management is very important, and analytical tools must be used to keep
up with the demand of snow removal. Various factors like the lead time for order,
reorder point, stock levels, and so forth, must always be monitored to keep the nec-
essary materials (e.g., salt) in stock. Keeping in mind the importance of providing
speedy response after a snowstorm, inventory management tools are indispensable for
planning snow removal operations. In addition, the existing methods do not provide
any visual feedback when it comes to routing and resource allocation. Visual feedback
not only improves the perception of the current assignments, but also helps review
the allocations for better planning in future allocations. Also, there is no integrated
use of weather information that could alert the snow removal crews and provide a
scenario-based decision support. Thus, snowplowing operations must be carefully
planned to provide optimum resource allocation and efficient routing.
Further, a literature review reveals that there is lack of extensive research in the
field of GIS-based winter maintenance decision support systems. One of the existing

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran, Ilavajhala, & Sugumaran

SDSS for snow removal planning is the “snow removal asset management system,”
SRAMS for short. Salim, Timmerman, Strauss, and Emch (2002) developed a SDSS
called “snow removal asset management system”(SRAMS). SRAMS is a stand-
alone, personal computer-based program that utilizes a rule-based expert system
for decision support. The test runs of SRAMS concluded that snow removal can be
planned efficiently, by reducing the deadhead times and optimizing resources for
snow removal. While this system is useful to a certain extent, it ignores many aspects
of resource allocation like the availability of drivers for a certain shift, the ability
of a driver to drive a specific category of a vehicle, inability to read live weather
data, providing scenario-based solutions, and so forth. Being a stand-alone program,
SRAMS cannot be accessed from any other computer. Also, SRAMS was developed
bearing a non-technical user in mind and therefore, there is a significant need to
develop strong winter operations maintenance system that offers expert advice and
decision support, while being easy-to-use for even a non-GIS professional.
Another decision support system for snow removal is the winter road maintenance
decision support system (Mahoney & Myers, 2003). The maintenance decision
support system (MDSS) focuses mainly on using weather data to design a decision
support system for maintaining roads during winters. While MDSS uses advanced
weather data, it lacks the ability to calculate the amounts of snow removal material
which are needed. Also, MDSS lacks an asset management module that would have
made the system much more efficient.
Similarly, a winter maintenance decision support system developed by the U.S.
Department of Transportation (USDOT) allows viewing road-wise weather data,
and various weather parameters like Visibility, Air Temperature, Snow Rate, Snow
Accumulation, and so forth (U.S. Department of Transportation, 2004). Also, an-
other system called “weather support to de-icing decision-making” (WSDDM)
uses commercial weather data in the form of Next Generation Radar WSR-88D and
METAR surface weather reports from automated surface observing system stations
and observers. The system also uses snow gauges on the ground to measure the
precipitation to provide decision support. The WSDDM essentially provides decision
support for de-icing operations that take place at airports. The drawbacks of this
system are the usage of commercial data and the lack of an intelligent component
(e.g., expert system) for making crucial decisions.
In summary, there are no well established intelligent SDSS available for snow
removal operations that integrate automatic routing, asset management, and real-
time weather data. Further, the systems discussed in the literature do not leverage
the potential of GIS using Web-based architectures and integrate expert system
components for intelligent decision support. Therefore, there is a significant need
for an intelligent system that provides analytical tools and decision support through
visual feedback for managing snowplowing operations with optimal resource al-
location and efficient routing.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intelligent Spatial Decision Support System 189

Methodology

The methodology for developing WebISDSS encompassed the following four steps
(shown in Figure 1): (a) data collection, (b) data analysis and modeling, (c) GIS and
intelligent software component integration, and (d) providing results and obtaining
feedback. Each of these steps is briefly described below.
The first step in developing WebISDSS is data collection. During data collection,
three categories of data were collected. These categories are expert knowledge, spa-
tial data, and non-spatial data. The aim of expert knowledge elicitation is to gather
knowledge of various snow removal operations like resource allocation, routing,
and so forth, from the officials responsible for handling snowplowing operations.

Figure 1. Overall methodology adapted for developing WebISDSS (Source: Ila-

vajhala, 2005)

Study Area

Data Collection
Expect GIS-Based
Non Spatial
Knowledge Data Collection
Data Collection
Elicitation

Data Analysis & Modeling

Asset
Intelligent Management Routing and
Components and Weather Mapping Tools
Retrieval Tools

Component Integration and

Web Interface

Results Feedback

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Sugumaran, Ilavajhala, & Sugumaran

Officials from the Street Departments of Cedar Falls and Waterloo, IA, along with
the county and Iowa DOT’s highway maintenance supervisor in Waterloo were
consulted to obtain the knowledge about snow removal in the study area of Black
Hawk County. Through a series of direct interviews, a variety of information regard-
ing the existing snow removal procedures was gathered.
The spatial data gathered is the county-wide roads and street dataset of Black Hawk
County, IA. This data is available from the Iowa DOT’s maps Web site (IDOT, 2005).
The dataset essentially contains the centerlines for public roads including interstates,
U.S. and state Highways, county roads, city streets, park roads, and institutional
roads. For providing effective routing solutions like providing shortest or quickest
path from the snowplow station to the snowplow route, the optimized road data set
from GDT Inc., was purchased. This data set is commercially available in a format
called spatial data engine (SDE). SDE is a format optimized to provide quick rout-
ing solutions. ArcIMS utilizes the SDE data through its RouteServer extension. The
SDE data, in combination with ArcIMS, can be used to provide various kinds of
routing and logistics options such as vehicle routing with multiple stops, interstate
preference, and so forth.
The non-spatial data contains complementary data that could be used in combination
with spatial data. There are two kinds of non-spatial information: (1) roads-related
asset data, and (2) weather data. The road-based non-spatial data comes as a set of
database files (DBF) that contain pertinent information like traffic volumes, number
of lanes, lane length, and so forth. All the non-spatial data is available along with
spatial road data from Iowa DOT’s maps Web site in the form of DBF. These DBF
are linked to the shapefile by the unique segment ID of each road.
The weather data is obtained from the Internet using the rich simple syndication
or RSS feeds and XML technologies. RSS is an XML-based document format for
syndicating news and other timely news-like information (NOAA, 2003). The rea-
son for choosing weather feeds is that they provide a dynamic, seamless interface
for obtaining live weather data without page reloads. Also, the weather feeds from
most sources, including the ones from NOAA, are free of cost.
The data gathered through the various data collection procedures is analyzed and
modeled in order to design the core components for WebISDSS, namely, the intel-
ligence component (expert system) and the analytical tools (spatial and non-spatial
tools). A major emphasis of this chapter is the use of an intelligent software com-
ponent for decision support and integrating it with a Web-based GIS. The expert
system component of WebISDSS provides decision support using the knowledge
gathered from snow removal experts and real-time weather data. The functional-
ity provided by the expert system includes providing advised decision-making for
generating shortest paths and prioritized routes, allocating resources optimally for
snow plowing, and generating suggestions based on real-time weather data.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

The analytical tools available in WebISDSS provide functionality such as basic

navigation, routing, resource management, and so forth. These tools utilize spatial
and non-spatial data. The intelligent component and the analytical tools are inte-
grated into a single interface that could be accessed through the World Wide Web.
Therefore, a Web site is designed for publishing the road (street) map of Black Hawk
County, IA. The Web site, in effect, will serve both as a user interface and a way to
publish the resultant maps and show the results. Various routes are color coded for
effectively presenting the route information visually for better understanding.

Architecture and Implementation

WebISDSS is designed using Web-based client/server architecture and rule-based

artificial intelligence techniques. The implementation combines GIS and Web-based
programming components from two leading technologies, Environmental Systems
Research Institute (ESRI) and Microsoft. For developing the current prototype,
ESRI ArcIMS ActiveX Connector is used in combination with Microsoft Internet
information server and active server pages technology. The overall architecture of
WebISDSS is shown in Figure 2. It contains the following two main components:
(a) Web-based client interface, and (b) the server. These two components are briefly
described as follows.

The.Web-Based.Client.Interface

The primary functions of the Web-based client are providing the user interface and
facilitating communication between the user and the server components. WebISDSS
offers user interface by the means of a Web browser. The client corresponds with
the server through requests; the server processes clients’ requests and sends back
a response.

The Server

The server is the “backbone” of the entire system and is composed of a set of com-
ponents that work in tandem to process the user requests and provide a response.
The server has the following components: (a) Web server, (b) spatial server, (c)
non-spatial databases, (d) intelligent component, and (e) real-time weather retrieval
component. A brief description of these components is given as follows.
The primary component on the server-side is the Web Server. It is the “central
location” from where the user requests are received and processed or delegated
further to other components like the spatial server or analytical tools, and so forth.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
192 Sugumaran, Ilavajhala, & Sugumaran

The Web server for WebISDSS is the Microsoft Internet information server (IIS).
The server is hosted on a Microsoft Windows XP Professional machine. All server
side programming to create the analytical tools is accomplished using Microsoft
active server pages (ASP) technology. The primary reason for choosing Microsoft
technologies is their tight integration and ease-of-use.
The Spatial server is responsible for processing client requests that involve rendering
and presenting maps and map-related information to the user. The spatial server is
implemented using ESRI ArcIMS, the most popular Internet mapping software. The
communication between the map display (client) and the spatial server components
is performed via a standardized and proprietary XML format called ArcXML. WebI-
SDSS utilizes ArcIMS ActiveX Object Connector v. 4.1. The ActiveX connector
acts as an intermediary between the Web server and the spatial server by converting
the user requests over HTTP into ArcXML statements, the language that the spatial
server understands.
For providing routing solutions, the ArcIMS RouteServer extension is used. ArcIMS
RouteServer comes as an additional component with ArcIMS and provides routing
by using commercial data from TeleAtlas, Inc. This data is available in an optimized
format called Spatial Data Engine (SDE) format. Advantages of using RouteServer
are quick and efficient routing, complete integration with ArcIMS for use over the

Figure 2. The overall architecture of WebISDSS (Source: Ilavajhala, 2005)

User Interface Elements

Web Browser

HTTP Requests /
Responses

Internet

Data Spatial Requests /

Knowledge Responses
Intelligent
Component Web Server Spatial Server

Real-Time
Weather Information
Spatial Data

Weather Retrieval
Component

Spatial Data

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

Web, and provision for obtaining fully-customized routes (e.g., percentage of in-
terstates of travel, addition of multiple stops, etc.). All the routes are color-coded,
according to a pre-determined priority, to take advantage of the visualization that
GIS-based systems provide.
ESRI software was chosen over other vendors’ products because we found ESRI
products to be stable, scalable, and reliable. A detailed comparison of various In-
ternet mapping software products from different vendors is beyond the scope of
this chapter. Also, ESRI ArcIMS provides a programming model (object model)
that is compatible with our server-side scripting language (ASP) and Web server
(IIS), respectively.
The intelligent component developed for WebISDSS is a rule-based expert system.
The knowledge gathered from the snow removal experts was carefully sifted and
categorized into three knowledge categories, routing, resource allocation, and weather
data. Categorizing the knowledge provided modularity and structure for storing the
knowledge. The knowledge gathered was entirely conditions-based, in the form of a
set of “actions” resulting from a set of “conditions.” This sort of knowledge can be
best represented as “if-then” structures. The intelligent component thus incorporates
the expert knowledge and an inference engine. The expert knowledge is stored in
a repository called the “knowledge base” and fed into the inference engine that
provides a set of results or suggestions.
Rule Machine Corporation’s Visual Rule Studio offers an easy and efficient way
to code knowledge as “business rules”, providing for encoding expert knowledge
in the form of “if-then” rules. Each rule has a left-hand side (LHS) or conditions,
and right-hand side (RHS) or results. The RHS is evaluated only when the LHS is
true. These rules can be “fired” depending upon various data that is supplied and
can help make a decision, rivaling a human expert. The knowledge-base is coded
using Procedural Rule Language (PRL), a proprietary language of Rules Machines
Corporation.
The asset databases contain information about various transportation assets that are
utilized in snow removal activities such as characteristics of snowplows, capacity,
mileage for equipment maintenance, odometer reading prior to assignment of the
machine, details of the materials available at the central storage for snow removal,
available quantity, unit cost of the material, reorder point, assignment of the opera-
tor to preferred machines, and so forth. All the asset databases are implemented in
DBASE IV database file (DBF) format and are managed through the Web-based
client.
As mentioned previously, the weather data for WebISDSS is obtained using the
RSS and XML feeds technology. The weather data is embedded into the system by
reading the free RSS and XML Weather feeds provided by www.weatherroom.com
and NOAA’s National Weather Service (http://www.nws.noaa.gov). These XML
feeds essentially provide an XML document that encodes various weather param-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran, Ilavajhala, & Sugumaran

eters including the air temperature, wind speeds, wind direction, wind-chill, and
visibility. Live as well as the forecast weather data can be obtained and presented
to the user in this fashion by reading a particular weather feed from the Internet ,
and loading the information contained in the weather feed document. The current
weather conditions help generate suggestions for material assignment and help
determine the allocation of resources. The forecast data is used to generate alerts.
The weather information can also be used for generating scenario-based solutions.
Further, the live weather data, in combination with the encoded rules, help make
intelligent decisions for snow removal and resource management and allocation.

Results

Application.Example:.........................................................
Planning.Snow.Removal.Using.WebISDSS

One of the main objectives of WebISDSS is to provide easy-to-use Web-based

interface and intelligent decision support for planning snow removal operations.
To achieve this goal, the system uses the knowledge gathered from snow removal
experts as well as real-time weather data. It provides an uncluttered interface, divided
neatly into various “areas.” For example, the menu and the tool bar appear on top,
and the map and map layers area are beneath the menu, and a “message area” dis-
plays detailed messages to the users. The interface is mainly menu-driven, making
it very easy to use. All the menu options are given appropriate, non-technical names
so that a naïve computer user can also utilize the system. Figure 3 shows the main
interface of WebISDSS and its organization into various “areas.”
The system generates verbose and detailed alerts, warnings, tips, and other messages
that help a non-technical user better understand the system. Most of the commonly-
used menu options are provided as tools on a separate tool bar and can be used by
clicking on the appropriate icon. When the user clicks on a particular tool, a message
is displayed in the message area of the window. A detailed help system is available,
which avoids technical jargon and provides the user with an easy guide to use the
system. The salient features and functionalities of the menu interface of WebISDSS
are briefly described as follows.

•. Weather.menu:The Weather menu lets the user set and view real-time weather
information. The weather menu is shown in Figure 4. The user can choose to
view live or forecast weather information and also store it. Weather conditions
can be set either manually by entering a set of parameters or by automatically

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

Figure 3. The main interface of WebISDSS (Source: Ilavajhala, 2005)

Figure 4. The “Weather” menu interface of WebISDSS

obtaining them from the Web. The weather menu also provides an option to
view live weather through NOAA NEXRAD radar image for the state of Iowa
(Figure 5) or Forecast data. Figure 6 shows a screenshot of forecast weather
retrieved by WebISDSS.
•. Route.menu:.Route menu provides options for creating, deleting, and load-
ing route information. The user can choose to create a route by clicking on
the “start route” option, and then clicking on the road segments to snowplow.
Clicking “end route” menu item will save the route, and generate driving
directions from the station to the route by creating the shortest or quickest

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran, Ilavajhala, & Sugumaran

Figure 5. NEXRAD radar image obtained from NOAA

Figure 6. Screenshot of weather forecast information retrieved by WebISDSS

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

route, as chosen by the user. The user can also delete a route or a route seg-
ment. Deletion of route can either be temporary (deletion from the map) or
permanent (deletion from the disk).
•. Vehicles.and.Drivers.menu:.The Vehicles and Drivers menu provides options
to assign drivers and vehicles. Also available from this menu is an option to
manage drivers and vehicles by adding, deleting, or editing driver or vehicle
information.
•. Materials.menu:.Materials menu gives options related to materials manage-
ment and control. Options include inventory analysis, material assignment,
addition/deletion/editing snowplowing materials such as salt, sand, and so
forth.
•. Help.menu: The Help menu lets the user obtain information about how to use
the system. The “Using WebISDSS” option shows the user the “Help central”
screen with various options for getting help. Also, the help menu shows the
product and copyright information.

System. Walkthrough

The expert system provides suggestions based on user input. These suggestions
include initiation of snow plowing operations, material to utilize for snowplowing,
amount of material that must be used, number of lines to snowplow, and material
assignment. To begin with, the user sets the weather conditions either by entering
weather parameters like current conditions, temperature, dew point, wind speed,
and so forth, or by using the system to read weather conditions from the Internet.
The real-time weather retrieval component reads the encoded current and forecast
weather data through the Web and provides a readable output to the user. Further,
these suggestions are used for assisting and planning resource allocation, initiat-
ing snowplowing operations, and so forth. Thus, the real-time weather retrieval
component works with the Web server and the intelligent component to provide
suggested actions depending upon real-time weather information retrieved from the
Web. These weather conditions are stored in a database file (DBF).
The expert system’s inference engine examines the weather conditions stored in the
database. The inference engine then produces a set of suggestions and stores these
suggestions on the server’s hard drive. These suggestions advise the user whether
to initiate a snowplowing operation, which materials to use for snowplowing, and
how much of each material should be used. Depending on the precipitation, the
expert system also produces suggestions to either initiate anti-icing or de-icing op-
erations. Other parameters that are used in producing suggestions are visibility and
dew point. The system advises the exercise of caution if the visibility is less than a

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran, Ilavajhala, & Sugumaran

Figure 7. Sample rules for material allocation

mile. The dew points are used to forecast precipitation and to caution the users to
wet the bride decks to prevent icing.
Once the weather conditions are accessed and stored in the database, routes can
be created. The WebISDSS calculates route parameters needed for snowplowing
such as the number of lane miles, total route length, and so forth. These parameters
are stored in a database along with other route parameters like route identification
number, and so forth. These parameters are further used for making suggestions
for material allocation.
After the routes are created, material can be assigned to these routes. Based on the
weather conditions stored in the database, and various route parameters such as the
route length, and so forth, the expert system provides suggestions on how much
material should be used for snowplowing. Figure 7 shows part of the “rules” that
the expert system uses to provide suggestions on material allocation and Figure 8
shows the material allocation interface of WebISDSS. Also, the expert system cal-
culates the amount of material needed per route depending upon the precipitation
levels. These calculations are based on a set of guidelines that the Iowa DOT uses
to calculate material allocation.
WebISDSS also provides a quick way to view all current assignments in one screen.
The menu option “Map > Show Current Assignments” provides a screen with cur-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 

Figure 8. Material assignment interface of WebISDSS

rent vehicle assignments, driver assignments, weather conditions, and material

estimations. Figure 9 shows one such screenshot.

Summary. and. Future. Work

We have presented a Web-based intelligent spatial decision support system (WebI-

SDSS) in this chapter that overcomes the disadvantages of existing systems by
providing intelligent decision support for effectively planning snow removal
operations. The system has been designed using ArcIMS ActiveX Connector with
ArcIMS RouteServer extension and Web technologies such as ASP, XML, and RSS.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
00 Sugumaran, Ilavajhala, & Sugumaran

Figure 9. Viewing all current assignments

The system also integrates an intelligent software component with geo-spatial and
analytical techniques for providing real-time decision support.
There is ample scope for further refining the WebISDSS system described in this
chapter. Our future work is aimed at making WebISDSS more efficient and effec-
tive. For extending WebISDSS, ArcGIS Server 9.x can be used in conjunction with
ArcObjects to provide advanced functionality like adding new features (e.g., new
roads) and advanced geo-analytical tools. Additional solutions based on “intel-
ligent technologies” such as intelligent software agents can be explored to extend
the fundamental rule-based system to a full-scale “intelligent agent” based system.
Also, the use of commercial weather data service that provides frequent updates
and segment-wise conditions can improve the efficiency of the current prototype by
generating very accurate snowplow routes and by cutting costs further. The weather
conditions can also be shown on the map to provide a visual display of various
road weather conditions around the county. Furthermore, advanced algorithms
can be implemented for sophisticated inventory analysis for better management of
materials. Similarly, improved algorithms can be employed for driver and vehicle
assignment purposes. Currently, the prototype is optimized to work with the Mi-
crosoft Internet Explorer browser. It can be expanded to work with other browsers
such as Netscape, Mozilla, and Firefox by carefully considering the cross-browser
functionality issues.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Development of a Web-Based Intellgent Spatal Decson Support System 0

There is plenty of research and development yet to be done in the field of Web-
based intelligent decision support systems. The prospect of employing intelligent
autonomous agents for decision support is very promising. Advances in the fields of
artificial intelligence and human computer interaction will have a big impact on how
spatial decision support systems behave. Another emerging trend is the availability
of decision support systems though a Web-based interface. This ensures cross-plat-
form compatibility and ease-of-use. Further, these interfaces can be extended to
work from portable devices like palmtops, cell phones, and so forth. In the coming
years, the Web will change dramatically, reaching a wide population via a variety of
handheld, smart devices. The future will also usher us into new ways of interaction
with computing devices in a radically different way from today’s interfaces.
The future for spatial decision support systems is promising considering the ex-
plosion of spatial data that we are experiencing. These spatial decision support
systems provide many types of scenario-based decision support for governmental
and non-governmental organizations (e.g., utility installation planning, right-of-
way management, demographic research, etc.) and will be capable of providing
advanced spatial analysis to further assist in decision-making. For example, the
future spatial decision support systems will include active raster and vector data
processing to provide instant results. This is a huge improvement from the various
online mapping systems that use only raster data for providing driving directions,
locations, and aerial photography.

Acknowledgments

The authors would like to thank Midwest Transportation Consortium (MTC), Iowa
DOT, and the University of Northern Iowa for sponsoring this research. We further
extend our thanks to Mr. Russ Frisch, Highway Maintenance Supervisor 3 of Iowa
DOT, Mr. Steve Decker of Waterloo Street Department, and Mr. Darrell Fanning
and Mr. Steve Ravn of Cedar Falls Streets Department for their time and input.

References

Eklund, W. P., Kirkby, S., & Pollitt, S. (1996). A dynamic multi-source Dijkstra’s
algorithm for vehicle routing. In Proceedings of the Australian and New Zealand
Conference on Intelligent Information Systems (ANZIIS ’96) (pp. 329-333).
Adelaide, Australia: IEEE Press.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Sugumaran, Ilavajhala, & Sugumaran

Hintz, G., Kettlewell, A., Shambarger, E., & Sweeney, T. (2001). Milwaukee’s snow
and ice control service. La Follette Policy Report, 12, 16.
Iowa Department of Transportation (IDOT). (2005). IDOT digital maps. Retrieved
August 29, 2004, from http://www.dot.stae.ia.us/
Ilavajhala, S. (2005). A Web-based intelligent spatial decision support system for
planning snow removal operations. Unpublished master’s thesis, submitted to
the Computer Science Department at the University of Northern Iowa, Cedar
Falls, Iowa.
Mahoney, P. W., & Myers, W. (2003). The winter road maintenance decision support
system (MDSS) project update and future plans. In Proceedings of the 19th
Conference on IIPS, 2003. Retrieved June 17, 2004, from http://ams.confex.
com/ams/annual2003/techprogram/paper_51856.htm
NOAA (National Oceanic and Atmospheric Administration). (2003). NOAA's NWS
RSS Library. Retrieved February 25, 2005, from http://www.nws.noaa.gov/
Odell, J. J., Parunak, V. D. H., Fleischer, M., & Brueckner, S. (2003). Modeling
agents and their environment. The Physical Environment Journal of Object
Technology, 2, 43-51.
Salim, M. D., Timmerman, A. M., Strauss, T., & Emch, M. (2002). Artificial-intel-
ligence-based optimization of the management of snow removal assets and re-
sources (MTC Final Report). Cedar Falls, IA: University of Northern Iowa.
Sengupta, R., Bennett, A. D., & Armstrong, P. M. (2000). Agent-oriented modeling
environment for spatial decision support. In Proceedings of the First Interna-
tional Conference on Geographic Information Science. Retrieved December
26, 2004, from http://www.giscience.org/GIScience2000/papers/188-Sengupta.
pdf
Shahriari, N. & Tao, C. V. (2002). Applications of agent technology in GIS. Journal
of Geographic Information Science, 8, 78-85.
Tsou, M. (2002). Adopting software agents for Internet mapping and distributed
GIServices. In Proceedings of the ESRI International User Conference, 2002
(pp. 234). San Diego, CA: ESRI.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 0

Chapter.IX

Conservation.Studio:
Dynamically.Modeling.and.
Mapping.Habitat.Suitability
Todd G. Olson, Landx, Inc., USA
Bran N. Hlton, Claremont Graduate Unversty, USA

Abstract

Habitat suitability modeling in some form is required for virtually every regulatory
action and conservation planning process that involves rare and/or endangered
species or habitat types. Conservation Studio is a spatial information system that
automates the entire process of conservation modeling, simulation, and planning.
Conservation Studio consists of four software modules: data acquisition interface,
habitat suitability analyst, conservation criteria developer, and implementation
modeler, the latter of which has been developed as a working prototype. The imple-
mentation modeler models the outcome of using tradable conservation credits to
conserve habitat resources in a specified geographical plan area. Future design and
development activities for this spatial information system are discussed.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Olson & Hlton

Introduction

Habitat suitability modeling in some form is required for virtually every regulatory
action and conservation planning process that involves rare and/or endangered spe-
cies or habitat types, yet no software application currently exists that automates the
entire process of developing such models. Consequently, such modeling efforts are
costly, and the answers that they are intended to provide can be long-delayed when
they are most in demand, at the beginning stages of a conservation project as the
interested parties are attempting to assess the scope of the project. The following
sections describe an ongoing research and development project entitled “Conserva-
tion Studio,” a spatial information system for conservation modeling, simulation,
and planning, to overcome these problems.
Conservation Studio guides non-experts through the entire process of generating
habitat suitability models; no current spatial information system performs this
comprehensive function. This spatial information system can be used to support
decision-makers in various earth science applications, such as invasive species
management, community growth, and water quality management (NASA, 2005).
It can also be used by conservationists, environmental regulators, parties subject to
environmental regulation, forest managers, and others who need to generate habitat
suitability models relevant to their activities.
Furthermore, this spatial information system provides an example of how remotely-
sensed earth observation data can be made useful to practitioners who are not spe-
cialists in geospatial data management or geostatistical analysis.
The significance of this spatial information system is that it:

• Automates the creation of habitat suitability models, making it possible for

practitioners to perform this task without the need for expertise in various
technical disciplines
• Dramatically decreases the time and expense required to generate such mod-
els
• Brings scientific rigor to a process that is often completed in an ad hoc fashion
due to the difficulty and expense of completing it in a more scientifically-de-
fensible manner

No existing software application automates the entire process of developing habitat

suitability models. Currently, several different software tools are required to perform
such analyses, and the user must assemble these tools. Even after the tools have
been assembled, they are not wholly integrated, and they are not specific to the
task of developing habitat suitability models. Furthermore, the data available for

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 0

use with these tools is often disparate and exists in various incompatible formats.
Consequently, significant expertise in the disciplines of geographic information
systems (GIS), database management, geostatistics, and conservation biology is
currently required to perform habitat suitability analyses.
Conservation Studio converts a task that requires an entire team of experts into a
task that can be performed by a non-expert with moderate experience in conservation
planning and conservation biology. Consequently, the time and expense of perform-
ing these analyses are dramatically reduced, while the quality of such analyses are
greatly increased by helping the user focus on conservation issues rather than on
numerous technical problems involved with performing the task.

Background

Habitat.Suitability.Modeling

Two approaches to habitat suitability modeling have been employed widely in the
United States: (1) the Habitat Suitability Index (HSI) approach, based on the Habitat
Evaluation Procedures developed by the U. S. Fish and Wildlife Service in 1980, and
(2) the correlation of vegetative cover and vertebrate species occurrences used by the
National Gap Analysis Program (Scott & Jennings, 1998), which is sponsored and
coordinated by the Biological Resources Division of the U.S. Geological Survey.
The HSI approach was developed in 1980 before the ready availability of GIS tools,
and therefore relies heavily on manual processes that require expert input. The gap
analysis approach, developed in 1987, was designed to identify biodiversity correlates
on a very large scale; however, its methods are not typically suited to predictive
modeling on a finer scale. In spite of the limitations of these two approaches, the
documented experience gained through their application over the years is instructive
and has guided the design and development of Conservation Studio.
Besides the two major, systematic habitat suitability modeling efforts described
above, numerous habitat suitability modeling projects have been documented that
take varying approaches to data selection, data acquisition, and data analysis. These
cases have also been studied for their unique perspectives on the problem.

Geostatistical.Analysis.Applied.to.Habitat........................
Suitability.Models

With the maturation of GIS software, the ability to perform computerized geostatistical
analysis to develop and test habitat suitability models is rapidly increasing (Larson &
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Olson & Hlton

Sengupta, 2004; Nikolakaki, 2004; Store & Jokimäki, 2003). Consequently, the body
of literature in this area is growing quickly (Goovaerts, 1997; Guisan, Edwards, &
Hastie, 2002; Scott, Heglund, & Morrison, 2002). For instance, NatureServe has a
project under way to evaluate various approaches to what it refers to as “predictive
range mapping” (NatureServe, 2005b). Also, geostatistical software that interoper-
ates with GIS software is becoming increasingly available, as exemplified by such
packages as FRAGSTAT, Variowin, GEOEas, Surfer, Gstat, GSLIB, SpaceStat,
S-PLUS, and Spatial Statistics Toolbox for Matlab.

Data.Model

NatureServe has published a “biodiversity data model” that is used extensively across
the United States and in other parts of the Western Hemisphere for purposes of rep-
resenting ecological data of many types and at many scales (NatureServe, 2005a).
Such data are gathered and made available through “Natural Heritage Programs”
that are administered by states and other units of government. This data model has
informed the design of both the input and output formats for Conservation Studio
to facilitate the acquisition of species occurrence and other relevant data, as well
as to be able to produce output that can be readily used by others.

Data.Interoperability

The implementation of interoperable solutions and applications for geospatial services,

data, and applications is the focus of the ongoing work of the OpenGIS Consortium
(OGC) (Open GIS Consortium Inc., 2003b). The OGC has specified multiple data
services that provide ready access to collections of spatial data over the Web. These
services include the Web map service (WMS), the Web feature service (WFS), and
the Web coverage service (WCS) to access spatial data. These services are enabled
by a number of underlying technologies such as registry, portrayal, processing, and
data services. Additional OGC services also support privacy and access controls
based on authenticated user identity.
Two key issues outlined in the Decision Support Priority Application Theme of the
OGC are: (1) defining methods for decision analysts to organize, store, and share
geo-analysis projects; and, (2) extending the access of decision support clients to
predictive modeling in addition to observations (Open GIS Consortium Inc., 2003a).
The OGC Web Services described above address these issues and would enable
the development of Web-based decision support tools. The OGC defines decision
support tools and systems as “interactive computer-based systems designed to help
people and organizations retrieve, summarize, analyze data/information and con-
duct predictive analysis on scenarios that enable enhanced capacity to make better

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 0

decisions” (Open GIS Consortium, Inc., 2003a). In a similar manner, Conservation

Studio will serve as a decision support tool built on these services.

Conservation.Studio

Conservation Studio consists of four software modules: data acquisition interface,

habitat suitability analyst, conservation criteria developer, and implementation
modeler. The first module, currently under development, would allow the user to
easily acquire data over the Internet that is relevant to a specified project. The sec-
ond module would apply geostatistical analyses to the acquired data and other data
supplied by the user to identify correlates of habitat suitability and map predicted
suitable habitat. The third module would assist the user in establishing quantitative
criteria that can be said to meet specified conservation objectives for a resource. The
fourth module, which has been developed to a working prototype, assists the user
in describing and modeling an implementation plan for the resource.
The combination of these four modules, when fully developed, would extend the
use of remotely-sensed earth observation and other spatial data beyond the habitat
suitability modeling process. It would assist a user step-by-step through the entire
conservation planning process to create a practical, science-based plan for conserv-
ing a given species, set of species, or natural community. This spatial information
system would provide not only habitat suitability analyses for various earth science
applications, but also tools to assist decision-makers in developing and implement-
ing plans for the conservation of species and natural communities. Figure 1 below
illustrates the relationship among the four software modules.
Furthermore, Conservation Studio could be extended for use in a GPS-enabled
hand-held device to allow for real-time analysis in the field. Through the habitat
suitability analyst, a field biologist could obtain instant feedback and suggestions
for furthering the fieldwork to refine the habitat suitability model. Development of
this capability will also be part of future research and development activities.

Data Acquisition Interface

Based on a specified geographical area (a “study area”) as the only required user
input, the data acquisition interface will: (1) directly acquire over the Internet
(without needing to use a separate Web browser) spatial data layers from NASA
and other diverse sources that are specifically relevant to creating predictive habitat
suitability models for species or natural communities (“resources”) in the study
area; (2) directly acquire over the Internet any available species-specific data for
biological resources known to exist in the study area or specified by the user; and

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Olson & Hlton

Figure 1. Conservation studio system architecture

(3) convert the acquired data into a raster format that is easily manipulated in the
habitat suitability analyst module described next.

Habitat.Suitability.Analyst

The habitat suitability analyst will automate the process of: (1) performing geosta-
tistical analyses on the data obtained using the Data Acquisition Interface to find
correlates of habitat suitability for a given resource in the plan area; (2) suggesting
geographic locations for new field work based on analytical results; (3) incorporat-
ing new field data and repeating steps #1 and #2 in an iterative fashion until the
statistical errors are reduced to an acceptable level of tolerance; and (4) creating a
spatial model based on the statistical analyses that predicts the locations and quality
of suitable habitat for the resource.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 0

Conservation.Criteria.Developer

Conservation criteria developer will assist the user in establishing quantitative criteria
that can be said to meet specified conservation objectives for a resource. The criteria
may include such factors as quantity and quality of habitat to be preserved (qual-
ity being based on the habitat suitability model developed using habitat suitability
analyst), the patch sizes of the preserved habitat, the shape of the preserved patches,
the edge effect on the preserved patches, the distribution of the habitat across the
historic or extant range of the resource, and so forth. The criteria will be based on a
combination of generalized principles of: (1) species and habitat needs inferred from
the habitat suitability results, (2) general principles of conservation biology, and (3)
species-specific habitat requirements supplied from available databases and by the
user. This component could incorporate a population viability analysis tool.

Implementation.Modeler

Implementation modeler assists the user in describing and modeling an imple-

mentation plan for the resource. The modeling is conducted in two steps. First, the
user describes an implementation strategy for testing using tools provided in the
software. Second, the software analyzes the specified strategy using “Monte Carlo”
analysis techniques to generate multiple outcomes with given ranges of variability
in input variables. Each outcome consists of a grid of cells in which some cells are
conserved and others are developed for economic uses. Statistical analyses can be
performed on the outcomes to assess the likelihood that the various conservation
criteria developed using conservation criteria developer will be achieved, the eco-
nomic efficiency of the plan, and so forth. Various implementation strategies can be
modeled for the same project to compare the tendencies of the different strategies to
produce different conservation and economic outcomes. The next section describes,
in detail, the implementation modeler.

Implementation. Modeler

The.“Habitat.Transaction.Method”.and.Implementation.
Modeler.Software

The lead author, along with others, has developed a conservation planning imple-
mentation methodology known as the “habitat transaction method,” which employs
tradable conservation credits to implement large-scale, multiple-landowner conserva-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Olson & Hlton

tion plans (Glickfeld, Jacques, Kieser, & Olson, 1995; Olson, 1996; Olson, Murphy,
& Thornton, 1993; Sohn & Cohen, 1996). The lead author has also co-developed,
in concert with software engineer Andrew Morgan, a prototype of the implementa-
tion modeler, which models the outcome of using tradable conservation credits to
conserve habitat resources in a specified geographical plan area. He has used the
software in his conservation planning consulting practice both to demonstrate the
tradable credit concept and to model outcomes of conservation plans under con-
sideration. Implementation modeler utilizes grid-based geospatial data layers that
represent habitat values and economic development values for each cell in the plan
area, and it graphically models potential patterns of development and conservation,
resulting in an outcome data layer. The current version is written in Java and runs
on Windows, Mac OS X, and Linux.
As it is further developed, implementation modeler will allow the user to model
the use of several different implementation strategies for meeting a specified set of
conservation objectives. In addition to tradable conservation credits, other strate-
gies that could be modeled include direct regulation, development fee, transferable
development rights, and combinations of techniques. The user could then compare
outcomes of the different strategies both in terms of likelihood of meeting the
conservation objectives and the economic efficiency in meeting the conservation
objectives.
The implementation modeler allows the user to describe: (1) existing land use and
open space patterns within the study area, (2) the relative development value of the
remaining undeveloped land, (3) the relative conservation value of the remaining
developed land, and (4) a set of rules governing future development and conservation
that describe the conservation strategy. The implementation modeler then models
future land development and conservation patterns based on: (1) the relative develop-
ment value of land, (2) the conservation rules that apply to new development under
the specified strategy, (3) the cost of complying with the rules in order to develop
land, and (4) a degree of randomness in decision-making to take into account factors
that the model does not address explicitly. Because of the randomness introduced
into the model, each outcome will be different. Each outcome consists of both a
map of lands projected to be protected, developed, and conserved as well as a set
of statistics to provide a quantitative analysis of the conservation and economic
results. Performing much iteration allows patterns of new development and new
conservation to emerge that can be subjected to statistical analysis and comparison
among conservation strategies.
Once the user has provided the map-based information described above, the user is
then able to make parameter selections for a model run that include:

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 

1. Conservation rules that apply to new development under the specified strategy,
such as conservation set-aside requirements and/or development fees that must
be paid, and so forth
2. Factors that affect the value of land for development over time, such as adja-
cency to other development and associated infrastructure
3. Factors that affect the conservation value of conserved or undeveloped land
over time, such as habitat fragmentation and edge effects
4. A degree of randomness to be assumed in decision making to take into account
factors that the model does not address explicitly

The current version of implementation modeler utilizes an elementary version

of the parameter inputs; a key priority for further development is to increase the
robustness of the model with respect to the input and modeling of the above input
parameters.
Figure 2 is a screenshot illustrating future habitat conditions as predicted by specified
parameter selections using the implementation modeler (Map tab); areas shaded green
are protected/conservation areas while those shaded blue are developed areas.

Figure 2. Conservation Studio implementation modeler — results of simulation

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Olson & Hlton

Applications.of.Conservation.Studio

Conservation Studio should be useful to conservation planning practitioners in both

governmental and non-governmental sectors. Potential applications for each sector
are discussed in detail below.

Governmental

Government agencies have a need to determine the location of the habitat of sensitive
species and habitat communities on a regular basis. Specific government applica-
tions that would benefit from Conservation Studio include:

•. Endangered.species.range.modeling: developing range maps for species to be

listed as threatened or endangered under the Federal Endangered Species Act
(the “ESA”) and some state endangered species acts (U.S. Fish and Wildlife
Service, or “USFWS”)
•. Critical.habitat.designations: designating “critical habitat” for threatened
and endangered species as required by the Federal ESA (USFWS)
•. Jeopardy.determinations: analyzing the threat to the habitat of a threatened
or endangered species caused by a proposed Federal project to determine
whether the project would jeopardize the continued existence of the species
or adversely modify critical habitat for the species (USFWS)
•. Federal.land.planning: locating the habitat of sensitive species for purposes
of developing land use plans for Federally-owned land (e.g., Bureau of Land
Management, Forest Service, National Park Service)
•. Tribal.land.planning: mapping sensitive biological resources to protect the
natural heritage of tribal lands and to comply with the ESA (various Native
American tribes)
•. State.and.local.resource.regulation: supporting the enforcement of state and
local biological resource regulations by modeling the existence of protected
habitat (states, counties, and cities)
•. Non-Federal.regional.conservation.planning: mapping of habitat resources
under the jurisdictions of states and counties for purposes of long-range plan-
ning; for example, California is undertaking the California Legacy Project,
which has approximately $10 billion in bond funds earmarked for parks and
conservation and is attempting to map its habitat resources to prioritize the
use of these funds
•. Development.of.habitat.conservation.plans: analyzing the existence and
location of suitable habitat for the species for purposes of developing “habitat

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 

conservation plans” (HCPs) under the ESA; in exchange for developing an HCP,
the applicant can obtain authority to “take” a certain amount of endangered
species habitat so long as the species in question is sufficiently conserved by
the plan (counties and cities; see also private sector HCPs as follows)
•. Evaluation.of.habitat.conservation.plans: evaluating HCPs that are proposed
under the ESA by either public (see bullet immediately previous and “Non-
Governmental” described later) (USFWS)

Habitat suitability modeling is necessary or useful in various earth science applica-

tions. Some examples of how Conservation Studio would support this include:

• Modeling habitat for both invasive non-native species and the native species
that are threatened by such invasive species
• Modeling the location of habitat that may be impacted by community
growth
• Discovering correlations between habitat suitability and water availability/
quality

Non-Governmental

The private sector also has regular need to perform habitat suitability modeling,
such as in the following applications:

•. Development.of.habitat.conservation.plans: performing analysis in support

of private HCPs under the ESA (landowners, industry groups)
•. Development.of.conservation.banks: analyzing the quality and importance
of habitat in a proposed “conservation bank” to establish the number of miti-
gation “credits” that the bank owner will obtain in exchange for protecting
sensitive habitat; such credits can be used under programs administered by the
USFWS (for endangered species habitat), the Army Corps of Engineers (for
wetlands habitat), and other regulators to mitigate project impacts or sold to
third parties as mitigation for their project impacts (landowners)
•. Prioritization.of.land.conservancy.acquisitions: mapping of the habitat of
rare and endangered species and natural communities for purposes of priori-
tizing land acquisitions and other conservation activities (private, non-profit
land conservancies)
•. Environmental.consulting: supporting as a private environmental consultant
to any of the activities listed above (consultants)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Olson & Hlton

The previous examples illustrate how Conservation Studio can make earth science
observations accessible to practitioners who do not have necessary technical exper-
tise by automating the entire process from data acquisition through data analysis
and map-based data presentation. The concept of habitat suitability modeling is
intuitive, and both the data inputs into the analysis and the data outputs from the
analysis are visual. As a result, Conservation Studio demonstrates how the use of
remotely-sensed data in a visual manner can be easily grasped.

Future.Design.and.Development.Activities

In its current phase of development Conservation Studio consists of a prototype

of the implementation modeler module described above. The goal of this project
is to fully develop the four application modules of Conservation Studio, data ac-
quisition interface, habitat suitability analyst, conservation criteria developer, and
implementation modeler, into a complete, unified spatial information system. How-
ever, spatial information system development, and indeed all information system
development, is impacted by the rapid development of emerging technologies and
various governmental regulations. As a result, the objectives of this project will
address these issues by:

•. Applying.standards-based.technologies: Selection of standards-based tech-

nologies to facilitate interoperability with software and systems external to
Conservation Studio (International Organization for Standardization, 2000;
Open GIS Consortium Inc., 2002, 2003b);
•. Incorporating.open.source.software: Integration of open source software
whenever possible to minimize the deployment cost of the Conservation Studio
(Bollinger, 2003; Câmara & Onsrud, 2004; Open Source Initiative, 2005; The
GNU Project, 2005); and
•. Developing. an. easy-to-use. interface: Emphasis on system usability and
quality in use in the design of the Conservation Studio user interface in order
to make its component modules useful to a wide variety of users with varying
levels of expertise in its underlying technologies and processes (Bevan, 1999;
Brooke, 1996; International Organization for Standardization, 1997; Losavio,
2002).

The following is the research and development methodology that will be used in
the design and development of Conservation Studio.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 

1. Conservation Studio Spatial Information System

Design the overall Conservation Studio spatial information system to incorporate

the data acquisition interface, the habitat suitability analyst, the conservation cri-
teria developer, and the implementation modeler modules, including the following
components:

• Graphical user interface

• File system
• Map data display
• Statistical graphing display
• Method for integrating component software such as the data acquisition in-
terface and habitat suitability analyst
• Potential deployment under a Web services model

2. Data Acquisition Interface

Design the data acquisition interface software module, including the following
components:

• Spatial data server for locating, previewing, and downloading spatial and other
data relevant to the user project
• E-commerce tool to facilitate the direct purchase of data through the proposed
client software
• Conversion utility for reading necessary data types and converting them to
the native format that will be used in the habitat suitability analyst

3. Habitat Suitability Analyst

Design the habitat suitability analyst software module, including the following
components:

• Internal data format(s)

• Project set-up (with “interview” of user on project characteristics, information
about the data sets, project definition of habitat suitability, etc.)
• Geostatistical query strategy for finding habitat suitability correlates

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Olson & Hlton

• Geostatistical analysis functions

• Analysis and presentation of geostatistical results, including tabular and graphi-
cal presentations, and suggestions for additional field surveying
• Interactive habitat suitability modeler that allows the user to combine the
geostatistical results with other input to define a habitat suitability model
• Habitat suitability mapper and GIS data exporter

4. Conservation Criteria Developer

Design the conservation criteria developer software module, including the follow-
ing components:

• Raster map display and interface

• GIS data acquisition and rasterization
• Interface for describing implementation strategy rules
• Interface for describing landowner and government behavioral assumptions
• Monte Carlo modeler and data capture with animated display of model in
progress over time
• Outcome description in the form of both maps and statistics
• Outcome analysis for multiple implementation strategy variations

References

Bevan, N. (1999). Quality in use: Meeting user needs for quality. The Journal of
Systems and Software, 49, 89-96.
Bollinger, T. (2003). Use of free and open source software (FOSS) in the U.S. De-
partment of Defense (No. MITRE Report Number MP 02 W0000101 v1.2.04).
The MITRE Corporation.
Brooke, J. (1996). SUS: A ‘quick and dirty’ usability scale. In P. Jordan, B. Thomas,
B. Weerdmeester, & I. McClelland (Eds.), Usability evaluation in industry (pp.
189-194). London: Taylor & Francis Ltd.
Câmara, G., & Onsrud, H. (2004). Open-source geographic information systems
software: Myths and realities. Paper presented at the Open Access and the Public
Domain in Digital Data and Information for Science, Washington, DC.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Conservaton Studo 

Glickfeld, M., Jacques, S., Kieser, W., & Olson, T. (1995). Implementation techniques
and strategies for conservation plans. Land Use & Envirornment Forum, 4,
12-28.
Goovaerts, P. (1997). Geostatistics for natural resources evaluation. New York:
Oxford University Press.
Guisan, A., Edwards, T. C., & Hastie, T. (2002). Generalized linear and generalized
additive models in studies of species distributions: Setting the scene. Ecologi-
cal Modelling, 157, 89-100.
International Organization for Standardization. (1997). Human-centered design
processes for interactive systems (No. ISO 13407). Geneva.
International Organization for Standardization. (2000). Draft business plan of ISO/
TC 211 — Geographic information/geomatics. Geneva.
Larson, B. D., & Sengupta, R. R. (2004). A spatial decision support system to iden-
tify species-specific critical habitats based on size and accessibility using U.S.
GAP data. Environmental Modelling and Software, 19(1), 7-18.
Losavio, F. (2002). Standard quality model to design software architecture. Journal
of Object Technology, 1(4), 165-178.
NASA. (2005). Applied sciences directorate — national applications. Retrieved
from http://www.asd.ssc.nasa.gov/aaps.aspx
NatureServe. (2005a). Biodiversity data model. Retrieved from http://www.nature-
serve.org/prodServices/biodatamodel.jsp
NatureServe. (2005b). Predictive range mapping. Retrieved from http://www.
natureserve.org/aboutUs/northamerica.jsp
Nikolakaki, P. (2004). A GIS site-selection process for habitat creation: Estimat-
ing connectivity of habitat patches. Landscape and Urban Planning, 68(1),
77-94.
Olson, T. G. (1996). Biodiversity and private property: Conflict or opportunity?
In W. J. Snape (Ed.), Biodiversity and the law (pp. 67–79). Washington, DC:
Island Press.
Olson, T. G., Murphy, D. D., & Thornton, R. D. (1993). The habitat transaction
method: A proposal for creating tradable credits in endangered species habitat.
In H. Fischer & W. E. Hudson (Eds.), Building economic incentives into the en-
dangered species act (pp. 27–36). Washington, DC: Defenders of Wildlife.
Open GIS Consortium, Inc. (2002). The OpenGIS Abstract Specification, Topic
12: OpenGIS Service Architecture, Version 4.3. Wayland, MA: Open GIS
Consortium, Inc.
Open GIS Consortium, Inc. (2003a). Decision support priority application theme.
Wayland, MA: Open GIS Consortium, Inc.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Olson & Hlton

Open GIS Consortium, Inc. (2003b). OpenGIS reference model. Wayland, MA:
Open GIS Consortium, Inc.
Open Source Initiative. (2005). Retrieved from http://www.opensource.org/index.
php
Scott, J. M., & Jennings, M. D. (1998). Large-area mapping of biodiversity. Annals
of the Missouri Botanical Garden, 85(1), 34–47.
Scott, M. J., Heglund, P. J., & Morrison, M. L. (2002). Predicting species occur-
rences: Issues of accuracy and scale. Covelo, CA: Island Press.
Sohn, D., & Cohen, M. (1996). From smokestacks to species: Extending the trad-
able permit approach from air pollution to habitat conservation. Stanford
Environmental Law Journal, 15(2), 405–451.
Store, R., & Jokimäki, J. (2003). A GIS-based multi-scale approach to habitat suit-
ability modeling. Ecological Modelling, 169, 1-15.
The GNU Project. (2005). Retrieved from http://www.gnu.org

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Chapter.X

GIS-Based.Site.Suitability.
Decision.Support.System.
for.Planning Confined
Animal.Feeding.
Operations.in.Iowa
Ramanathan Sugumaran, Unversty of Northern Iowa, USA
Bran Bakker, Aeral Servces, Inc., USA

Abstract

Confined animal feeding operations (CAFOs) are becoming increasingly common

on the Iowa landscape. They produce large amounts of byproducts that can cause
a threat to the surrounding environment of the areas of production. Thus, there is a
need for careful planning, particularly the selection of suitable locations for future
CAFO development. In addition to Iowa state regulations, selection of locations
for CAFOs require multiple parameters like, locations of manmade structures that
include roads, residences, businesses, wells, and so forth, and location of natural
features such as rivers and lakes. Currently, locations for CAFOs are chosen
manually using paper hard copies with producer’s preference and are restricted
by state of Iowa guidelines. There is no decision support system available to aid in

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Sugumaran & Bakker

selecting an appropriate location for the development of future CAFO structures.

The purpose of this chapter is to demonstrate how a decision support tool was de-
veloped to aid CAFO managers and producers in selecting appropriate locations
for animal confinements using geographic information system (GIS) technology
and CAFO regulations in Iowa.

Background

In the past half century, the production of livestock has gone from being an activity
carried out on small family farms, to an industrialized activity completed by very
large corporations using specialized labor and confinement facilities. The number
of farms in America has been reduced drastically from over 5,500,000 in 1950, to
just over 2,000,000 at the end of the 1990s (Hallberg, 2001). While the number
of farms decreased, production per farm increased to keep up with food demands.
Much animal production is now carried out in CAFO facilities. These confinement
facilities are generally very large and are often hundreds of feet long and can con-
tain hundreds or even thousands of animals. It is commonly known throughout the
Midwest United States that these large factory-style CAFOs create many benefits
to local economies while also creating many threats to the environment where they
are located. Rural areas are in desperate need of the economic benefits that animal
feeding operations bring to local communities. Research shows that an estimated
89,000 jobs in the state of Iowa are directly or indirectly related to the hog industry,
and an estimated $700 million of income is earned by farmers and workers directly
employed in the hog production and processing industry, while an additional $1.4
billion of personal income is indirectly linked to Iowa’s hog industry (Otto, Orazam,
& Huffman, 1998).
Despite the positive economic benefits created by CAFOs, many environmental
concerns have been raised. Many individuals are concerned about air and water
pollution created by animal confinements. Several studies have examined and
reported the environmental impacts of livestock production on neighboring com-
munities (Abeles-Allison, 1990; Kim, Goldsmith, & Thomas, 2005; Taff, Tiffany,
& Weisberg, 1996). Researchers have also found various problems associated with
air pollution caused by CAFOs (Wing & Wolf, 1999). Their results show that per-
sons living near the hog farms where smell is noticeable had increased feelings of
tension, depression, anger, fatigue, and confusion as compared to a control group
(Schiffman, Miller, Suggs, & Graham, 1994). Physical symptoms have also been
linked to air pollution from CAFOs, showing that persons living near large hog
farms suffer from significantly higher levels of upper respiratory and gastrointestinal
ailments than people living near large cattle farms or in non-livestock farming areas
(Wing & Wolf, 1999). Surface and ground water is also impacted by the CAFO.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

From 1995 to 1998, there were at least 1,000 spills or other pollution incidents at
livestock feedlots in ten states, and 200 manure-related fish kills that resulted in
the death of thirteen million fish (Frey, Hopper, & Fredregill, 2000). Studies from
Illinois showed that the streptococcus bacteria were detected at least once from
groundwater wells located below manure pits (Krapac et al., 2002).
These above-mentioned issues clearly show the need for proper planning of CAFO
locations. The result of poor planning of CAFO locations was most strongly dem-
onstrated in September of 1999, when Hurricane Floyd produced nearly 20 inches
of rain over North Carolina, the second largest hog producing state in the United
States. Fifty CAFOs were flooded, and thirty thousand swine and nearly 3 million
poultry were reported dead. The catastrophe drew attention to 185 animal operations
that were built within a 100-year floodplain prior to 1997 rules which outlawed such
construction (“Hurricane Floyd,” 2000).
Since CAFO facilities create benefits and risks in the locations where they are
constructed, careful and intelligent planning of the facilities is the only solution
that protects the environment, local economies, and a farmer’s right to produce.
Selection of locations for CAFO require integration and analyses of multiple pa-
rameters like, manmade structures including roads, residences, businesses, wells,
and so forth, and also requires natural features like rivers and lakes. Recently, GIS
proved to be an excellent decision support tool to help evaluate general land suit-
ability by integrating multiple location information. GIS is an information system
that is used to manipulate, analyze, and output geospatial data, in order to support
decision-making for planning and management purposes (Warren County, 2005).
Only a couple researchers have attempted to show the importance of GIS in CAFO
site selection. These studies are explained in the next section.

Site Suitability Studies for Confined Animal

Feeding. Operations

An early decision support system for livestock facility planning was created by Jain,
Tim, and Jolley (1995) at Iowa State University (ISU). This system was created using
ARC/INFO to aid in determining land areas that are suitable for various livestock
production strategies. Factors taken into account by the decision support system
(DSS) included environmental, aesthetic, and economic constraints in the study area
of the Lake Icaria watershed in southern Iowa. The ISU system used a point analysis
technique to determine how far a set of acceptable solutions, weighted according
to a specified criterion, deviated from a set of ideal solutions or feasible regions.
Since the ideal solution is rarely available, the technique involves the use of a best
compromise solution that minimizes the distance from the theoretical ideal solution

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

(Jain, Tim, & Jolley, 1995). Parameters taken into effect in the model included: soil
drainage class, stream proximity, soil permeability, road proximity, land slope, and
aspect. When run, the ISU model selected parcels of land which were best suited
for various livestock production strategies including small, medium, and large beef
and hog operations. The ISU study did not take into account current state or federal
regulation placed on confined feeding operations. It also focused on physical land
features, and not manmade features. Furthermore, instead of focusing simply on
CAFOs, it looked at animal production on a broader level, including small, medium,
and large livestock production operations. It did not focus on large confined animal
feeding operations.
GIS tools for siting animal feeding operations have also been developed by Covington,
Kloot, and Atkins (2000) in conjunction with the USDA and University of South
Carolina. The system can be used to identify whether an area meets legal require-
ments for siting a CAFO or for manure application according to South Carolina
regulations. The system can also qualify legally-sited areas based on environmental
factors, and can assist in managing the relocation of facility byproducts. The South
Carolina DSS was created using ArcView GIS. Instead of looking at area-wide suit-
ability, the DSS tool is designed to evaluate a specific location. It primarily takes into
account setback distances which are individually entered by the user. The setback
distances are then used to create a grid for the selection of CAFO locations, and
waste utilization areas. The South Carolina DSS includes an economic component,
allowing the user to input information about animal weight and waste management.
It is designed to help producers make economic decisions about livestock produc-
tion. A later version was developed to help in manure management, specifically to
aid in determining field areas, land-application areas, and volume of manure that
could be applied to a location (Covington, Kloot, & Taduri, 2000).
A study by Worley, Rupert, and Risse (2001) addressed the effects of property line
and water buffers on land availability for animal feeding operations. Using GIS, the
authors found that available land was decreased to 63% of total land with a 100-foot
buffer, and as little as 7% of total land was available with a 500-foot buffer. The
study used a simple buffer analysis to determine where 1,000 or more animal-unit
swine production facilities could be constructed. The study assumed that a 1,000
unit swine facility requires 5 acres for the facility and 100 acres of crop land to
effectively apply manure. Property boundaries were buffered at 30.5 and 152.5
meters, and water boundaries were buffered at 30.5, 61, and 152.5 meters. The
study made some broad assumptions, including the notion that all waste had to be
spread onto one field, and that setback distances applied to fields where application
of manure would occur. The study also did not incorporate many other objects that
must have a buffer zone under State of Iowa regulations, including homes, public
areas, and schools.
Each reviewed system failed to fully meet the current needs of a producer selecting
a location for a new CAFO facility. While the ISU system failed to consider state

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

laws when selecting suitable locations for CAFOs, the Worley et al. (2001) study
was not designed to aid in specific site selection. The South Carolina system was
exceedingly broad and reached far beyond simple site selection. It also required user
input of setback distances. The main goal of this research is to create a GIS-based
suitability tool to aid in CAFO site selection, while also aiding in the understanding
of how CAFO sites impact neighborhoods and are impacted by state regulations.

CAFO. Decision. Support. System. Development

Study.Area.and.Data.Used

In order to understand the effects of CAFO regulation, it is important to look at a

rural and urban county to compare how the amount of urban space affects the land
available for CAFO development. It is also important to consider how the types and
amounts of natural resources affect land availability for animal feeding operations.
In this study, a case study using data from two Iowa counties: Black Hawk (an
urban county) and Grundy (a rural county) were selected. Figure 1 shows a map of
the study area. The primary data provider for both counties was the county govern-
ment itself. Other data providers include the Iowa DNR and Iowa DOT. In addition,
2004 USDA National Agriculture Imagery Program (NAIP) photography was also
used to analyze and verify the county data sources. Much of the data needed were
preprocessed due to compatibility issues between the decision support tool and the
original county data.

Decision.Support.Tools.Development

The GIS-based analysis tools were developed using ArcGIS software. ArcGIS software
is the industry standard GIS software developed by Environmental Systems Research
Institute (ESRI). ArcGIS allows users to customize the GIS interface using Visual
Basic for Applications (VBA), a simple object-oriented programming language. VBA
is a very popular programming language because of its ability to be used in rapid
development of applications. Furthermore, ESRI has written a special VBA library
of functions, called ArcObjects, which allows a user to automate commands and
functions available through the regular ArcGIS interface. By automating complex
spatial analyses, novice GIS users can complete tasks that would otherwise require
advanced GIS knowledge. In addition, the ArcGIS platform was chosen because it
is the standard GIS application used by most government agencies, such as many
of the anticipated users of this application, including county governments, USDA
service offices, and Iowa DNR.
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

Figure 1. Study area — Black Hawk and Grundy County

Analysis.Techniques

To determine the most appropriate location for animal feeding operations, two
methods, the setback analysis and multi-criteria decision analysis, were developed.
Figure 2 shows the flowchart for the overall GIS-based site suitability analysis.

Setback.Analysis

Prior to selecting a suitable location for a CAFO, a developer must understand which
locations would be legally eligible for a confinement facility. As mentioned previ-
ously, CAFOs must be constructed beyond state-mandated distances from various
objects. Therefore, a logical first step to selecting a CAFO location must include a
setback analysis to determine legal locations for CAFO construction. To complete
the setback analysis, a GIS-based buffer analysis was chosen in this study. A buf-
fer is a map feature that is created at a specific distance around other features on a
map (Wyatt & Ralphs, 2003). Using buffers, an exclusionary method of analyzing

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Figure 2. Flowchart for GIS-based site suitability analysis

potential CAFO locations can be completed where locations within state-mandated

setback distances can be excluded from consideration for CAFO development.
Figure 3 shows the steps involved in completing the setback analysis to determine
locations that are legally eligible for a CAFO facility in Iowa. The analysis starts
with State’s distance requirements, followed by checking appropriate GIS layers,
creating buffers, and displaying the results.

Multi-Criteria.Decision.Analysis

Once legal locations have been determined by the setback analysis, a multi-criteria
evaluation (MCE) can then be run on the remaining locations to better understand
which legal locations are best suited for CAFO development. For this study, a simple

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
226 Sugumaran & Bakker

weighting MCE has been chosen. This weighting method was chosen because it
is simple to understand even by novice users and also widely used for the similar
type of site suitability study. For example, Malczewski (1999) used this approach
for locating a nuclear power plant. Furthermore, the simple formula is easy to
implement through VBA programming. Using formulas by Malczewski (1999) to
compare suitable locations, the simple additive weighting method evaluates each
alternative Ai by the following formula:

Figure 3. Flowchart for setback analysis

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Ai = Σj [ wj * xij]

where xij is the score of the ith alternative with respect to the jth attribute, and weight
wi is a normalized weight, so that Σ [wj] = 1.
Two simple weighting methods: (a) rank sum and (b) ratio estimation procedure
have been selected to determine the weights to be applied to the simple additive

Figure 4. Flowchart for multi-criteria analysis

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

weighting evaluation. The rank sum weighting method assigns weights by ranking
each criterion in order by a decision-maker’s preference. Using rank sum, a user
would select the most important criteria, followed by the second most important
criteria, and so on, until all criteria under consideration has been selected. The weight
is then calculated by the following formula:

Wj = n-rj + 1/Σ (n-rk + 1)

where wj is the normalized weight for the jth criterion, n is the number of criteria
under consideration, and rj is the rank position of the criterion. The criterion under
consideration is normalized by the sum of all weights.
The ratio estimation weighting method allows for additional user input. To use this
method, the user must first rank the criteria in the same manner as the rank sum
method. Once desired criteria have been ranked, the user has the ability to assign the
criteria a ratio or percentage of importance, at the level he or she feels the criterion
under consideration compares to the most important criterion. Figure 4 shows a
flowchart describing the steps involved in the multi-criteria evaluation.
The analysis begins by retrieving the layers of interest, weighting method, and
weights from the user (Figure 4). Once the layers and weighting criteria have been
determined, the analysis can begin. Upon completion of the analyses, the results
are displayed as raster layers. A distance raster is simply an image comprised of
pixels, and each pixel has a value that represents the distance from the pixel to
the nearest object of interest. Once a distance raster image is created for each GIS
layer, the analysis is completed by creating a composite raster image where the
pixel value of the composite image is a function of the sum of the weighted value
of each distance raster layer, and the weighted value of each distance raster layer
is computed by multiplying the weight assigned to that distance raster layer by the
value of the distance raster.

Application. Examples

The CAFO suitability analysis tools developed in this study was tested with two types
of application examples. The first application, a county-wide suitability analysis, is
a type of analysis that might be used by a county planner who wants to understand
the general picture of suitability across the county. The second application, a loca-
tion-specific analysis, is a type of analysis that a farm service organization like a
USDA Service Center might use with a producer to find a suitable location at the
farm level for a CAFO development.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

County-Wide.Suitability.Analysis

As shown in Figure 2, two types of analysis tools (setback and multi-criteria) need
to be run to determine the availability of suitable lands for CAFO development.

Setback.Analysis

The setback analysis user interface developed in this study is given in Figure 5,
and that can be launched from ArcMap software. A setback analysis uses various

Figure 5. Legal locations setback analysis user interface

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Sugumaran & Bakker

CAFO facility types and numbers of animal units. Firstly, the user must select the
type of CAFO structure to be constructed. Five types of structures are listed in the
drop-down menu corresponding with state regulations. Secondly, the user enters the
number of animal units (AU) to be housed in the CAFO facility. Once both drop-
down boxes have been populated, the setback analysis form populates all remaining
setback distance textboxes based on the state legal requirements (Figure 5). In the
final choice, the user must decide whether to consider locations within incorporated
areas for development or not. In general, the user will not be able to construct a
CAFO facility within incorporated areas due to zoning laws; however, in the rare
case a person would actually want to develop within an incorporated area, this tool
gives the user the power to create such an analysis. Once the decision parameters
have been entered, the user must simply select the “Show Legal CAFO Locations”
button to complete the setback analyses. The results of suitable locations will be
displayed in ArcGIS.
In this example, we ran an analysis for a very typical facility constructed by large
hog operations today with the following options: type of CAFO structure: “a formed
and covered manure storage structure”, and total animal unit capacity = “1000-
2999”. The result is shown in Figure 6. In Black Hawk County, 11% of the county
is eligible for such a development, with various opportunities for the development
throughout rural areas of the county. Grundy County has even more opportunities
to develop typical large CAFOs with 22% of the land available for development.
As with Black Hawk County, these locations are spread throughout rural areas of
the county. Table 1 shows different scenarios with the area available (in percentage)
for the CAFO development with various facility types and number of animals for
both counties.
Results from this study clearly indicate that while setback distances do strongly af-
fect land availability for CAFO facilities, Iowa law does not affect land availability
as much as other literature indicates.

Table 1. Land available for CAFO development

Options Black.Hawk..(Urban) Grundy.County.(Rural)

Earthen Lagoon,
0.4% 1.7%
3000+ Animal Units

Formed, Covered Manure Storage, 1000-

11% 22%
2999 Animal Units

Formed, Covered Manure Storage, < 500

67% 85%
Animal Units

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Figure 6. Land available for a typical large CAFO (Black Hawk County top, Grundy
County bottom)

Not suitable Suitable

Multi-Criteria.Analysis

The “Weighted Distance Analysis” tool (Figure 7) allows the user to apply weights
to individual layers. To complete this analysis, the user must begin by selecting the
layers to be included in the analysis. By double-clicking on a layer listed in the left

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

text box, the layer name disappears from the textbox, and is placed by a number
under the “Layer” column. The user can select as many layers as he or she wants
to include in the analysis. The user must select the layers of interest in a rank order,
selecting the most important layer first, then the second most important layer, and
so on until all layers of interest have been selected. Once the layers have been
selected, the user can select one of two weighting methods to weight the layers,
either the rank weight method or ratio weight method. The formulas behind these
weighting methods, discussed in the previous section, assign each layer a weight
of importance less than one, and all weights of all layers add to one.
The rank-weight method requires no input from the user. Once the user selects the
layers to be considered in the analysis in order of importance, the form automatically
calculates the weights. To run this analysis, the user must select the rank weight
tab, and click on the “Create Weighted Distance Map” button. The ratio weight
analysis requires more user input. To run this analysis, the user must enter a ratio
of less than 100 for each layer listed below the first layer, where the ratio assigned
by the user indicates the user’s preference of the first layer compared to the layer
being weighted. Once a ratio has been entered, the form automatically calculates
a normalized weight. After all layers have been normalized, the user can select the
“Create Weighted Distance Map” button to complete the analysis.

Figure 7. Weighted distance analysis user interface

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Figure 8. Ratio-weight analysis — Grundy County (right) Black Hawk County

(left)

A multi-criteria evaluation analysis was also run for both Black Hawk and Grundy
County. For this evaluation, all state-mandated setback objects are created equally
with no preference to one over another. Using the ratio estimation procedure, each
feature was given an identical ratio-weight, and the model was run. This type of
analysis might be performed to simply determine overall suitability of a county
for CAFO development. An official might use results from this type of analysis to
encourage a producer to build in one part of a county over another. Figure 8 shows

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

Figure 9. Aerial photograph of Porkhaven CAFO and neighboring residences

(Source: Iowa Geographic Map Server, 2004)

the results of that type of multi-criteria analysis for both Black Hawk and Grundy
County, respectively. In the result map, areas of lighter colors are more suitable and
areas of darker colors are less suitable for animal confinements. In Black Hawk
County, the best locations for CAFOs, determined by the weighted distance analysis,
are found primarily in the southeastern and north central parts of the county (Figure
8). The streets of urban communities stand out on top of areas represented by dark
color. The darker colors indicate the lack of suitability for animal confinements in
and around the urban areas. Strips of lighter tones between the urban communities
indicate areas of suitability within the counties.
Similarly, Grundy County results indicate areas of greater and lesser suitability to
animal confinements. Northern Grundy County, with its lack of urban areas seems
to be most suited for CAFO development, as well as a corridor through east central
Grundy County. The especially dark tones in southern Grundy County seem to
indicate higher densities of residences in the Beaman/Conrad region of southern
Grundy County.

Location-Specific Application Example

One location in Black Hawk County was chosen as a case study to show how the
CAFO tools could be used in choosing a site for an animal feeding operation at
farm level. The analysis completed in the case study would be similar to the type of

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Figure 10. Results of a setback distance analysis

analysis a farm services center might complete to help a producer choose the best
location for a new CAFO. The chosen location for this case study has an existing
animal confinement in the region. This analysis has been able to show how the lo-
cation of the confinement compares with what the CAFO tools would recommend
as a best location. In Black Hawk County, a facility registered as Porkhaven Farm,
LLP-Beck Site was chosen for the analysis. It is registered for 1,600 animal units,
and is considered a Swine Grow to Finish Facility. It is located approximately 5
miles south-southeast of Hudson, Iowa, in Eagle Township, section 7, in southern
Black Hawk County. The Porkhaven facility (Figure 9) is located over a half mile
from its nearest neighboring residence, and over one and a half miles from all other
objects requiring setback distances. The nearest water resource is a small stream
over one and a half miles to the northeast. The facility lies approximately 800 feet
beyond the nearest setback buffer (Figure 10).
When the ratio estimation procedure is run using equal weighting for all objects, the
results clearly indicate that the facility has been placed in an ideal location (Figure
11). The darker colors indicate areas that are less suitable for animal confinements.
Results from a combined setback analysis and multi-criteria analysis also indicate
that the planner involved in identifying this CAFO facility did an excellent job
in locating the facility in a place which was not only legal, but was truly the best
location in the overall area.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

Figure 11. Results of a ratio estimation analysis of area surrounding Porkhaven

CAFO facility

Conclusion. and. Future. Directions

The main goal of this research was to develop a customized GIS-based suitability
tool to aid in CAFO site selection, while also aiding in the understanding of how
CAFO sites impact neighborhoods and are impacted by state regulations. This proj-
ect has used the ArcGIS software and customization through VBA to complete the
research objectives. The primary analysis tools created for this application include
a setback analysis, and a multi-criteria analysis. The setback analysis tool aids in
determining legal locations for CAFO facilities. It uses current setback distance
requirements mandated by the state of Iowa. The multi-criteria analysis tool aids
a producer in determining the best possible location for an animal confinement. It
allows a user to develop different scenarios by assigning weights to multiple criteria
to determine an optimal location. Each analysis tool has been tested using real data
from the county governments. Results from the tests show how the tools developed
can be used in determining suitable locations for animal confinements. The results
show how setback distances limit the amount of land available for animal confine-
ments, and also indicate that the amount of land available for a confinement facil-
ity is more limited in urban counties as compared to rural counties. The research

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

clearly showed the usefulness of GIS-based decision support systems in planning

CAFO locations.
Future direction would be the migration of this type of system into a Web-based en-
vironment to allow for broader access to the decision support tools. This application
has shown how desktop GIS packages can be used to implement sophisticated models
for site suitability studies relating to animal confinements. While this application
has great potential for use in farm service organizations, most livestock producers
do not have regular access to desktop GIS packages; therefore, an Internet-GIS
based environment would lend itself well for the development of a decision sup-
port system for finding locations for animal confinements. Furthermore, none of the
existing systems attempt to incorporate plume modeling to understand how odors
emanate from a proposed facility. The future study will integrate an odor model in
GIS-based suitability DSS in identifying CAFO ideal location.

Acknowledgment

The work was funded by the STORM Project at the University of Northern Iowa.

References

Abeles-Allison, M. (1990). An analysis of local benefits and costs associated with

hog operations in Michigan. Unpublished thesis, Department of Agricultural
Economics. Michigan State University.
Covington, E., Kloot, R. W., & Taduri, H. K. R. (2000). GIS decision support system
for animal feeding operations planning. Paper presented at the 2000 ESRI
User Conference Proceedings. Retrieved April 20, 2005, from http://gis.esri.
com/library/userconf/proc01/professional/papers/pap570/p570.htm
Frey, M., Hopper, R., & Fredregill, A. (2000). Spills and kills: Manure pollution and
America’s livestock feedlots. Retrieved January 18, 2004, from http://www.
cwn.org/docs/publications/spillkillmain.htm
Hallberg, M. C. (2001). Economic trends in U.S. agricultural and food systems
since World War II. Ames, IA: Iowa State Press.
House File 519. (1995). Iowa General Assembly. 76th General Assembly. Retrieved
February 10, 2004, from http://www.legis.state.ia.us/Legislation.html
Hurricane floyd floods farms, challenges regulators. (2000). AWARE NEWS, 5(1),
9. Retrieved January 19, 2004, from http://www.engr.uga.edu/service/exten-
sion/aware/vol5no_1.htm#_1_11

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Sugumaran & Bakker

Iowa Geographic Map Server. (2004). The USDA national agriculture imagery
program (NAIP). Retrieved April 10, 2005, from http://cairo.gis.iastate.edu/
map.html
Jain, D. K., Tim, U. S., & Jolly, R. (1995). Spatial decision support system for
planning sustainable livestock production. Computer Environment and Urban
Systems, 19(1), 57-75.
Kim, J., Goldsmith, P. D., & Thomas, M. H. (2005, May). Economic impact and social
cost of confined animal feeding operations: A comparison and compensation
analysis at the parcel level (Working paper). Champaign, IL: University of
Illinois at Urbana-Champaign. Retrieved August 15, 2006, from https://netfiles.
uiuc.edu/pgoldsmi/www/working_papers/jungik4%20.pdf
Krapac, I. G., Dey, W. S., Roy, W. R., Smyth, C. A., Storment, E., Sargent, S. L., et
al. (2002). Impact of swine manure pits on groundwater quality. Environmental
Pollution, 120(2), 475-492.
Malczewski, J. (1999). GIS and multicriteria decision analysis. New York: Wiley
Otto, D., Orazam, P., & Huffman, W. (1998). Community and economic impacts of
the Iowa hog industry. In Hays, D. (Ed.), Iowa’s pork industry — dollars and
scents (Chap. 6). Retrieved January 23, 2004, from http://www.econ.iastate.
edu/outreach/agriculture/livestock/pork_dollars_and_scents/chapter6/intro-
duction.html
Schiffman, S., Miller, E. A. S., Suggs, M. S., & Graham, B. G. (1994). The effect
of environmental odors emanating from commerical swine operations on the
mood of nearby residents. Brain Research Bulletin, 37(4), 369-375.
Taff, S. J., Tiffany, D. G., & Weisberg, W. (1996). Measured effects of feedlots on
residential property values in Minnesota: A report to the legislature (Staff Paper
Series, P96-12, 27). Minneapolis, MN: University of Minnesota Department
of Applied Economics.
Warren County, Ohio. (2005). What is GIS? Retrieved April 20, 2005, from http://
www.co.warren.oh.us/warrengis/definitions.htm
Wing, S., & Wolf, S. (1999). Intensive livestock operations, health, and quality of
lifeamong eastern North Carolina residents (Report prepared for the North
Carolina Department of Health and Human Services). Chapel Hill, NC: Uni-
versity of North Carolina at Chapel Hill.
Worley, J. W., Rupert, C., & Risse, L. M. (2001). Use of GIS to determine the effect
of property line and water buffers on land availability. Applied Engineering
in Agriculture, 17(1), 49-54.
Wyatt, P., & Ralphs, M. (2003). GIS land and property management. New York:
Spon.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 GIS-Based Sute Sutablty Decson Support System 

Section IV

Future Trends
and Technologies

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Brodersen & Nelsen

Chapter.XI

Geo-Communication,.
Web-Services,.and.Spatial.
Data.Infrastructure:.
An.Approach.Through.
Conceptual.Models
Lars Brodersen, Aalborg Unversty, Denmark
Anders Nelsen, Natonal Survey and Cadastre, Denmark

Abstract

The introduction of Web services as index-portals based on geo-information has

changed the conditions for both content and form of geo-communication. A high
number of players and interactions as well as a very high number of all kinds of
information and combinations of these characterise Web services, where maps are
only a part of the whole. This chapter discusses the relations between the different
components of SDI and geo-communication as well as the impact thereof. Discussed
is also a model for the organization of the passive components of the infrastructure;
that is, legislation, collaboration, standards, models, specifications, Web services,
and finally the information. Awareness of the complexity is necessary, and structure
is needed to make it possible for the geo-information community to pull together in
the same direction. Modern Web-based geo-communication and its infrastructure
looks very complex, and it will get even more complex. Therefore, there is a strong
need for theories and models that can describe this complex Web in the SDI and

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

geo-communication consisting of active components, passive components, users,

and information in order to make it possible to handle the complexity and to give
the necessary framework.

The. Chapter’s. Delimination

The major concern of this chapter is the requirements driven or user driven devel-
opment of SDI and geo-communication. Most GIS-, cartography- and SDI-litera-
ture lacks theories, models, and methodology for the systematic user requirement
assessment, which comprises user awareness, situation awareness (task, time and
place), and capability awareness.
This chapter describes conceptual models, that is, relations on a general level. This
chapter is not a description of technical implementation methodology, that is, actual
action or prototyping. The ideas presented in the chapter are of speculative nature.
They are mainly based on the author’s joint experience and empery from twenty-
five respectively thirty-seven years in the business. The theoretical aspects in the
chapter, particularly those regarding the geo-communication, are mainly based on
C. S. Peirce’s theories on semiotics and phenomenology.

Introduction

The role of geo-information and the distribution of geo-information have changed

dramatically since the introduction of Web services on the Internet. In the framework
of Web services, maps should be seen as a part of an index to further geo-informa-
tion. Maps are no longer an aim in themselves. In this context, Web services perform
the function as index-portals to further information. This index-function is based
on geo-information, for example, maps.
The introduction of Web services as index-portals based on geo-information has
changed the conditions for both content and form of geo-communication. A high
number of players and interactions (as well as a very high number of all kinds of
information and combinations of these) characterise Web services, where maps are
only a part of the whole. These new conditions demand new ways of modelling the
processes leading to geo-communication.
See Figure 1. A high number of players and interactions (as well as a very high
number of all kinds of information and combinations of these) characterise modern
geo-communication, where maps are only a part of the whole; these new conditions
demand new ways of modelling the processes leading to geo-communication.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Brodersen & Nelsen

Figure 1. Web services, a new geocommunication

What. is. Geo-Communication

The purpose of any communication is to conduct the behaviour of the user. This is
done by submitting detailed and precise information, on the basis of which the user
may act. Decision and action is conditioned by this supply of information, and the
following mental connection and integration with previous experience in the user’s
mind. In order that the producer may communicate the necessary information to
the user, the producer must be able to analyse the phenomenon of which the com-
munication consists and be able to describe the result of this analysis in detail. The
purpose of the analysis is to select that kind of information that forms a reasonable
basis for decision and the according action (Brodersen, 2005). Transmission of this
kind of information, in writing or in graphical form, is part of geo-communication.
All together geo-communication describes a value-chain from reality to decision
and the according action.
The user may want a basis for making decisions on a possible trip, that is, a sug-
gestion of an itinerary. For this purpose, the user starts a Web service intended for
this use. The user types the start point and the end point of the trip, date and time,
and after a short time, he will receive a number of proposals for the itinerary. On
this basis, he will be able to make the decision, “Yes” or “No,” to travel.
Figure 2 illustrates the above example of travel planning. The figure consist of the
passive elements of geo-communication (plus a user), that is, geo-information and
spatial data infrastructure (SDI). All processes shown in the illustration can be it-
erative. The illustration can be seen as a longitudinal section of the overall process.
Compare also Figure 5 showing the cross section of the geo-communication.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

Figure 2. Longitudinal geoinformation

See Figure 2. The elements of a geo-communication illustrating the example of

planning a trip; the user is looking for information enabling her to decide whether
to take the trip or not. The primary problem for the producer is to catch this problem
and to deliver the exact information having this meaning. The secondary problem
for the producer is to master the complex network of processes and their mutual
dependencies. All processes may be iterative. The illustration can be seen as a lon-
gitudinal section of the overall process. Compare also Figure 5 showing the cross
section of the geo-communication.
A Web service application in many ways can be considered like a “black box”: Users
don’t have to care about what happen behind the interface. Based on the business
logic of the service a number of requests are sent to certain service-to-service database
interfaces demanding certain geo-information, such as timetables, maps, road work,
and so forth. This geo-information can be dynamic as well as static. The dynamic
geo-information could be, for example, traffic conditions, weather conditions, road
works, and so forth. Depending on the value of the dynamic geo-information, the
need for certain static geo-information may vary. Static geo-information is, for
example, maps, routes, addresses, timetables, and so forth.
The information retrieval is triggered by the initial user action. Another set of busi-
ness rules is then used by the service to produce the desired output in the form of
tables, text, and graphics, which together express the itinerary to be presented via
the web application. It is important to note that the user does not ask for a certain
timetable or a certain map, but only for the meaning of these in relation to the trip
she wants to make. The meaning for the user is to have a basis on which first to
make the decision to travel or not to travel, and if she decides to travel, then to
know how to do it.
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Brodersen & Nelsen

By means of other algorithms, these different types of geo-information are con-

trolled against one another, resulting in a number of tables, text, and graphics, which
together acts as the wanted itinerary. It is important to note that the user does not
ask for a certain timetable or a certain map, but only for the meaning of these in
relation to the trip she wants to make. The meaning for the user is to have a basis
on which first to make the decision to travel or not to travel, and if she decides to
travel, then to know how to do it.

The. Structure. of. Geo-Communication

Compared to the “good old days” when maps were maps, new crucial aspects have
come into play. One new aspect is the fact that the service providers have become a
part of the geo-communication process with influence on the content. Another aspect
is that several new producers have become active on the market. A third aspect is
that there no longer is a given relation between producer and end user, as it was
the case in the “good old days.” A fourth aspect of the new way of modelling is the
distinction between active and passive components in the spatial data infrastructure.
All four aspects will be discussed in the following.
Example: Up until approximately 1990, there was in Denmark one producer of geo-
information. The Geodetic Institute’s production was linked strongly to the demands
of the military, that is, the military’s definition of meaning of the geo-information
was the basis for all production. More or less everybody else in the Danish society
had to be satisfied with those military-oriented maps and other products. This situ-
ation was due to the fact that the Geodetic Institute had a monopoly-like position in
map-production, partly supported by legislation, and partly because of the extremely
expensive means of production.

Figure 3. Up to approximately 1990 in Denmark: One producer, one definition of

meaning, and several users

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

Figure 4. A simple version of a modern Web-based geo-communication commu-

nity

Since 1990, things have changed dramatically. The monopoly has been removed.
The law concerned was cancelled, and the costs of the means of production were
reduced dramatically. Several producers are now active in producing geo-informa-
tion, and several Web service providers are carrying out the transmission of this
geo-information to a huge number of users. Not only have a lot of producers and
service providers become active. The most important aspect of the new era is that
there are thousands of new users.
In the “good old days,” the production of maps was extremely expensive, and therefore
it was kept for the few. Today the Web-based infrastructure eases the admission to
information so that nearly everybody can take part. This higher number of produc-
ers, service providers and users can be put into a diagram similar to the diagram in
Figure 3 showing a part of the complexity of the new situation; see Figure 4.
The Web in Figure 4 illustrates a simple version of a modern Web-based geo-com-
munication community; today there are several producers of geo-information and
several Web service providers carrying out geo-communication. On top of the SDI
a huge number of users are all trying to find that particular meaning of the geo-in-
formation that satisfy their particular needs. Therefore the producers have to cope
with several, different types of definitions of meaning.
The first point to be made here is that the world of geo-communication has become
extremely complex because of the higher number of producers, service providers,
and users. The Web in the Figure 4 illustrates a simple version of a Web-based geo-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Brodersen & Nelsen

Figure 5. Modern geo-communication should be seen as the combination of GIS

plus other elements; the diagram can be seen as a cross section of geo-communica-
tion, compare also Figure 2 with the longitudinal section of the overall process of
geo-communication.

communication community. The second point to be made here is that there is no

longer only one definition of meaning with which everybody has to be satisfied. All
users have the possibility to find a producer or a service provider who accepts to
take care of that particular definition of meaning demanded by that particular user.
The producers have lost control of the users’ behaviour.
Now, the good question is what to do about this increasing complexity? It reflects
a complete new way of dealing with geo-communication and geo-information, and
therefore new theories and models are needed. The way forward is to create models
and theories that describe the new structure in geo-communication and SDI. The
old theories and models are no longer sufficient.
Geographic information systems (GIS) is often understood as the combination of
software, data, and methods, as the red dotted line in Figure 5 shows. This might well
be so. But maybe this is no longer sufficient to be able to understand the complex
era of modern geo-communication based on SDI. Figure 2 can be seen as a longi-
tudinal section of the geo-communication process. The accompanying cross-section
of the geo-communication looks like the diagram in Figure 5. Geo-communication
should be seen and understood as consisting of GIS plus other elements, as Figure
5 illustrates.

Spatial. Data. Infrastructure. (SDI)

Spatial data infrastructure (SDI) is the framework for geo-communication. Where

geo-communication is the actual transmission of geo-information in writing or in

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

Figure 6. The combination of public services and the responsible organisations can
be seen as the infrastructure of a society; SDI is concerned with geo-communication
and the respective responsible organisations

graphical form, SDI is the framework that makes it possible to carry out geo-com-
munication. Therefore some kind of overlapping can be identified. The geo-com-
munication view-point is concerned with the transmission of the meaning of the
geo-information. The SDI view-point is concerned with the organisation and the
services, systems, and so forth, that make the transmission of geo-information
possible.
From an organisational view-point SDI can be seen as the combination of organisa-
tions and public services. Infrastructure consists of:

• The basic organisation, the system according to which a company, organisa-

tion, or other body is organised at the most basic level.
• Public services or systems,.the large-scale public systems, services, and facili-
ties of a country or region that are necessary for economic activity, including
power and water supplies, public transport, telecommunications, geo-com-
munication, roads, and schools.

From an activity view-point SDI can be defined as a combination of active compo-

nents and passive components:

• The active components in SDI are those organisations that get things running.
The active components have the responsibility, and they must be active. Oth-
erwise nothing will happen.

The active components in SDI are:

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Brodersen & Nelsen

Figure 7. The active components in SDI are those organisations that get things
running, the active components have the responsibility, and they must be active; the
passive components in SDI are those documents made by the active components,
the passive components are created to get the information distributed about the
active components activities

• International organisations like UN, NATO, and so forth

• Governments
• National mapping agencies
• Standardisation bodies
• Custodians for various services
• Producers of geo-information

The passive components in SDI are those documents that the active components
have to produce to get the information about their activities distributed. These docu-
ments, the passive components, are the following, here presented in their mutual
dependency. The mutual dependency is of iterative nature. The dependencies of
the passive components go from general to concrete; that is, from legislation to
geo-information.
The passive components in SDI are (here presented in their mutual dependency):

1. Legislation Ideas, feelings

2. Collaboration (MoUs) Ideas, feelings
3. Standards Principles, rules, relations
4. Models Principles, rules, relations
5. Specifications Principles, rules, relations
6. Services Action, concrete
7. Geo-information (data) Action, concrete
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

The active components and the passive components can be combined in one illus-
tration as shown in Figure 7.

The. Passive. Components. in. SDI

The passive components are those documents containing and presenting the results
of the activities of the active components. The passive components are dependent
upon the activities in the organisations. There-fore they are called passive. These
passive components are not active in themselves.

1. The legislation is made by the organisations, the active components. The

legislation must act on a general level taking care of:
• Enacting the framework for the deeper structures of SDI
• Setting the areas of responsibilities
2. Collaboration (Memorandum of Understanding, MoU) are policy state-
ments (position papers). Collaboration is based on the framework given by
the legislation. Without this framework it is basically not possible to establish
partnerships. Collaboration must act on a general level taking care of:
• Setting the area(s) of interest

Figure 8. A few examples of standards; standardization is a necessary basis for

activity within an SDI. Standards are general and often conceptually roomy, thus
applicable to a wide range of projects and their specifications.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Brodersen & Nelsen

• Establishing operational partnerships (within the framework of legisla-

tion)
• Inclusion of services as a full palette of joint government and commercial
theatres
• Agreements upon the intention of sharing meta-information on services
and registries
3. Standards are the necessary basis for activity within an SDI. Standards are
the logic and practical conclusions of the agreements made in (2) collabora-
tion. Collaboration is a declaration of intent. It expresses a convergence of
will between the parties, indicating an intended common line of action. Stan-
dards define in greater detail, what to work on. Standards answer questions
like “Which are the components or entities of the universe of discourse that
we are going to work on?”; “How do we define our activities?” and so forth.
Standards are general and can therefore be used for several concrete projects,
where a specification is concrete and valid for one project. A few examples
of standards (of which a few are Danish standards) can be seen in Figure 8.
4. Models describe how to use certain standards for a given project. Models
bridge the gap between standards and specifications. Models describe:
• Value model:
 Identification of content
• Business process engineering:
 Information and resource flow

Figure 9. Web services bridge the gap between producer’s databases and the users;
Web services are the technology making the use of geo-information possible

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

 Requirements-driven service development

• System use case:
 Application schemas
 General feature models
• The need for specifications:
 Implementation process
5. Specifications are descriptions of what has to be done in a certain project.
A specification can, for example, specify that things have to be done in ac-
cordance with certain standards. A specification is concrete and valid for one
given project; standards are general and valid for several projects. Specifica-
tions describe rules and contents for one given project:
• Categorisations and classifications of real-world phenomena (features)
within a standardised universe of discourse
• Definitions and descriptions of attributes on the level of the classified
features
• Definitions of the information flow; sources, update strategy; components
value assessment, and so forth
• Storage and security strategies

Figure 10. Geo-information and metadata in services in a global Web; user, pro-
ducer, and service provider are all parts of the service Web, each of them carrying
out their specific task (Idea: Dave McKellar, Canada)

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Brodersen & Nelsen

• Filtering and retrieval methodologies

• Strategies for multistage representation (incl. semantically and geometrical
granularity and generalisation)
• Symbolisation strategy, design manuals, and legend drafting

The passive components numbered 1 through 5 can all be carried out on a piece of
paper; nothing practical or actual has happened until here.

6. Services are the concrete, practical set-up of the passive components num-
bered 1 through 5. Services establish the technology; that is, the software,
the hardware, the user-interfaces, and so forth. Compare also Figure 2, the
longitudinal section of the geo-communication, as well as Figure 5, the cross
section of the geo-communication.
7. Metadata.and.information is the “fuel” to put into the machinery (the ser-
vices) once the services have been created. Metadata and Information is not
the technology. Metadata and information are the actual, practical, concrete
result of a certain production carried out in accordance with the characteristics
of the services, with the specification, with the model, with the standard, with
the MoU, and with the legislation.
• Metadata:
 Where to find the information (the data)
 Enables an analysis of the information’s fitness for use, the
chacteristics of the information
 Never create information (data) without metadata, and never
separate the two
• Information (data):
 Information is the fuel for value-added service
 Information (data) is not products

Conclusion

With the introduction of Web-based geo-communication, things have become most

complex, compared to the “good old days” when maps were maps. Decision and
action is conditioned by the supply of geo-information, and the following mental
connection and integration with previous experience in the user’s mind. In order that
the producer may communicate the necessary information to the user, the producer

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Geo-Communcaton, Web-Servces, and Spatal Data Infrastructure 

must be able to analyse the phenomenon of which the communication consists and
be able to describe the result of this analysis in detail. Transmission of this kind
of information, in writing or in graphical form, is part of geo-communication. Al-
together, geo-communication describes a value-chain from reality to decision and
the according action.
The basis for Web-based geo-communication is the spatial data infrastructure (SDI).
SDI consists of both active components and passive components. The active compo-
nents get things happening. The passive components are the documents describing
the results of the activities with the active components. The passive components are
the legislation, the agreements, the standards, the technology, the specifications, and
the information, which are the crucial elements of the infrastructure and with it, the
necessary basis for Web-based geo-communication. As there is a mutual dependency
between all the components, none of them can be left out. If just one component
is missing, the impact is that the geo-communication is based on a non-systematic
and non-conscious foundation.
Modern Web-based geo-communication and its infrastructure looks very complex.
That is surely true. We think that it will get even more complex. Therefore, there is
a strong need for theories and models that can describe the “Web” in order to make
it possible to handle the complexity and to give the necessary framework. There is
also a strong need for political consciousness about these things because it is from
there that the legislation comes.

References

Bernhardsen, T. (2002). Geographic information systems: An introduction. New

York: John Wiley & Sons.
Brodersen, L. (2005). Semiotik i geokommunikation — fra virkelighed til handling.
Frederikshavn, Denmark: Tankegang a·s.
Bødker, K., Kensing, F. & Simonsen, J. (2004). Participatory IT design: Designing
for business and workplace realities. Cambridge, MA: MIT Press.
Caquard, S. (2003). Internet, maps and public participation: Contemporary limits
and possibilities. In Maps and the Internet (345-358). Oxford, UK: Elsevier
Science Ltd.
Hjelmager, J., Delgado, T, Moellering, H, Cooper, A, Danko, D, Huet, M, Aalders,
H, & Martynenko, A. (2005). Developing a modelling for the spatial data
infrastructure. ICA Proceedings ICC2005.
ISO/TC211 Standardisation on Geoinformation and Geomatics. ISO-19100 series.
(n.d.). Retrieved from http://www.isotc211.org/

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Brodersen & Nelsen

Jepsen, R., Nordestgaard, A., Peterson, J. E., & Lassen, M. (2003). IT udvikling.
Frederikshavn, Denmark: Dafolo Forlag A/S.
Jiang, B. (2003). Beyond serving maps: Serving GIS functionality over the Internet.
In Maps and the Internet (147-158). Oxford, UK: Elsevier Science Ltd.
Koláčný, A. (1969). Cartographic information: A fundamental concept and term in
modern cartography. The Cartographic Journal, 6(1), 47-49.
Longley, P.A., Goodchild, M.F., Maguire, D.J., & Rhind D.W. (2005). Geographical
informations systems and science. Chichester, UK: John Wiley & Sons Ltd.
Nebert, D. D. (Ed.). (2004). Spatial data infrastructure cookbook. Global Spatial
Data Infrastructure (GSDI). Retrieved from http://www.gsdi.org/docs2004/
Cookbook/cookbookV2.0.pdf.
Peirce, C. S. (1992). The essential Peirce. Selected philosophical writings (Vol. 1).
IN, Indiana University Press.
Peirce, C. S. (1998). The essential Peirce. Selected philosophical writings (Vol. 2).
IN, Indiana University Press.
Percivall, G. (Ed.). (n.d.). OGC reference model (Ref. No. OGC 03-040 Version:
0.1.3),Open Geospatial Consortium, Inc.
Peterson, M. P. (2003). Maps and the Internet: An introduction. In Maps and the
Internet (1-16). Oxford, UK: Elsevier Science Ltd.
Peterson, M. P. (2003). Foundations of research in Internet cartography. In Maps
and the Internet (437-446). Oxford, UK: Elsevier Science Ltd.
Østergaard, M., & Olesen, J.D. (2004). Digital forkalkning. Frederikshavn, Den-
mark: Dafolo Forlag A/S.

Further. Reading

Brodersen (2005) is an attempt to develop a theory and models for a systematic

user requirements assessment. It is based on Charles Sanders Peirce’s theories on
semiotics and phenomenology (Peirce, 1992; Peirce, 1998).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

Chapter.XII

A.Data.Visualization.and.
Interpretation.System.for.
Sensor.Networks
Fengxan Fan, Kunmng Unversty, People’s Republc of Chna

Abstract

With the increase in applications for sensor networks, data manipulation and
representation have become a crucial component of sensor networks. This chapter
explores an implementation to process and interpret the data gathered by sensor
networks. In a project supported by SensIT program at DARPA, we have built wireless
sensor networks deployed to monitor rare plants or other endangered species. The
environmental data, such as temperature, rainfall, and sunlight, around the plants
are sent by the wireless sensor networks to a base station. The system presented
in this chapter combines database management technology, geographic informa-
tion system, and Web development technology to visualize the data gathered by the
wireless sensor networks. The integration of our data visualization tools and the
online collaborative discussion environment makes the system useful to different
communities of potential users.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Fan

Introduction

Of all the global problems in the biosphere we confront today, few would argue
that the extinction of species and destruction of ecosystems have the most serious
consequences, and they are irreversible. Worldwide, the preservation of rare spe-
cies presents a major challenge. In Hawaii, there are numerous species of plants
and animals. Many of them are found only in Hawaii and are currently threatened
or endangered.
In order to monitor the ecological environment and events around rare plants, the
Pods project at the University of Hawaii has started to build wireless ad-hoc sensor
networks.(Biagioni & Bridges, 2002). A sensor network is a computer network made
up of many spatially-distributed sensors which are used to monitor conditions, such
as temperature, sound, vibration, pressure, motion, or pollutants. These sensors are
usually small and inexpensive, so they can be deployed in large numbers. In a wire-
less ad hoc sensor network, the sensor nodes are self-contained units consisting of a
battery, radio, sensors, and other accessories. The nodes self-organize their networks,
rather than having a pre-programmed network topology. Every node in this system
can transmit data of its own and also forward data from other nodes (Bose, Morin,
Stojmenovic, & Urrutia, 2001; Nagar & Biagioni, 2002). In our project, we call
these network nodes pods. Each pod contains a micro-computer which is needed for

Figure 1. A Silene Hawaiiensis plant in flower

Figure 1. A Silene Hawaiiensis plant in flower

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

collecting and transferring the weather data, micro-sensors, and other accessories.
Currently the pod is designed to measure sunlight, temperature, wind, and rainfall.
Some pods are also equipped to take high-resolution images of the plants periodi-
cally. In addition, the pod is designed and constructed to be inexpensive and easily
camouflaged to avoid damage by curious visitors. The pods are deployed every few
hundred feet, thus form a wireless ad hoc sensor network. A new wireless routing
protocol (named multi-path on-demand routing protocol [MOR] has been designed
for the network to provide energy conservation and routing efficiency. This network
constitutes a monitoring system for scientists to observe the rare plants. On the Big
Island of Hawaii, we have already made preliminary deployments of pods to moni-
tor a rare plant species, Silene Hawaiiensis. Figure 1 is the picture of this rare plant
which was taken automatically by a pod sensor, and is believed to be the first ever
picture of a Silene Hawaiiensis in flower in the wild.
In this wireless ad hoc sensor network system, the collected data and images are
transferred from one pod to another. They eventually reach a special pod — the
base station. At the base station, the data are stored for further manipulation and
accessible via the Internet.
It needs to be pointed out that field sites where the rare plants live are sometimes
in harsh environmental condition or in remote areas. With the help of the data sent
back by the wireless sensor network, the ecologists and botanists can observe the
plants and their environmental conditions from the Internet, without disturbing the
site or unnecessarily attracting attention to the endangered species. Then they can
analyze and understand the reasons why the rare plants survive or disappear. In ad-
dition, the data transmission is near real-time, so the observers can decide whether
the situation needs a site visit to the rare plants.
The gathered data which are sent back by our wireless sensor networks are stored
in a database. Because the environmental weather data are recorded every few
minutes, manipulating and translating the vast amount of data is crucial to the end
users. Hence, we are developing an information interpretation system for the sen-
sor networks.
The objective of this information interpretation system is to convert raw climate
and weather data into visual formats that can be easily understood by people. The
system also provides an environment on the Internet for people to access the data
and to observe the area in which the sensor networks are deployed. In this system,
users can get current (near real-time) or historical views of environmental data and
thus derive conclusions based on substantial understanding of the available data.
We also expect that people can form online communities to exchange their conclu-
sions about their observations and to discuss their points of view interactively in
this system. By means of these features, we can fulfill our ultimate goal: not only
to gather the data from the area in which sensor networks are deployed, but also to
convey and translate them for scientists to analyze, interpret, and discuss.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Fan

In the development of this information interpretation system, we adopted and com-

bined several technologies: database management systems, geographic information
systems, dynamic Web programming, and human-computer interactive design
technology. One challenge we encountered is how to display the distribution of the
data attributes in a real-world map in ways that would be intuitive and meaningful.
After several trials and failures, we selected a geographic information system.(GIS)
as the main platform to visualize the data. Another challenge was how to handle
the human- computer interaction when people are accessing the data display page
on the Internet. To address this challenge we have cooperated with a human-com-
puter interactive learning project and focused on developing a collaborative online
discussion environment for the end users, scientists, students, or other people who
are interested in environmental monitoring and conservation.
This chapter discusses the above technologies in more detail below, as follows. First,
we focus on how we generate maps displaying the distribution of data attributes by
means of information visualization techniques. Then we discuss how to apply the
technology of usability engineering to develop an asynchronous interactive discus-
sion system. Because of space limitations, we focus our description on the key ideas
and technologies we successfully applied in our project.

Data. Visualization

In order to pursue the goal of providing the information for people to review and
examine, we applied to our information interpretation process the technology of
data visualization, in which visual features such as shapes and colors can be used
to code different attributes of the data. We need a software platform to execute this
function, so we selected the geographic resources analysis support system (GRASS)
geographic information system. We have used the global positioning system (GPS)
to collect the geographic position coordinates — longitude and latitude — for each
pod. This makes the application of GIS technology in our information interpretation
system possible, and thus becomes an innovative feature for this aspect of usage in
which we have combined GIS with sensor networks.
Figure 2 and Figure 3 are an example of the resulting weather data distribution map
accompanied with the appropriate legend.
This map is generated based on the data for one day which are sent back by the Pods
sensor networks in the area of the University of Hawaii at Manoa. The map of Manoa
is the background (the background map in Figure 3 comes from the Tiger mapping
service of the U.S. Census Bureau). In this map we use different colors to represent
the different levels of temperature and sunlight in the areas being monitored. The
rainfall is represented by rain drops with different densities of drops presenting

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

Figure 2. Legend for the weather data distribution map

LIGHT

Low Medium High

Low 1 2 3

TEMPERATURE Medium 4 5 6

High 7 8 9

the levels of rainfall. The map is intuitive and easy to understand. The temperature
increases from SW to NE. The central area is the sunniest, but also has some rain.
Most of the rainfall is measured in the SE corner.

Figure 3. One example map displaying the weather data spatial distribution which
is generated from our information interpretation system

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Fan

The level of the data is determined by comparing the data value at a specific pod
with the average value over the entire area under observation. Standard deviation
has been used as the threshold while comparing the different levels of the weather
data.
The maps can also be generated based on monthly and yearly statistical data. In
addition, if the observer wants to view the data for a specific time, the system can
provide the near real-time monitoring maps to the user without an obvious delay.
These features satisfy the diverse requirements from the observers or viewers.

The.Application.of.GRASS.GIS.Technology.in.the.Informa-
tion.Interpretation.System.

To develop the data visualization map using GRASS, we have used some particular
techniques, stated as follows:

•. Background.map.importation.and.rectification: While adopting GIS technol-

ogy for our project, we need to import real-world maps containing the locations
of the deployed pods as the background of the resulting weather distribution
maps. We also need to do image processing on these maps to convert the maps
to the GRASS GIS data format — raster file.

The imported maps also need to be rectified by transforming the coordinate system
to a standard geographical coordinate system, for example, the UTM coordinate
system. This is accomplished by adopting appropriate data manipulation models
provided by GRASS.
Once the rectified map is imported into GRASS, we can place the different pods at
the appropriate locations. The geographic positions of pods are obtained from GPS.
In Figure 3, we can see that four pods, labeled with yellow location names (uhpress,
stjohn, labschool, and hig) have been located on this map.

•. The.interface.between.the.PostgreSQL.database.and.the.GRASS.GIS:
Since we are dealing with a large amount of data, we use a database for data
storage. The data stored in the database includes rainfall, temperature, and
light level which are gathered from the areas under observation.

The existing interface between the PostgreSQL database and GRASS is not appli-
cable to our system, because it does not support our color coding scheme which is
explained in the following section. In order to solve this problem, we have developed

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

an interface to meet our requirements on the system. This interface is used to transfer
the data from the PostgreSQL database into the GRASS GIS for processing, and to
convert the data from the database format into category values that are suitable for
GRASS data manipulation. In addition, because only snapshots in time are available,
the aggregated data, such as average readings, should be calculated in some way.
Therefore, this interface has some special functions. It can access the database table
to retrieve the desired weather data collected by pods. It can also implement the
color coding scheme to convert the raw data to category values according to the
levels (low, medium, and high) when compared with the average value of the data
gathered by all pods in the entire area.

The.Color.Coding.Scheme.and.Voronoi.Diagrams

In this system, we have designed a color coding scheme for the weather data dis-
play map. The map is partitioned into areas, with each area holding a unique color
to represent the level of the weather data. The algorithm for generating Voronoi
diagram is used to divide the map into different portions.

•. The.color.coding.scheme:.The purpose of the color coding scheme is to use

different colors to represent the levels of attribute data such as temperature,
light, and rainfall. However, the data collected by the sensors are independent
of each other. This means we need to find a way to combine them in one dia-
gram and make it meaningful and intuitive to viewers.

The goal of information design is to help users perceive, interpret, and make sense of
what is happening (Rosson & John, 2002). In order to pursue this goal, we designed
a data representation scheme with intuition and perception as the main concerns. For
example, we use bright color to represent areas where the sun shines more brightly,
and we use colors such as blue for cold and red for warm temperatures. This is
strongly related to people’s perception and expectations, and it gives a good context
to interpret the displayed information. Over several iterations, we carefully designed
the color coding scheme according to people’s intuition. It is easy for viewers to
understand what is going on in that area with the colored display.
The color coding scheme is implemented by the interface between the PostgreSQL
database and the GRASS GIS. In the scheme, we use two bits to represent the
levels of temperature and sunlight. Therefore, binary integer 00 means the value
is medium. Integers 01 and 10 represent the value low and high respectively. The
combination of temperature and sunlight is imported to GRASS as category values
which are used to render the map with different colors. Table 1 shows the scheme

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Fan

in details. This table is the foundation of Figure 2 displaying the legend of the
rendered resulting map.
Some parameters are more easily represented using shapes and symbols than using
colors. In this case, as much as possible we use familiar shapes; for example, we
use rain drops to indicate the area where it is raining. Hence, by color coding and
shape representation we can efficiently convey detailed information to people who
wish to observe or examine the environmental conditions on the site of their areas
of interest.
We also apply two bits to represent the rainfall. But this two bit integer does not
change the color scheme; rather we use shape symbols resembling rain drops to
represent rainfall. If the area has different levels of rainfall, we can present the rain
drops in different densities within different portions of the area, as is shown in
Figure 3. This two bit binary integer is also converted into a category value by the
interface between database and GIS.
The standard deviation is applied to determine the levels of the weather data value.
For instance, if the.temperature gathered by one pod is more than a standard devia-

Table 1. Color coding scheme

Category.
Temperature Sunlight Color
Value

00 Medium 00 Medium 0000 (0) Green

00 Medium 01 Low 0001 (1) Dark green

00 Medium 10 High 0010 (2) Bright green

01 Low 00 Medium 0100 (4) Blue

01 Low 01 Low 0101 (5) Dark Blue

01 Low 10 High 0110 (6) Bright Blue

10 High 00 Medium 1000 (8) Red

10 High 01 Low 1001 (9) Dark Red

10 High 10 High 1010 (10) Bright Red

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

tion higher than the average of temperature over the entire area, it can be catego-
rized as high temperature. So, when our system shows an area on the map that has
higher brightness than the average for the whole map, an observer knows that the
brighter area is at least one standard deviation brighter than what is reported by the
other points on the map. The one standard deviation threshold is the default when
the user initially accesses the map display page. The standard deviation serves as
an initial value to distinguish the pods with higher or lower values from those with
approximately the average value. Such diagrams reliably identify extremes of hot
and cold. However, users with specific goals may want to use other threshold values
rather than the standard deviation (as suggested to us by an anonymous reviewer).
In the future, we would like to modify the system so the initial presentation uses the
standard deviation for the threshold and users are able to dynamically modify these
thresholds to produce different maps. In the section “Data Accessibility” we present
an asynchronous online environment for users to input the data interactively.

•. Voronoi. diagram. generation. and. rendering:. After the data have been
categorized based on our color coding scheme and transferred from the Post-
greSQL database into GRASS, we need to divide the whole region in which
we deployed monitoring pods into different areas. Then we could render the
areas with different colors specified in our color coding scheme.

One of the main algorithms we adopted in GRASS is for generating Voronoi diagrams,
in which each polygon covers one pod. As an important geometric data structure in
geographic analysis, Voronoi diagrams are pervasive in GIS. The Voronoi diagram
partitions a given space into distinct polygons (called Thiessen polygons [Preparata
& Shamos, 1985]), and each polygon covers the area that is nearest to one pod.
Thus, this method is called nearest neighbor interpolation. In GRASS we use the
sweepline algorithm developed by Steve J. Fortune (Fortune, 1987) for generating
Voronoi Diagrams. His algorithm avoids the difficult merge step of the divide and
conquer technique and runs with complexity O(nlogn). This property is important
for relatively complicated data processing because we are sometimes dealing with
large amounts of information.
Figure 4 illustrates an example of Voronoi diagrams generated by GRASS according
to the locations of the pods. Within a Voronoi diagram, the area where monitoring
pods are deployed is divided into tiled polygons. In the GRASS GIS, we render
this diagram based on our color coding scheme. For example, the polygon holding
the pod with high temperature and medium sunlight is rendered with red color, as
is shown in the upper right corner of Figure 3.
The main purpose of our information visualization is to display the weather data on
a real-world map. As mentioned before, we have imported into GRASS an exist-
ing real-world map which covers the area in which we deployed some observing

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Fan

Figure 4. One example Voronoi diagram generated by GRASS

pods. This map is used as the background on the resulting display. So, we merge
the real-world map with the rendered Voronoi diagram. Thus, we can display the
distribution of weather attribute data in a two-dimensional space, that is, a map. This
becomes a feasible way for people to view and examine the data in an acceptable
visualization format.

Data.Accessibility.

Availability.of.the.Visualized.Data

We have provided two means for users to view the map on the Internet. One is dy-
namic map generation via a CGI program. This function satisfies the usage of viewing
a map for a specific date and time. The other way is to create maps automatically
at a particular time by using a program launching system (Linux cron). With this
system and our especially designed programs, the daily, monthly, and yearly maps
can be created and added to Web pages automatically.

•. Dynamic. generation. of. maps:. As a widely-used technology for building

dynamic Web documents, the common gateway interface (CGI) is applied to
generate maps dynamically and display them on the Internet. The user can fill
out a form online specifying the date and time for which the user intends to
observe the data, then submit the form to the Web server. The CGI program
runs through the entire data visualization process. It begins with the interface
implementation to query the database and create the site file. It also runs the
GRASS GIS to manipulate the site file, generate the Voronoi diagram, then
convert and output the resulting map to an image file which is available from

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

the Internet. Although CGI is a traditional technology for Web development,

making the CGI program perform the entire task is rather challenging. First,
the CGI program must be able to run GRASS and execute the required ma-
nipulation to the site file imported by the interface. Second, the CGI program
needs a long time to query the database because the database table holds a large
amount of data collected by pods. This situation has been improved by mov-
ing some of the calculations (for example, average, maximum, and minimum
values) from the database management system, where they are relatively slow,
to the CGI program. Therefore, the CGI program can generate a map within
an acceptable time — around thirty seconds.
•. Automatic.generation.of.maps:.The ecologists and botanists are often more
interested in the statistical data representation, such as daily, monthly, and
yearly environmental data. In order to meet this requirement we developed a
series of programs to generate those three kinds of map using the statistical
results based on the data gathered by the wireless sensor networks. We also
applied a program launching system, cron, to start the program at a specific
time. For example, for the daily maps, we can start the map generation pro-
gram at 00:00am everyday. For monthly maps, the program can be launched
at 00:00am on the first day of every month. The program can also add the
generated maps to the corresponding daily, monthly, or yearly html files.

Interactive.Data.Access.Environment

One major advantage of user interface design technology based on usability engineer-
ing over traditional displays is the possibility for dynamic redisplay. Computer-based
displays can be dynamically restructured, changed in size, filtered, and animated
(Rosson & John, 2002). This indicates the feasibility of constructing an interactive
system to provide an online environment for viewers to query the data from the
database, observe the events on different sites, and interactively discuss their points
of view. We need various technologies of Web development and usability engineer-
ing to design and implement this functionality, and to do this we have cooperated
with a research team led by Dan Suthers of the University of Hawaii.

•. Combination.of.visualization.tools.—.an.online.asynchronous.collabora-
tive.discussion.environment:.We have developed various visualization tools.
One of them emphasizes the spatial distribution using maps that we described
above. Another one shows the chronological charting of the data. In addition,
we have collected a lot of images which are taken by the cameras embedded
in pods for observing the growth of the plants. These pictures are organized in
Web pages with dynamic accessibility. All of these tools are developed for the
Web, so they are available from the Internet. But they exist independently and

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Fan

reflect the different aspects of the plants’ condition and environment. In order
for end users to make use of these tools to view the collected data, we need to
combine these data visualization tools. Therefore, we need to design an online
asynchronous collaborative discussion. system to provide artifact-centered
discourse which can support online communities among the participants who
are viewing the data. In this system, the users can choose the data visualization
tools, send requests to the database, and get the visualized data display through
the Internet. In addition, they can make online discussions in the system with
the data visualization results, data distribution map, data charting, or other
visualized data formats, as the artifacts to support their arguments.

As has been indicated to us by Kim Bridges, a botanist at the University of Hawaii

and a Principal Investigator of the Pods project, the interpretation and analysis of
the data gathered so far in our observations of the rare plant Silene Hawaiiensis
has never been done in the field of ecology. As in all scientific studies, hypotheses
about the observing targets need to be proposed, tested against the available data,
and refined appropriately. In order to work with these hypotheses and reach the
proper conclusions, the users need an online asynchronous collaborative working
environment to exchange their ideas derived from their substantial observations.
Within the online community formed among the viewers in the system, users can
build discussion threads, such as the effect of temperature on the flowering of the
plant, the influence of rainfall on the plant, and so on. The professionals or students
can build their individual hypotheses and arguments based on the observations, by
accessing the data visualization tools to view the current or historical data stored
in the database. Users can then discuss their hypotheses asynchronously and inter-
actively in various threads of conversation, which may lead them to draw reliable
conclusions and to refine their theories.
The concept of knowledge representation has also been emphasized in this design
because the conversations about the data will be as multifaceted as the data itself.
One of the main goals of the information interpretation system is to provide an ef-
ficient platform for users to present their observation results. With relevant lines
of discussion topic being distributed across multiple conversational threads and
associated with different artifacts, the asynchronous collaborative discussion en-
vironment can achieve the ultimate goal of the information interpretation system.
With substantial observations and discussions, scientists can eventually obtain the
adequate information about the plant under observation. The development of this
system can help scientists analyze the ecological environment of the plants, or even
other species if we apply our system to observe them in the future.

•. Human-computer.interaction.design:.While designing the human-computer

interface for the asynchronous collaborative discussion system described above,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

we applied the technology of scenario-based usability engineering (Rosson &

John, 2002). We emphasize collaborative interactions among the discussion
communities or groups with usability as the main concern in the design process.
In our scenario-based usability engineering, we included requirement analy-
sis, information and interaction design, prototyping, and usability evaluation.
Connecting the different design stages is a series of user interaction scenarios.
The scenario is actually a description of people activities. Representing the
use of a system or application with a set of user interaction scenarios makes
the system’s use explicit (Rosson & John, 2002).

We started our usability engineering design from the interview with the potential
users. The purpose of the interview is to understand the current work activities while
observing the plant and their expectations on the system. Based on the interviews,
we create the problem scenarios, which are descriptions of the current observation
activities in the problem domain. We also guide the following design stages, activ-
ity design, information design and interaction design, with appropriate scenarios.
Those scenarios are elaborated at the stage of prototyping. Web pages are selected
as the tool for prototyping. Figure 5 is an example of the interfaces for discussion

Figure 5. One example of the interface for discussion with a charting of the tem-
perature over time as the discussion artifact

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Fan

of the effect of rainfall on the plant. Below the chart on this figure (not shown in
Figure 5, but available in the actual interface) is a discussion area where the com-
munity members can post messages. The resulting system allows users to choose
a visualization tool from the function list as is shown in Figure 5. Then the system
switches to perform the selected visualization function. When the visualized data
form is presented, the user can add it to the discussion environment as an artifact to
support his or her arguments.The usability evaluation is conducted based on these
sketched prototypes. After several circles of evaluation and redesign, the prototypes
have become closer to usability objectives. In addition, we also adopted other us-
ability engineering methods, for example, usage centered design (Constantine &
Lockwood, 1999), in our design process. We have created a series of essential use
cases and navigation maps which give guidance to the entire design process.
In brief, the goal of the design for the online asynchronous collaborative discussion
environment is to apply the technology of usability engineering to our information
interpretation system. Therefore, end users will be able to reach reasonable and
accurate conclusions based on the data collected by the pods.

Related.Work

This chapter provides a relatively detailed description and rationale of the data
interpretation system designed for the wireless sensor network project which is
undertaken at the University of Hawaii. An article by Biagioni and Bridges (2002),
co-authored by two Principal Investigators, summarizes the goals of the project. As
a subproject of this wireless ad hoc sensor network project, our data interpretation
system has adopted a number of technologies, such as database management systems,
geographical information systems, dynamic Web programming, and human-computer
interactive design, for the information interpretation. We have also applied usability
engineering technologies to support collaborative online discussion.
Since we have potentially large amounts of data to display, we have carefully fol-
lowed the tenets for user interface design to allow users to focus on the data being
reported rather than on how to interpret the data. These principles encouraged us to
use intuitive representations such as colors for temperature and drops for rainfall,
Voronoi diagrams to identify the two-dimensional space that corresponds to a given
measurement, and customizable thresholds to provide useful summaries.
The GRASS GIS is used for data management, image processing, graphics pro-
duction, raster/vector spatial modeling, and visualization of many types of data.
It is capable of reading and writing maps and data to many popular proprietary
GIS packages including ARC/Info and Idrisi. Users wishing to write their own ap-
plication programs can do so by examining existing source code, interfacing with
the documented GIS libraries, and using the GRASS modules. This allows more
sophisticated functionality to be integrated in GRASS. The application of GRASS

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 A Data Vsualzaton and Interpretaton System for Sensor Networks 

GIS in our information interpretation system enables the combination of geographic

information system with sensor networks, thus becoming a unique technique for
visualizing weather data collected by sensor networks.
Interpolation based on Voronoi diagrams (Preparata & Shamos, 1985) is a well-
known technique in cartography. They tend to be involved in situations where a
space should be partitioned into “spheres of influence.” This makes applying Voronoi
diagrams for climatology feasible, and it has become a widely-used method in this
field (Thiessen & Alter, 1911). This interpolation method becomes an important
feature of our information interpretation system for sensor networks.
Usability engineering (Constantine & Lockwood, 1999; Rosson & John, 2002) re-
quires that a focus on user needs and capabilities pervade the entire design process.
While adopting usability engineering technologies in the design of our information
interpretation system, we need to consider both what information is significant to
the user and what information can easily be communicated to the user. A reviewer
has suggested generating the environmental data distribution map according to the
user’s setting rather than using the standard deviation to determine thresholds. This
emphasizes the user’s role and will make the system more interactive. In addition, a
distribution chart of the gathered data will help to set the threshold. We look forward
to implementing and evaluating this in future versions of the system.

Conclusion

While we are developing this information interpretation system specifically for the
Pods wireless sensor network project, we also expect that it can be applied to more
generic sensor networks, including wired networks. The end result should be to
make the sensor networks friendlier to users and more flexible to meet the require-
ments of different applications. It successfully conveys the information gathered
by the wireless sensor networks to ecologists, botanists, or other researchers. It also
provides a system for them to view the environmental condition around the target of
their observations. Based on the substantial understanding of the collected data, they
can also discuss and exchange their viewpoints through the collaborative discussion
environment provided by this information interpretation system.

References

Biagioni, E. S., & Bridges, K. (2002). The application of remote sensor technology
to assist the recovery of rare and endangered species. International Journal of
High Performance Computing Applications, 16(3), 315-324.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Fan

Bose, P., Morin, P., Stojmenovic, I., & Urrutia, J. (2001). Routing with guaranteed
delivery in ad hoc wireless networks. Wireless Networks, 7(6), 609–616.
Constantine, L. L., & Lockwood, L. A. D. (1999). Software for use: A practical
guide to the essential models and methods of usage-centered design. Reading,
MA: Addison-Wesley.
Fortune, S. J. (1987). A sweepline algorithm for Voronoi diagrams. Algorithmica,
2, 153-174.
Nagar, N., & Biagioni, E. S. (2002). Open issues in routing techniques in ad hoc
wireless sensor networks. In Proceedings of the International Conference on
Parallel and Distributed Processing Techniques and Applications (PDPTA),
Las Vegas, NV (Vol. 4, pp. 1867-1873). CSREA Press.
Preparata, F. P., & Shamos, M. I. (1985). Computational geometry: An introduction.
New York: Springer-Verlag.
Rosson, M. B., & John, M. C. (2002). Usability engineering: Scenario-based de-
velopment of human-computer interaction. UK: Academic Press.
Thiessen, A. H., & Alter, J. C. (1911). Climatological data for July, 1911: District
no. 10, Great Basin. Monthly Weather Review, 1082-1089.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Chapter.XIII

Towards.a.Global.
Real-Time.Enterprise
Peter Ibach, Humboldt Unversty Berln, Germany
Mroslaw Malek, Humbolt Unversty Berln, Germany
Gerrt Tamm, Unversty of Appled Scences Erfurt, Germany

Abstract

Frictionless interoperation of business processes even across enterprise boundar-

ies, complete up-to-date view of the overall status, and full remote control over the
business parameters for individuals in charge — this is the holy grail of a “global
real-time enterprise”. Yet a vision, a number of enabling technologies brought us
closer to accomplishing the challenges: sensing the position of mobile objects and
processes status, distributing the status information with “zero latency,” discover-
ing it according to specific demands across organization boundaries, providing
and securing uniform service-oriented access to all kinds of entities ranging from
smart items to business processes, and aggregating the overwhelming variety of
elementary services to form high-value composite services. In this chapter, we
overview the enabling technologies that drive the development and further discuss
market factors, security and privacy concerns, and standardization processes that
have to be considered. Then we propose our SEMALON approach, the SEMAntic

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

LOcation Network, intended as a basic infrastructure for discovery and composition

of location-based services. Finally we describe our experiences from a case study
implementation, the NOMADS Campus, which is a distributed spatial information
system for our campus at Humboldt University, Berlin.

Introduction

The “Internet of things” (Gershenfeld, Krikorian, & Cohen, 2004) with billions
and soon trillions of seamlessly interconnected devices is about to take over, and
we expect for the next years a literally exploding number of services that not only
provide information about physical objects, originating from Web pages, database
entries, or sensors, but also allow to trigger activities and control the objects by
some actuators. To support processes in the physical world by information technol-
ogy, a location-based mapping that connects physical objects and their correlated
information is required. Spatial interrelationship is what will put mobile users in the
position to navigate through the growing complexity and dynamics of physical and
informational spaces. Through the spatial organization of physical and informational
objects, virtual and real spaces will tightly interconnect.

The. Vision:.A. Global. Real-Time. Enterprise

In such a ubiquitous computing scenario, a “global real-time enterprise” envi-

sions that positions and other status information of all objects and corresponding
processes can be monitored and controlled in real-time — both, internally as well
as across enterprise boundaries. Clearly, a number of prerequisites have to be sat-
isfied, including enabling technologies as well as economic regulations and open
standards. Actual market trends, global competition, integration requirements, and
standardization in information technology are driving more and more enterprises to
adopt their coordination model. The organizational structures should be extremely
flexible and enable the integration of suppliers and customers processes. Real-time
enterprises supply information just in time to customers, suppliers, employees, and
partners, and integrate processes, systems, and media over organisational bounda-
ries. Emerging technologies, network economics, and global standards are the main
accelerators for the proliferation of real-time enterprises. Information systems of
real-time enterprises have to support permanent change of processes, data formats,
and interface specifications.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

The vision of a global real-time enterprise with end-to-end dynamic value Webs
becomes a reality. All processes and information flow within the enterprise as well
as external processes involving customers, suppliers, and partners will be instanta-
neously synchronized at all enterprise systems. Each service is transparent and can
be managed with an IT-service-catalog. When a new Web-based service is available
in the catalog, billing services and customer service are automatically added, and
the service can be used right from the moment that it was added to the catalog. Col-
laboration within real-time enterprises based on RFID and the electronic product
code will extremely reduce the transaction costs. Outdoor and indoor positioning
systems based on Internet-driven standardization (initiatives around XML and Web
services) make innovative business models possible.
Enabling technologies, first to mention among them are XML, RFID, and the EPC
Network, seamless positioning technologies, and open service oriented architec-
tures, develop at exponential pace and changing current information and business
situation dramatically.

Current.Applications/Architectures.and.their.Limitations

Although there are numerous proprietary applications that deal very well with location
information, interoperability of location information across application boundaries
in a standardized open format over the Internet is still to be completed. Considering
location semantics and mobility, the situation is even worse.
Present location-based services (LBS) are mostly bound to a specific technology
reflecting the preferences of the service provider. Figure 1 shows two exemplary
applications popular in the German LBS market: (1) The Jamba Finder allows cell
phone users to look for objects nearby, for example, public buildings, fuel stations,
cinemas, or restaurants. (2) Navigation systems as deployed in cars or independently
usable via GPS equipped PDAs enjoy rapidly growing popularity.
Typically, proprietary protocols and interfaces are employed in these LBS applica-
tions to aggregate the different system components for positioning, networking,
content, or payment services. In many cases, these components are glued together
to form a monolithic and inflexible system. If such a system has to be adapted to
new conditions, it very likely requires entire reengineering.
Let us consider a position-sensing service, for example, a satellite-based GPS. If a
mobile device moves from outdoor to indoor environments, the signal will likely
become unavailable and position sensing will fail. Without the location informa-
tion expected from this subservice, composite services depending on it will become
unavailable as well. To arrive at seamless operation, on-the-fly switchover to an
alternative position-sensing service using a different technology is required. To
choose from multiple possible position-sensing services, the decision has to consider
service availability, quality of service properties, and costs.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

Figure 1. Examples of current LBS value chains with participating service provid-
ers; multiple barriers hamper flexible service composition
Locaton Moble Devce
Infrastructure Moble Network Technology Provder or Content Portal Advertsement /
Provder Operator Provder Vendor Provder Operator Payment

Semens T-D T-D, Cell Noka, Jamba AG Jamba -

Jamba!
Finder

() ID Semens, AG
Sony, etc.
- - GPS Semens Tele Atlas, - -
Navigation
Systems

VDO, NavTeq,
() Poneer, etc.
Garmn,
etc.

no flexibility

In the near future, most mobile and wearable devices are expected to have multiple
available position-sensing technologies such as GPS, GSM, WLAN, and Bluetooth.
Nevertheless, new technologies, like at present WiMax or RFID, are emerging. Thus,
hardware devices and software components, their interfaces and architecture, have
to be able to deal with changing conditions. Thus, adaptivity, the ability to cope
with continuously-changing conditions, is crucial to make mobile location-based
services highly available and overall successful.
Lots of research has focused on location-based services combining the concept of
location-aware computing with distributed geographic information services based
on Internet standards (Hazas, Scott, & Krumm, 2004; Hodes, 2003; Peng & Tsou,
2004; Rao & Minakakis, 2003; Reichenbacher, 2004). Unfortunately, a number of
specific interoperability barriers exist in current LBS value chains, resulting in the
“Multi-X Problem”:

• Multiple connection technologies (GSM, UMTS, WLAN, Bluetooth, …)

• Multiple location technologies (GPS, Cell-ID, WLAN, Bluetooth, RFID,
…)
• Multiple hardware, software, and service providers
• Multiple operating systems, programming languages, and system architec-
tures
• Multiple application-specific ontologies describing location semantics
• Multiple content depending on specific location and granularity demands

Flexible service composition requires interoperability despite increasing multiplicity.

Web Service standards seem promising to solve this challenging problem.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Enabling.Technologies

Location-based services have been hyped as the “killer application” during the
Internet bubble, whereas true market developments could not accomplish the
exaggerated expectations. But with the advances of mobile devices, position sens-
ing, and wireless connectivity, the market for Location-based Services is rapidly
developing, particularly in the field of geographic, telematic, touristic, and logistic
information systems.

Seamless.Outdoor.and.Indoor.Positioning.Based.on...............
Commodity.Communication.Standards

Wireless emergency services require the ability to pinpoint the location of a cell
phone placing an emergency call, for example, for firebrigade, ambulance, or police.
E911 Phase II legislation in the U.S. requires cell phone companies to be able to
locate handsets within 150 meters by 2006. E112 initiatives in Europe are similar.
Positioning techniques now are maturing to provide accurate positioning in outdoor
and indoor environments at affordable cost, small size, and low power consump-
tion. Hamerhead, for example, is a single chip assisted GPS solution at €6.50 and
sufficiently sensitive that it works in most indoor environments. Infineon expects a
market of more than 700 million mobile phones to be sold in 2008 where 25% of
those will be equipped with A-GPS functionality.
Commodity mobile devices, such as laptops, PDAs, and cell phones can sense their
position even without extra GPS receivers. Intel’s PlaceLab project therefore has
mapped the positions of millions of existing GSM, WLAN, or Bluetooth base sta-
tions all over the world. Their experiments in the greater Seattle area indicate 20 to
40 meter median accuracy and close to 100% coverage exploiting “radio beacons
in the wild” (LaMarca et al., 2003).
At Humboldt University, Berlin we developed a WLAN positioning system called
MagicMap that can perform software-only positioning within a few meters average
deviation both outdoors as well as indoors (Ibach, Hübner, & Schweigert, 2004a;
Ibach et al., 2005a). Related wireless positioning systems (WPS) have recently been
released by companies such as Skyhook, Newbury Networks, or Cisco.1 These new
positioning techniques improve traditional satellite-based or telco-centered localiza-
tion in many practical scenarios and are capable to bootstrap the broad adoption of
location-aware computing.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

Open.Service-Oriented.Computing.and.Web.services

Enterprise applications were initially developed on closed, homogeneous mainframe

architectures. In the explosively growing heterogeneous landscape of IT systems
in the 1980s and ’90s, integration of intra- and inter-company business processes
became one of the most important and most cost-intensive tasks of the IT economy.
Due to missing or non-transparent standards, many enterprises pursued integration
by extremely expensive ad hoc solutions. With the spreading of the Internet and
the increasing importance of electronic business, open Internet-oriented solutions
have emerged. Enterprise-internal monolithic software was broken into smaller,
autonomous, and flexible components. This enabled the access to services not only
enterprise-internally, but along the whole value chain to suppliers, distributors, and
customers. We characterize this observation as a shift from rigid systems to flexible
service-oriented architectures.
In service-oriented computing, resources are accessed via services. Services ex-
pose well specified interfaces and are the basic building blocks for flexible and
efficient composition of more complex applications. The fundamental concept is
the composition of systems by extensive reuse of commodity software/hardware
components. Many approaches share this very general concept of compositionality
(see Figure 2).
However, a number of differences, for example, in wording, perception, implementa-
tion, and practical use, are indicating advantages of the service-oriented paradigm

Figure 2. Compositional architectures play an increasingly important role in value

chains due to improved possibilities of interoperability, integration, composability,
flexibility, reusability and thus increased efficiency at reduced total cost of opera-
tion
Value Chan

Coordnatng Servces
Framework

Deployment
Interface

Busness Process

Specfc
Interfaces

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Figure 3. Assembly of a 96 bit electronic product code

over previous approaches that were focusing on components, objects, modules, or

other compositional entities. At the forefront, Web services and grid technologies are
attracting a lot of attention accompanied by mixed opinions whether the expectations
in reusability, composability, flexibility, maintainability, and return on investment
that previous approaches have struggled with can finally be accomplished. See, for
example, Erl (2004) and Linthicum (2003) for the growing importance of Web Ser-
vices in enterprise application integration, and Bloomberg and Schmelzer (2002) and
Gershenfeld et al. (2004) for a detailed discussion of pros and cons comparing Web
Services to preceding concepts like CORBA. Commonly, the following advantages
are attributed to Web Services (still waiting for further empirical inspection):

• Improved degree of interoperability and flexibility (barrier-free computing),

across protocols, interfaces, programming languages, devices, connection
lines, operation systems, platforms, enterprise boundaries, vendors, and service
providers, through loose coupling based on the eXtensible Markup Language
(XML)
• Service aggregation using choreography languages supports “two-stage
programming” including flow control, exception handling, and transactional
processing
• Integrated directory services such as UDDI or WS-Discovery
• Enhanced protocols for propagation, discovery, and invocation of “lightweight”
services for embedded devices with limited processing and communication
capacity in ad hoc networks
• Asynchronous dependable messaging and security and privacy support for
identification, authorization, access control, and secure data transmissions

Web services are intended to facilitate the application-to-application interaction

extending established Internet technologies. Thereby Web services and grid concepts
are converging, guided by the open grid services architecture (OGSA). Its goal is
to overcome the two predominant challenges at the same time: uniform access to
services and processing resources.
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

Web services and grid toolkits like the Globus Toolkit or the Emerging Technology
Toolkit have helped to establish standards. Component-based software for embed-
ded systems (Müller, Stich, & Zeidler, 2001) and lightweight services (Milanovic,
Richling, & Malek, 2004; Schwan, Poellabauer, Eisenhauer, Pande, & Pu, 2002)
expanded the domain to span from distributed client-server applications and globally
networked e-business processes down to next generation heterogeneous embedded
systems. These developments paved the way towards the general paradigm of ser-
vice-oriented computing where all kinds of entities are providing, using, searching,
or mediating services while efficiently exploiting available resources. Driving the
widespread acceptance of the service-oriented paradigm, Location-based Services
might reveal the enormous economic potential of dynamic value Webs in mobile
business (Ibach, Tamm, & Horbank, 2005b).

RFID.and.the.EPC.Network

Automatic object identification is a fundamental requirement for efficient supply

chain management and usually solved by labeling objects with a bar code. However,
bar code labels are using line-of-sight technology, that is, a scanner has to see the bar
code to read it. Using radio frequency identification (RFID), in contrast, labels can
be read via radio communication as long as they are within the range of a reader.
With passive labels that derive their energy from the reader’s RF signal, a distance
of 5 cm up to 10 m is achievable, depending on physical characteristics comprising
used antennas and wavelength. RFID labels supporting collision-free communication
protocols provide further significant advantage over conventional labels: They allow
for bulk readouts and enable trolley or container scanning in a single pass. Current
anti-collision RFID chips are approaching prices below $0.05 in mass production,
making it affordable to label almost any commercial item electronically. Although
RFID tags probably will never replace bar code labels completely, apparently, the
broad use of RFID technology in supply chain management is forthcoming.

The.Electronic.Product.Code

As they can be read out through optical barriers, RFID chips enable the automatic
collection of huge amounts of data at various locations in a supply chain. Com-
ing along with the RFID technology, a new numbering scheme to label items was
developed.
The identification number follows the electronic product code (EPC) standard
(Harrison, 2003). It is designed to uniquely identify commercial goods. Each EPC
consists of three parts: (1) a number identifying the manufacturer, (2) a number for
the product type, plus (3) a serial number for each individual item (see Figure 3).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

While the EPC manager and the object class represent the same information as stored
in traditional European article numbers (EAN), the serial number adds item-level
granularity to product numbering. This allows for identifying each object individu-
ally, and is ideal for using it as a primary key and for integrating data about objects
from various databases.

The.EPC.Network

The standardized Electronic Product Code that uniquely identifies each object is
stored on the object’s RFID tag. Some applications would certainly benefit from
additional information directly stored on the object’s tag, for example, prize, size,
weight, place of origin, or date of expiration. However, capacity for additional
information is very limited. Current chips in the range of $0.05 comprise a stor-
age capacity of typical 128 bits. Of course there are further reasons, for example,
availability and security issues, why certain pieces of information should not be
stored on the object’s tag. Just imagine, for example, information about transport
damage or about the object’s actual location. Therefore, the EPC network provides
the infrastructure to maintain such additional product-related information. Thus,
considerable information about each individual item can be stored, retrieved, and
secured by appropriate access control facilities.

Data.Discovery.within.the.EPC.Network

Since each EPC is globally unique, it can be used as a key for database lookup. For
discovery of EPC-related data, the EPC Network provides a number of services that
seamlessly integrate into Internet discovery standards: At first, the object naming
service (ONS) is employed to retrieve data resources related to a specific EPC. The
ONS returns the different data resources containing information about the specific
object, whereby each resource is denoted by a uniform recourse name (URN). For
each URN, the universal naming service (UNS) resolves the URLs, describing the
physical locations of the data. Multiple locations are feasible, since the same data
can be stored at different places. Finally, a URL is translated into an IP address by
the DNS to arrive at an EPC information service (EPCIS) endpoint that manages the
access to the designated data. The architecture is fully distributed, that is, requests
to ONS, UNS, or DNS are delegated to nearby servers while appropriate caching
and synchronization protocols reduce network traffic and response times.
Using this discovery mechanism, data related to an EPC can be located. Usually,
the URL of at least the EPC information service of the product manufacturer is
expected to be returned when retrieving information via an object’s EPC. However,
to improve supply chain efficiency, it is intended that various suppliers, distributors,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ibach, Malek, & Tamm

retailers, contractors taking care of logistics, and other partners within a supply chain
gather information and make it available by providing additional EPC information
services. Companies may maintain their EPC information services on their own or
delegate them to outsourcing partners.
The Physical Markup Language (PML) has been specified to standardize those pieces
of information and to make them machine understandable and interchangeable. PML
is XML-based whereby special conventions to express spatio-temporal annotations
or containment information have been predefined by the Auto-ID Center. However,
PML is designed to be expandable.
As a tagged item moves, usually because of some logistics processes, it passes
several instances within the supply chain. Each instance might gather additional
information about the item. For example, a retailer may want to store the time at
which a certain product has arrived at his storehouse.
Due to this, more than one EPC information service will provide information about
an object as it moves through the supply chain. In order to locate these diverse EPC
information services, a dynamic approach is needed.
In “EPC Information Service — Data Model and Queries” by the Auto-ID Center
(Harrison, 2003) two possible ways of handling that issue have been suggested.
One solution is to organize EPCIS, holding information about a certain EPC, in a
double-linked list. Each EPCIS points to its predecessor as well as to its successor

Figure 4. Players in the supply chain offer product information via individual in-
formation services (EPCIS); the ONS and the EPC discovery service help to locate
services providing information related to a specific object

Points to
URL of EPC Discovery
ONS Service

Points to URL of
Request about
an EPC

EPCIS EPCIS EPCIS

Requesting role: Local Local Local

DB DB DB
Manufacturer Distributor Retailer

Movement across the supply chain

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

in the supply chain. To retrieve the most recent information about an object, each
instance in the list has to be consulted for the successor until the list’s end is reached.
However, this approach only works as long as all of the participating EPCIS are
available. Moreover, due to the expected inefficiency, alternatively a central service
has been investigated. This central service holds a list of all EPCIS that are referring
to an EPC. Once an EPCIS provides information related to an EPC, it has to register
at this central service. VeriSign has announced an initiative to establish such a kind
of service, called EPC discovery service. An overview of the mentioned services
and how they are organized in the EPC Network is given in Figure 4.

Information. Economy

Dynamic.Value.Webs

Web services appear promising to facilitate interoperability and enable dynamic value
Webs, the on-demand aggregation of services even across enterprise boundaries.
This is a prerequisite when striving for a global real-time enterprise. A lot of work
on service-oriented computing refers to a vision that enables users to formulate
abstract requirements, which are then performed by adaptive, self-configuring ser-
vices (Milanovic & Malek, 2005). This in particular implies services that are able
to find and to call other services automatically. Approaching this vision, software
development based on Web service standards is expected to exhibit advantages over
classical software development. However, experiences have shown that this vision
of future Web services is still difficult to realize. As yet, the volume of transactions
in the Web service market has by far not reached the expected level. Information
asymmetries and uncertainties among suppliers, aggregators, and potential customers
are the predominant causes for the modest development and enforcement of Web
services (Tamm & Wuensche, 2003).
This chapter investigates the economic impact of service-oriented architectures
for building dynamic value Webs in mobile environments using adaptive location-
based services. It addresses the following three predominant difficulties: (1) service
adaptivity to changing conditions in mobile environments, (2) interoperability in-
cluding higher levels of semantics, and (3) assuring trustworthiness to all affected
parties. Web services are modular self-describing software components (Alonso,
Casati, Kuno, & Machiraju, 2004). Due to their standardization they may be used
individually or may be aggregated to software bundles. Web services standards fa-
cilitate enterprise application integration (EAI), offer new ways of interconnecting
both consumer and business partners, and provide a comprehensive framework to
enable the evolution of dynamic value Webs.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

Web services are digital goods which can be aggregated to dynamic value Webs on
the basis of a service-oriented architecture. Compared to physical products, Web
services are thus very flexible. A fast growing market for Web services was forecast.
Unfortunately, the market for Web service has so far not lived up to the predicted
expectations. Possible reasons for this relatively slow adoption may originate from
the different perceived risks of suppliers and customers, which often rise from a
lack of experience and knowledge about Web services. Suppliers’ problems arise
regarding configuration, display, and communication of Web services, contract, and
escalation management (Tamm et al., 2003). Also, pricing, billing, and liability is-
sues remain to be a problem. However, a major problem is that potential customers
do not know where their data is saved, what is done with it, how stable the service
is, and which of the promised privacy properties it keeps.
The following section uses information economics as a methodical approach for
the scientific analysis of the information problems described above and for the
development of strategies to overcome these problems. The theory of information
economics is part of the new institutional economics. It is based on the idea of
classifying the detectability of the quality of a product on the basis of the different
consequences of information asymmetries. According to that, several strategies
will be developed, aiming at a reduction of the classified types of uncertainties in
the context of quality evaluation. The characteristics of the mobile service bundle
are attributed to different information economics-based types. Then, the strategies
recommended shall be deployed in order to facilitate the quality evaluation and the
perception of the characteristics of the mobile service bundle. The following section
describes the fatal consequences likely to occur if the actors on the mobile service
market do not succeed in reducing information asymmetries.
Potential users equipped with only little experience and knowledge about the mobile
service model lack methods to assess the quality of mobile services supplied. The
lack of transparency perceived with regard to the quality of the performance char-
acteristics of the mobile services bundle and the providers of a mobile services, as
well as according to partners of the providers, leads to an information asymmetry
on the demand side, and thus to the small number of contracts placed.
The information asymmetries prevailing between the supply and the demand side do
not only cause high transaction costs, but can ultimately even result in a complete
market failure.
Figure 5 relates and summarizes the main causes to the emergence of information
asymmetries on the Web service market (Tamm et al., 2003).

Markets.vs..Hierarchies

Web services use the infrastructure of electronic networks like the Internet, which
introduces new combinations of business characteristics described as Internet economy

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Figure 5. Influence factors of the information asymmetry on the Web service mar-
ket

(Picot, 2004). Most differences compared to the “old economy” are advantages in
the reduction of transaction costs. Malone, Yates, and Benjamin characterize the
results of using new technologies in economic processes: “By reducing the costs
of coordination, information technology will lead to an overall shift toward propor-
tionately more use of markets — rather than hierarchies — to coordinate economic
activity” (Malone, Yates, & Benjamin, 1997, p. 484).
Most companies on the Internet concentrate on their core competencies for speci-
fied services. That leads to a strong division of labor which creates learning-curve
effects and increasing economies of scale and scope. Compared with physical
markets, a Web service aggregator for dynamic value Webs faces relatively low
search expenses, a better market transparency on the Internet, and moderate pur-
chasing costs of elementary Web services. It will therefore outsource a lot and set
its boundaries very tight.
Transaction costs in electronic markets like the Internet are lower than in physical
markets. Nevertheless, providers of digital goods, especially suppliers of value
Webs, must cope with the low willingness to pay for their products. Therefore, the
Web service supplier has to consider strategies to decrease the transaction costs on
the supply side. An important aspect of lowering transaction costs is the presence of
trust between business partners. Trust develops in partnerships over a long period of
time in which the partners are content with the quality of supplied products as well

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

as prices. Costs for searching, negotiating, integrating, and especially monitoring

do not apply here. However, other suppliers on the Internet might offer a better
quality or a better price for the same elementary Web service. Hence, even in long
business partnerships, carriers of dynamic value Webs have to constantly watch
new market developments.
Another method of lowering transaction costs is the increased speed of contracting
between Web service suppliers and carriers of dynamic value Webs. In the short run,
one approach is to standardize elementary Web services in order to make them easier
to handle and easier to understand. In the long run, complex automated electronic
markets might be developed which are able to handle a lot of different Web services
very fast and cheap. Then, standardization has to turn into product differentiation.
In this context, two groups of products can be distinguished: Contract goods and
exchange goods (Alchian & Woodward, 1998). Products of the first group are com-
plex and often need description and negotiation before purchase. The second group
contains products with clearly delimited and simple properties. Here transaction
costs are lower because no complicated negotiation is required. Standardizing a
Web service corresponds to the procedure of transforming a contract good into an
exchange good. New approaches to speed up the contracting process in the Web
service market are contractually assured specifications regarding, for example, qual-
ity of service levels, policy assertions, or pricing models. The premise is that the
properties of services are described, fixed, and understood by all interacting parties
to build up dynamic value Webs. Following, we describe the principle advantages
that Web service standards provide for service coordination and composition and
investigate the applicability to set up adaptive location-based services.

The.Need.for.Standards

Standardization of IT services and harmonization of the IT landscape are the main

challenges in IT business today. From the economic perspective, IT services have
to be described in a transparent and comparable way. Customers of IT suppliers
expect increasing productivity, return on investment, and additional benefit by using
information technology. IT services which support governmental and commercial
business processes, have to be monitored including key performance indicators (ef-
ficiency) and key goal indicators (effectiveness). From the technological perspective,
standards (e.g., XML, Web services) have to be developed and widely accepted
in specific branches. Based on such standard software modules, interoperable IT
architectures can be established. Most of the available spatial information is locked
in non-standard systems, useless for interoperable architectures and real-time busi-
ness process integration.
The Open Geospatial Consortium (OGC) develops, discusses, and releases techno-
logical standards for geospatial information systems. The OGC is a global industry

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

consortium that envisions a world in which everyone benefits from geographic

information and services made available across any network, application, or plat-
form (Reed, 2005). Inherent in this vision is the requirement for standardization in
geospatial technology and services. In this chapter we describe the current OGC
standards based on Web service technology.

Geography.Markup.Language.

The Geography Markup Language is one of the most important standards for
geospatial information systems. Geography Markup Language is an XML gram-
mar written in XML schema for modeling, transport, and storage of geographic
information. GML provides a variety of objects for describing geography including
features, coordinate reference systems, geometry, topology, time, units of measure,
and generalized values.
The main characteristics of GML are features. A geographic feature is “an abstrac-
tion of a real world phenomenon; it is a geographic feature if it is associated with
a location relative to the Earth” (Cox, Daisey, Lake, Portele, & Whiteside, 2005, p.
xviii). Features are geographical characteristics of objects (e.g., landscapes, cities).
Each feature includes geometrical (e.g., point, circle, polygon) and non-geometrical
characteristics. Both types can be used optionally. A digital representation of the
real world can be thought of as a set of features. The state of a feature is defined by
a set of properties, where each property can be thought of as a triple (name, type,
value). The number of properties a feature may have, together with their names and
types, are determined by its type definition.
Geographic features in GML include coverages and observations as subtypes. Cov-
erage is a sub-type of feature that has a coverage function with a spatial domain
and a value set range of homogeneous two to n-dimensional tuples. A coverage can
represent one feature or a collection of features. An observation models the act of
observing, often with a camera, a person, or some form of instrument. An observa-
tion is considered to be a GML feature with a time at which the observation took
place, and with a value for the observation. A reference system provides a scale of
measurement for assigning values to a location, time, or other descriptive quantity
or quality. A coordinate reference system consists of a set of coordinate system
axes that is related to the earth through a date that defines the size and shape of the
earth. Geometries in GML indicate the coordinate reference system in which their
measurements have been made. The parent geometry element of a geometric com-
plex or geometric aggregate makes this indication for its constituent geometries. A
temporal reference system provides standard units for measuring time and describing
temporal length or duration. A measure dictionary provides definitions of physical
quantities, such as length, temperature, and pressure, and of conversions between
different dictionary systems.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

Figure 6 illustrates the advantages of GML. Existing services (e.g., map services)
and additional data (e.g., length of road or number of lanes) can be integrated. By
the usage of XML schema, existing data can be proofed and evaluated.

Open. Geospatial. Services

OGC Web Services Common Specification

The OGC Web services common specification (OWS) (Whiteside, 2005) specifies
many of the aspects that are, or should be, common to all or multiple OWS inter-
face implementation specifications. Those specifications currently include the Web
map service (WMS), Web feature service (WFS), Web coverage service (WCS),
and catalog service (CAT). These common aspects include: operation request and
response contents; parameters included in operation requests and responses; and
encoding of operation requests and responses. The OWS specification is a norma-
tive reference for all subsets of OWS. OWS describes common specifications for
the following topics:

• Normative references
• Terms and definitions

Figure 6. GML example describing “Highway between Berlin and Hamburg”

<Road id="R456">
<description> Highway between Berlin and Hamburg</description>
<name>A 24</name>
<nLanes>4</nLanes>
<surfaceTreatment>bitumen</surfaceTreatment>
<destination xlink:href="http://some.big.org/places/G6421„ />
<pavement>
<Polygon> ... </Polygon>
</pavement>
<centreLine> <Curve> ... </Curve> </centreLine>
</Road>

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

• Conventions
• GetCapabilities operation
• Exception reports
• All operations except GetCapabilities, minimum abilities
• operation parameters
• Operation request and response encoding
• Guidance for OWS implementation specifications
• XML schemas

In the following sections we will give an insight into WMS and WFS.

OGC.Web.Map.Service

OGC Web map service (WMS) (Sonnet, 2005) specifies how individual map servers
describe and provide their map content. Since December, 2005, WMS is available
as ISO 19128 standard.
WMS provides the operations GetCapabilities, GetMap, and GetFeatureInfo. WMS
supports the just-in-time creation and presentation of maps based on personal re-
quirements. With WMS it is possible to combine several different map information
sources, which are remote and heterogeneous, to a specific and individual map in-
formation service. The present context specification states how a specific grouping
of one or more maps from one or more map servers can be described in a portable,
platform-independent format for storage in a repository or for transmission between
clients. The WMS standard addresses basic Web computing, image access, display,
and manipulation capabilities. It specifies the request and response protocols for
Web-based client/mapserver interactions. Web mapping refers, at a minimum, to
the following actions (McKee & Kottman, 2003):

• A client makes requests to one or more catalog servers to discover URIs con-
taining desired information
• Catalog servers return URLs and also information about methods by which
the discovered information at each URL can be accessed
• The client locates one or more servers containing the desired information,
using OGC’s catalog server technology, and invokes them simultaneously
• As directed by the client, each map server accesses the information requested
from it, and renders it suitable for displaying one or more layers in a map
composed of many layers. Map servers provide the display-ready information
to the client, which then display it.
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

The description of WMS can be realized with Web map context documents. A context
document includes information about the server providing layer in the overall map,
the bounding box, and map projection shared by all the maps, sufficient operational
metadata for client software to reproduce the map, and ancillary metadata used to
annotate or describe the maps and their provenance for the benefit of human view-
ers. A context document is structured using eXtensible Markup Language (XML).
XML schema is used for the validation of WMS context documents.

OGC.Web.Feature.Service

In addition to the WMS, the OGC Web feature service (Vretanos, 2005) allows
users to define and model personalized views on geospatial data and to save these
views as features. By using Geography Markup Language (GML) for OGC Web
feature service geospatial data can be recovered and updated. The requirements for
a Web feature service are:

1. The interfaces must be defined in XML.

2. GML must be used to express features within the interface.
3. At a minimum a WFS must be able to present features using GML.
4. The predicate or filter language will be defined in XML and be derived from
CQL as defined in the OpenGIS catalogue interface implementation specifica-
tion.
5. The datastore used to store geographic features should be opaque to client ap-
plications and their only view of the data should be through the WFS interface.
The use of a subset of XPath expressions for referencing properties.

OGC Web feature service (WFS) defines interfaces for data access and manipula-
tion operations on geographic features using HTTP as the distributed computing
platform and XML for data description. Via these interfaces, a Web user or service
can combine, use, and manage geographical data by invoking the following WFS
operations on geographic features and elements:

• Create a new feature instance

• Delete a feature instance
• Update a feature instance
• Lock a feature instance

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Get or query features based on spatial and non-spatial constraints Web feature service
allows a client to request spatial data, not a map, from a service. The spatial data is
encoded in the GML. The WFS specifies interfaces that unambiguously model the
behavior of feature identifiers. When a map layer (or “feature collection”) contains
multiple identical features, when features move, and/or when it is possible to ac-
cess many different maps, it is sometimes necessary to have “feature identifiers”
that help identify a particular feature in different maps as being the same feature.
Different software systems with different feature identification approaches need a
way to communicate such information, and that is what this specification will pro-
vide. WFS also specify common interfaces for communication about relationships
between features. WFS include functions for the exchange of features and data of
features. The specification of WFS describes in detail the following functions:

1.. GetCapabilities: describes the characteristics (functions, services) of a WFS.

Respond and request data are based on XML
2.. DescribeFeatureType: describes meta data of WFS structure
3.. GetFeature: requests feature data from the WFS
4.. Transaction: adds, changes, and deletes data
5.. LockFeature: locks feature data to ensure consistency if used concurrent to
other requests

OGC.Web.Coverage.Service

The Web coverage service (WCS) (Evans, 2003) provides access to detailed in-
formation about the geospatial data. With WCS, users of the geospatial data can
get detailed information about requirements for client-site rendering, coverage of
multi-geospatial data sources and obligatory transformation rules.
The Web coverage service describes the geospatial data and supports users with
detailed descriptions of the geospatial data source. With this additional information,
users of the geospatial data are able to create complex queries against the geospatial
data sources. Instead of requesting graphical maps, users of the WCS can request
the original geospatial data source with original semantics. Based on the original
geospatial data, users of the WCS are very flexible to use the data for transforma-
tion, based on time and space criteria.
The Web coverage service provides the operations GetCapabilities, GetCoverage
and DescribeCoverage. With the GetCapabilities operation, users get a XML-docu-
ment, which includes a main description of the service. The XML-document also
describes the data source from which the user can request coverage. It is possible
to run the GetCapabilities operation once and store the XML-document for future

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ibach, Malek, & Tamm

reuse in single or multiple sessions. When the GetCapabilities operation is not able
to respond with the XML-description document, an alternative source (e.g. image
database) can be offered. With the DescribeCoverage operation, users can request a
complete description of one or more coverages served by a selected WCS server.
With the GetCoverage operation, users request a single coverage layer of the ge-
ospatial data (e.g., sliced in time, elevation, or range components).

SEMALON.—.the.SEMAntic.LOcation.Network

As part of NOMADS, Networks of Mobile Adaptive Dependable Systems (Malek,

2003; Malek, 2004), which is a project for building a comprehensive service-ori-
ented framework for ubiquitous computing, we are developing the Semantic Loca-
tion Network (SEMALON) based on open Web Services standards. In SEMALON,
all resources are uniformly accessed via services. Services expose interfaces that
can be semantically interpreted using ontologies. Multiple layers contain objects,
their locations, and associated services (see Figure 7). Physical resources might
be stationary, such as streets and buildings, or mobile, such as mobile devices or
RFID-tagged items. Informational resources comprise. Internet pages, objects in
databases, and services.

Spatial.Semantics

For semantic location determination, we distinguish the following LBS classes:

• Location-based services can be provided by some immobile unit, for example,

a museum or a botanical garden. Typically, such immobile units provide sta-
tionary.LBS which are fixed to a certain location. A common problem is to
semantically detect the location, and find or filter stationary services related
to that location. For example, a user’s movement in a museum can tell that he
might be interested in information about a specific exhibition object (e.g., he
moves to that object and then, while looking at it stops moving for some sec-
onds). A location-aware device could then request the appropriate service.
• Likewise, some immobile units may provide general.LBS that are location-
independently accessible but require a location parameter. Examples are a
regional weather forecasting service or a service that processes queries like
“where is the next subway station?”
• Regarding mobile.LBS, the location is a parameter describing the context of
a mobile device. Imagine a user traveling with his laptop: If the laptop recog-
nized the availability of a specific LAN connection, it could figure out where

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Figure 7. SEMALON — a globally scalable semantic network of location-based

services
servces wth
regonal scope

layers wth
specfc objects
and ther locatons

servces assocated
wth specfc
objects/locatons

physcal space nformatonal space

it is located (e.g., in the user’s office) and adapt its behavior (e.g., synchronize
certain files).
• Finally, interdependent.LBS require multiple related location parameters, for
example, a people-finding service that guides mobile users to meet at some
specific place.

All these cases demand appropriate semantic interpretation of location. To accom-

plish semantic interoperability, one has to agree on suitable ontologies which define
objects and relations for each specific application area.

Spatial.Ontologies

Typically, locations are represented by geographic coordinates. In common use is

the projection according to the World Geographic System, 1984 (WGS84); other
projections comprise Universal Transverse Mercator, Swissgrid, Gauss-Krüger-grid,
or the military grid reference system, which can be interchangeably converted by
software algorithms.
Going beyond bare geo-coordinates or free-form textual descriptions, spatial on-
tologies can be used to define objects and relations by means of spatial semantics.
A widely-accepted ontology that models physical objects and their location is the
Geography Markup Language (GML), standardized by the OpenGIS Consortium
and used in the geographic information system (GIS). The Physical Markup Lan-
guage (PML) of the EPC network, standardized by the Auto-ID Center, is intended

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

for product classification, but also allows for spatio-temporal annotations for object
tracking and supply chain management. The World Wide Web Consortium is extend-
ing the resource definition framework (RDF) to relate Web content to its associated
physical location. The DARPAAgent Markup Language (DAML) combines multiple
schemes for location description.
Using GML, DAML, or RDF, complex schemes can be designed. Related elements
can be grouped and hierarchically structured to represent different aspects of loca-
tion information (see Table 1).
Typically, discovery by means of spatial semantics is done describing some known
objects and, based on those, query for other spatially-related objects. In previous
examples we have used radial distance to a central point. Table 2 shows how this
can be expanded regarding two- or three-dimensional objects, different prepositions,
and custom distance measures.

Ontology.Translation

In our approach, ontology translation is handled by dedicated ontology translation

service (OTS) instances. Inputs are: source ontology instances, including a refer-
ence to the source ontology, and references to one or multiple target ontologies. It
outputs the information translated into the requested target ontologies. Typically,
ontology translation is pursued in three steps:

1.. Discovery: manually, automatically, or semi-automatically defining the rela-

tions between ontologies
2.. Representation: A language to represent the relations between the ontolo-
gies

Table 1. GML example describing the Sony Tower at Potsdamer Platz in Berlin,
comprising address, surface area, and geo-coordinates
<exp:Building fid = “Sony Tower”> ← name of the building
<exp:noFloors>26</exp:noFloors> ← number of floors
<exp:use>Commercial</exp:use> ← commercial type of use
<exp:surfaceArea>216700</exp:surfaceArea> ← surface area in m2
<exp:frontsOn>Neue Potsdamer Straße </exp:frontsOn> ← street
<gml:locationOf>
… ← geo-coordinates, in WGS84 standard
</gml:locationOf>
</exp:Building>

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

Table 2. Describing spatial relations between objects

Object Preposition Distance Use.Case.Example:.look.up.…

… WLAN hot spots within a radius of 50
undirected meters
meters
point … holiday destinations south of your
south flying time
location
undirected walking time … restaurants within 5 minutes of walk
undirected driving time … SOS-telephones along a highway
polygon … customs facilities west of the country’s
west driving time
border
undirected boolean … people inside a specific room
polyhedrons undirected boolean … adjacent offices in a tower block
below walking time … parking levels below the first floor

3.. Execution: Changing instances of the source ontology to instances of target

ontology

One approach for ontology translation is to provide an explicit m-n mapping for any
given pair of m source to n target ontologies. This potentially achieves maximum
translation quality, but the required number of mappings grows quadratically, that
is, at O(m∙n). At the other extreme, the m-1-n translation introduces an intermediate
ontology into which all source ontologies are translated and from which all target
ontologies are derived (see Figure 8). This minimizes the number of required map-
pings to linear size, O(m+n), but for many cases, it results in unacceptable loss of
information. Therefore, we pursue a hybrid approach, where the best path over a
manageable number of intermediate ontologies is chosen.
If a suitable translation path can be found, the source information can be exploited,
for example, to trigger certain activities. Imagine, for example, a user who wants his
mobile phone to automatically activate the hands-free speaking system inside a car
or mute when inside a theater. Usually, a cellular phone cannot tell from geo-coor-
dinates, determined, for example, by a GPS signal or by its cell ID, that it is inside
a theater. But if the location description contains information that allows deriving
from given geo-coordinates that the location is a theater, belonging to the category
<silent space>, the “mute feature” could be automated (see Figure 9).
Accordingly, the location <Prater>, a theater in Berlin, would need to indicate that
it is a <theater> in the <places> scheme, which defines that a <theater> is a <silent
space> (see Table 3). Nested ontology translations of such kind are a major chal-
lenge of the Semantic Web.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
294 Ibach, Malek, & Tamm

Figure 8. Ontology translation as a hybrid of m-n and m-1-n mapping

Ontology Translation

m-n m-1-n hybrid

translation translation translation

source intermediate target

ontologies ontologies ontologies

Adaptive Location-Based Services (ALBS)

From the perspective of a mobile user, the environment is ever-changing as he moves

from one location to another. Adaptivity to location characteristics is essential for
mobile service availability. In our approach, adaptivity of a composite location-
based service — we call these services adaptive location-based services (ALBS)
— is accomplished by choosing the appropriate chain of subservices for composi-
tion (see Figure 10). Prerequisites are general discoverability, interoperability, and
composability of subservices through standardized communication protocols and
directory services.
Lots of research has focused on location-based services combining the concept of
location-aware computing with distributed geographic information services based on
Internet standards (Hazas, Scott, & Krumm, 2004; Hodes, 2003; Peng & Tsou, 2003;
Pinto, Boas, & José, 2003; Rao & Minakakis, 2003; Schiller & Voisard, 2004). In
particular, the Open Geospatial Consortium (www.opengeospatial.org) is focusing
on Web service standards for publishing, finding, and binding geospatial services.
In the Web services standard, interoperable communication is accomplished by
exchanging XML data over HTTP. But Web services are not restricted to the WWW
or a specific protocol. Rather, it is a promising solution for adaptive application
synthesis in distributed, dynamically-changing environments. The notion of Web
Services goes beyond prescribed client-server communication. Emerging standards
for directory services such as UDDI (UDDI, 2005) or Web services choreography
(BPEL4WS, 2005) allow for dynamic discovery of services and composition of
multiple Web services to fully-fledged distributed applications.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Time Enterprise 295

Figure 9. Nested ontology translation from WGS84 geo-coordinates into the value
<mute> in the <CellPhoneProfiles> ontology

  
Value: Geo-coordinates Theater Silent Space Mute

CellPhone
Ontology: WGS84 Places Places/Theater
Profiles

Table 3. Example RDF description of location <Prater>, a theater in Berlin, indi-

cation location semantics
<rdf:RDF xmlns:rdf=”…” xmlns:plc=”…” ← XML namespaces
xmlns:geo=”…” xmlns:scp=”…”>
<plc:Places rdfID=”…”> ← place attributes
<plc:category>theater</plc:category> ← it is a theater
</plc:Places>
<scp:Scope rdfID=”…”> ← object’s extent:
<scp:type>undirected</scp:type> ← radial distance of 20
<scp:distance>20</scp:distance> ← meters to
<scp:metric>m</scp:metric>
</scp:Scope> ← the point with geo-coordinates:
<geo:Point rdfID=”…”> ← Latitude: N52° 32.39’
<geo:lat>52.539833</geo:lat> ← Longitude: E13° 24.64’
<geo:long>13.410666</geo:long> ← Altitude: 91 meters above sea level (WGS84 standard)
<geo:alt>91.000</geo:alt>
</geo:Point>
</rdf:RDF>

Consider a location-based service that requires some input, for example, accurate
position information or the user’s choice of payment. The user might present these
data to the LBS manually. Likewise, this information might be the result of a preced-
ing Web service which, for example, reads the geographic position from an attached
GPS device. In case of payment, information about user’s choice of payment could
be sent to an accounting service which, for example, uses a direct debit authoriza-
tion. For service composition it is not necessary to know how the accounting is
actually performed or how the access to the GPS device is implemented, as long
as one can trust the responsible Web services. Authorization and trust will be fun-
damental for the success of location-based services and Web services composition.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

Moreover, protecting privacy regarding the user’s trace of information is a severe

issue. Further dangers of intrusion to take care of are service spamming, where un-
desired services are propagated, and service spoofing, where insecure services are
offered under disguised identity. Ongoing developments in the Web Services Trust
Language (WS-Trust) therefore accommodate a wide variety of security models.
However, proper service specification as well as justification of a specification’s
trustworthiness will remain a major challenge to ensure Web service interoperability
across enterprise boundaries.

Using.Web.Services.for.ALBS.Implementation

We use Web services standards to implement the appropriate selection of subservices

and to process their composition. These comprise the service interface description
in the Web Services Description Language (WSDL). In an interface description
the port type specifies the service’s request/response behavior. A service instance
is accessed through a port. Each port has to bind to a port type and has to support
additional binding information, for example, the used protocol.
For each application to be composed of subservices, a flow through required port
types and optional port types guides the composition process. This flow can be
specified using choreography languages (e.g., WSCL or BPEL4WS). Ongoing
developments in Web services choreography incorporate far-reaching flow control
where, for example, a group of services can be processed transactional or the flow
may branch with respect to optional services availability or in case exception oc-
curs. The composition process can be graphically expressed by a path through a
network of accessible ports.
The composition process is triggered at service invocation. Whenever an ALBS
is invoked, it is dynamically composed of suitable ports. Among the ports of each
port type, the best match will be taken with respect to the specific context that de-
termines availability and suitability of each port. Therefore, the Web services policy
framework defines general purpose mechanism for associating policy expressions
with subjects. Thus, specific property assertions can easily be attached to ports and
registered in the repository.
For successful composition, at least one port of each required port type has to be
accessible. In the example presented in Figure 10, there are five port types:

•. Connection: Services of this type allow for access to available networks.

This could be WLAN, LAN, Bluetooth, GSM, GPRS, or UMTS connections.
Properties attached to these ports comprise information about bandwidth, costs,
power consumption, encryption, or latency time.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Time Enterprise 297

Figure 10. Building dynamic value Webs by on-demand composition of adaptive

location-based services

services providing the same

interfaces (port types)

Semantic
Position Location
Connection Sensing Determination Content Accounting

Bluetooth Cell-ID Ontology 1 Flat

UMTS GPS Ontology 2

PPT
RFID
WLAN Ontology 3
composite
PPV
WLAN service

service instances at disposal (ports)

• Position sensing: This port type provides location information. Ports can be
GPS-receivers, services based on stationary RFID-tags, or Bluetooth trans-
mitters. Other services, for example, based on WLAN-positioning or Cell-ID
in cellular networks, are possible candidates as well. The properties should
contain information about the accuracy of the location information. Exten-
sions of position sensing services might be able to recognize direction, speed,
and variance of movement. (WLAN and Bluetooth positioning base on signal
strengths of different access points. For each position, these signal strengths
exhibit a certain characteristics that can be translated into location informa-
tion by a signal characteristics map. Since the signal strengths vary, the map
needs periodic update. Nevertheless, coverage and accuracy of the positioning
may be insufficient for some LBS. However, this way PDAs, laptops, or other
mobile devices can locate themselves independently of GPS availability.)
• Semantic location determination: Information about location semantics is
offered by this port type. Input is the context-specific sensor data (containing
geographic location, RFID numbers, available services, or other specific char-
acteristics that could be utilized to reason about the position). The response
includes the semantic position according to a given ontology.
• Content: This port type offers content for a given geographic or semantic
location. It receives a message with the location information which then is
processed. The returned message contains information (text, pictures, audio,
or video if requested) about the given location. To process the location infor-
mation semantically, some common ontology is required.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Ibach, Malek, & Tamm

•. Accounting: Accounting port type allows on-demand billing of services

used.

Interoperability.and.Trust

Obviously, a major challenge is the appropriate standardization of the above service

interfaces to enable interoperability across enterprise boundaries on one hand and
to keep flexibility to allow further evolution and differentiation of the services on
the other hand. a typical tradeoff between tight and loose coupling. Tight coupling,
with all participating end points set up to interoperate properly, might be easier to
implement and more efficient in specific cases. However, this approach will likely
result in poor reusability and adaptivity in case of varying and unforeseeable de-
ployment conditions. Here, Web services with their potential flexibility based on
XML are expected to be advantageous.
In the earlier sections we have sketched how this might be approached in a restricted
context, here, in the case of location-based services. However, to generally succeed
in dynamic value Webs it requires getting over serious hurdles:

• To increase reusability and adaptivity, service specification on higher levels

of semantics is required. Not only is the interpretation of such specifications
challenging by itself. It additionally gives rise to multiple ways of interpreta-
tion, probably resulting in unintended behavior.
• In case of independent actors, opportunistic or even malicious behavior comes
into play. Means (e.g., reputation systems) for coordination and cooperation
become essential to prevent the market of dynamic value Webs from failing.

Higher levels of semantics need to consider support for contracts, contract ne-
gotiation, and assessment of fulfillment or violation in case of discord. Possible
solutions range from peer-to-peer feedback systems to decentralized authorities
that may impose reliable reputations or sanctions on participants. One of the most
severe issues might be the support for reputation systems that aim at establishing
trust. Supporting prerequisites are: security, identifiability, authenticity, timeliness,
correctness, and other issues of dependability.
Plug and play in a global scale of interacting software components and hardware
devices is not only a problem of connectivity, but moreover of making interacting
units flexible and interoperable at a semantic level. Interacting entities need to “un-
derstand” interfaces and specifications and adapt such that they are able to utilize
each other’s abilities appropriately. Managing communication at higher levels of
semantics is the key towards integration of information pieces, software components,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 

business processes, and global distribution. Ultimately, the information infrastructure

should carry the whole spectrum of human information exchange beyond people’s
limited physical reach in space and time.
The issues to be solved spread manifold up from the physical layer, where the raw
data is transferred, to the higher levels of semantics, where data is associated with
meaning. This extends from the communication infrastructure and end-user devices
(comprising, e.g, increased communication reach, coverage, bandwidth, and usability
of devices) to high interoperability and seamless operation in dynamic and mobile
scenarios. Finding desired services as well as assessing and assembling them to
comprehensive problem solutions is highly demanding. Additionally, to cope with
the information asymmetry arising from uncertainty, ambiguity, and complexity
inherent to dynamic value Webs, creates further challenges.

Case. Study:. NOMADS. Campus. —. a. Distributed.

Spatial. Information. System. for. Berlin.Adlershof

We developed an adaptive location-based service prototype. It provides the basic

functionality of a mobile information system for the “WISTA Adlershof Science
and Technology Park” in Berlin, where natural science departments of the Hum-
boldt University are located. A widely available WLAN-infrastructure gives mobile
devices access to remote services and allows the setup of WLAN-based position-
sensing services. We have defined some basic port types allowing all companies
from the WISTA area to independently provide their location-based information
and services. In our current demonstration setup, three institutes provide a WLAN
position-sensing service and a map service as depicted in Figure 11.

Use.Case.Scenario

A user may use our NOMADS Campus system outdoors or indoors by a mobile
device that visualizes the actual position in a map. Using the system at some specific
position, the mobile device first discovers all services available for that specific
position. In the example path shown in Figure 11, the user at the outdoor position
<1> has no network connection and thus can only utilize local services. It discovers
two local services: a position-sensing service that calculates the position by an at-
tached GPS receiver and a map service where the map content is stored in the local
memory (see Figure 12). However, the map cannot comprise detailed information
since the necessary amount of data will quickly exceed the local storage capacity.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
00 Ibach, Malek, & Tamm

Figure 11. Example path of a mobile user visiting the WISTA science and technology
park in Berlin, the path comprises three situations denoted by <1>, <2>, and <3>
that require adaptive re-compostition of services and differ in availability of network
connection (WLAN), type of position sensing, and source of the map

Figure 12. In the first case the user has no network connection and utilizes local
services only

No remote servces are

avalable or wthn range,
therefore Aggregator
composeses LBS from
local servces.

WLAN: PositionSensing (GPS): PositionSensing (WLAN): MapService:

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 0

Figure 13. Mixed utilization of local and remote services in the reach of a WLAN
network

Aggregator queres newly

dscovered remote servces
for ther propertes and
decdes to use one local
and one remote servce
snce ther propertes
match best.

WLAN: PositionSensing (GPS): PositionSensing (WLAN): MapService:

Obviously, the locally provided map also cannot reflect dynamics, for example, if
building occupations have changed.
At position <2> the mobile user has WLAN network access (see Figure 13). Thus,
he can utilize remote services in addition to the local services. In our example, he
discovers a remote map service and a remote position-sensing service using WLAN
signal trilateration which we provide by our tool MagicMap (Ibach et al., 2004a;
Ibach et al., 2005a). Adaptive service composition decides for utilizing the GPS
position sensing service which is, at this outdoor position, more accurate than the
WLAN position sensing service. In addition, the remote map service is chosen since
it offers detailed and up-to-date area information. To keep pace with dynamics in
the area map, we are working on a Web crawler that scans the regional Websites
for location information and, if location can be determined, assigns an icon to the
location with a link to the Website.
If the user enters a building (see Figure 14), the GPS signal becomes unavailable and
the adaptive service composition decides to utilize the remote WLAN position-sens-
ing service instead. Note that the remote services at this position are provided be the
responsible institution, in this example, the computer science institute. Therefore, the
map service can provide an accurate indoor map containing individually-compiled
information and services, for example, room occupations.
Location-based services will also be available for “virtual travelers.” They explore
the WISTA map on the Internet that visualizes location-specific information and
stationary services. By point-and-click, it is possible to directly access these station-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ibach, Malek, & Tamm

Figure 14. In this example indoor situation, the mobile device utilizes remote
services only

2. GetPositionResponse
- N52°25'94''
- E13°32'35''
1. GetPositionRequest
One local servce becomes 3. GetMapRequest
unavalable and another - N52°25'94''
- E13°32'35''
remote servce matches - radius 100m
better than the other
local one, therefore
composton s done only
from remote servces. 4. GetMapResponse
- URL to map

WLAN: PositionSensing (GPS): PositionSensing (WLAN): MapService:

ary LBS or to forward the specific position to some general LBS. That way LBS
link virtual to physical spaces. For example, if a user is visiting the Internet site of
a company, the company’s physical location will be determined and can serve as
input for subsequent LBS. Vice versa there are LBS that link from physical to virtual
spaces, for example, one that processes instructions such as “show me the Website
of the restaurants located within 5 minutes’ walk”. In future, a position-sensing
service will be able to determine the semantic position within the virtual space as
well. For example, if position sensing detects that the user is visiting some product
information site, it can take him to product-related offers, for example, test reports
or best price comparisons.

Run-Time.Adaptation

The sequence chart (see Figure 15) shows the message sequence of service interac-
tion. The setup in this example consists of a service instance supervising the appli-
cation control flow, the registry, for example, an UDDI-implementation, two ports
connecting to position-sensing services (a GPS service and a WLAN positioning
service), and two content ports. The example indicates how the composite service
remains viable if some of its ports (here, the GPS positioning service) become
temporarily unavailable, and how on-the-fly switchover to a replacement service
(here, a WLAN positioning service2) takes place.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 0

Figure 15. Sequence chart of service interaction

The first sequence (messages 1-8) shows a service request using the available GPS
service:

1. Search the registry for position-sensing ports

2. Registry returns a GPS-receiver service port
3. Request position from returned port
4. GPS-receiver returns position
5. Search the registry for content port
6. Registry returns port of Content Provider 1
7. Request content from returned port
8. Content Provider 1 returns content data, for example, a city map in which the
building is located

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ibach, Malek, & Tamm

Before Message 9 is being sent, possibly the mobile user is entering a building,
where the GPS device cannot receive the satellite signal and therefore unregisters
its service from the registry. Supposing an in-house WLAN positioning service
becomes available, the second sequence (9-16) shows the service request after this
context change:

9. Search the registry for position-sensing port

10. Registry returns port of WLAN-positioning service
11. Request position from returned port
12. WLAN-positioning service returns position
13. Search the registry for content provider port
14. Registry returns port of Content Provider 1 and Content Provider 2
15. Supposing semantic information is available that indicates the user is inside
the building, Content Provider 2 providing corresponding content will be
prioritized and requested
16. Content Provider 2 returns content data, for example, a map that provides
location-based guidance inside the building

As the sequence chart indicates, adaptivity results from context-sensitive service

composition. Instead, the messaging behavior of each subservice remains indepen-
dent of context changes. This is possible because ports of the same port type can
be interchangeably replaced without interfering with the port’s WSDL-prescribed
request/response behavior.
Traditional monolithic LBS typically do not provide this degree of context adaptivity
(here, to switch to WLAN positioning in case the GPS becomes unavailable) without
being explicitly designed for every possible change of interoperation. Furthermore,
they hardly adapt to emerging technologies that were not foreseeable at design time.
In contrast, provided that messaging behavior of new services remains compatible
with the given type definition, ALBS can adapt to changing or newly-emerging
conditions without extra programming effort.

System.Architecture

The fundamental concept of the service-oriented paradigm is to enable uniform

access to all resources via services, including mobile or embedded devices, and
hardware resources, for example, GPS receivers or WLAN adapters. In dynamic
environments, where network topology, connections, and bandwidth are unstable and
connected devices may have limited resource power, this requires specific methods
for service propagation, discovery, invocation, and processing:

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 0

•. WS-Discovery: The Web services dynamic discovery (WS-Discovery) standard

defines a multicast protocol to propagate and locate services on ad hoc networks
in peer-to-peer manner. It supports announcement of both service offers and
service requests. Efficient algorithms (caching, multicast listening, discovery
proxies, message forwarding, filtering, scope adjustment, and multicast sup-
pression) keep network traffic for announcing and probing manageable. Thus,
the protocol scales to a large number of endpoints.
•. Lightweight.services: For efficient invocation and processing of Web services
on embedded devices with limited processing and communication power,
“lightweight” services utilize specific real-time protocols, programming lan-
guages, scheduling algorithms, message queuing policies, or XML coding and
parsing schemes.

In our example, the mobile device broadcasts its request to the resources within
local reach and collects the service announcements. The GPS receiver announces a
service for position sensing and the WLAN adapter announces two services, one for
position sensing and one for connection (see Figure 10). These ports are stored in the
local registry cache. Retrieved entries from the global registry are cached as well.
To locate a service, the discovery service is instructed to retrieve the corresponding
port type. Additionally, the discovery service can look for certain assertions to be
satisfied. Furthermore, the ALBS application communicates with local services the
same way it does with remote services. All services are propagated, discovered, and
invoked by standard Web services protocols.

Outlook. and. Conclusion

We have presented a vision and a roadmap leading to creation of a “global real-time

enterprise.” We reviewed potential enabling technologies and examined them to assess
how well they fit in a global enterprise environment. We specifically investigated
the applicability of Web Services standards in the domain of mobile computing,
in particular in the field of location-based services. We described how to flexibly
compose elementary services along typical value chains using Web services tech-
nology and achieve high adaptivity at the composite service level. Furthermore, we
outlined how location information can be processed semantically by using ontolo-
gies. As a case study, we have developed a distributed spatial information system,
NOMADS Campus, which allows composition of services available on our campus
and surrounding area according to the actual position.
We anticipate that the methodology will be applicable to future context-aware com-
puting in distributed, heterogeneous, and dynamic environments at a great degree

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
306 Ibach, Malek, & Tamm

Figure 16. The 0 allows for location-aware computing based on universal service-
oriented communication

Registry
MOBILE
Local Registry search
DEVICE services

search local register Content

Content
services Provider 22
Provider
service
WLAN
WLAN ALBS
ALBS Content
Content
Connection
Connection Application
Application Provider 11
Provider
message
GPS
GPS passing
Positioning
Positioning local
message passing
RFID
RFID
Positioning
Positioning
ENVIRONMENT

of interoperability. We are convinced that barrierless interoperability on location

semantics will tightly interconnect physical and virtual spaces and have a catalytic
impact on the markets of location-based services and mobile commerce.
The use of open Web services standards for location-based services has its appealing
strengths. And, as it turns out, Web services standards are quite suitable to overcome
a number of interoperability hurdles and to enable the evolution of dynamic value
Webs. However, a number of desirable mobility characteristics were not originally
provisioned in these standards and the integral Web services support for mobility
and ad hoc adaptivity to changing conditions is still under development. Additional
problems remain for future research to make dynamic value Webs flourish, finally
forming a global real-time enterprise. Among the most demanding, we see widely-
accepted ontologies for semantic interoperability, contracts for dependable service
specification and composition, and reputation system for justification of specifica-
tion trustworthiness.

References

Alchian, A. A., & Woodward, S. (1998). The firm is dead; Long live the firm: A
review of Oliver E. Williamsons’ The economic institutions of capitalism.
Journal of Economic Literature, 26, 65-79.
Alonso, G., Casati, F., Kuno, H., & Machiraju, V. (2004). Web services: Concepts,
architectures, and applications. Springer-Verlag.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 0

Bloomberg, J., & Schmelzer, R. (2002). The pros and cons of Web services. Zap-
Think Report. Retrieved from www.zapthink.com
BPEL4WS. (2005). Business Process Execution Language for Web Services, Version 1.1.
Retrieved from www-106.ibm.com/developerworks/library/ws-bpel
Cox, S., Daisey, P., Lake, R., Portele, C., & Whiteside, A. (2004). OpenGIS® Ge-
ography Markup Language (GML) Implementation Specification, Version 3.0
(OpenGIS Project Doc. No. OGC 02-023r4).
Erl, T. (2004). Service-oriented architecture: A field guide to integrating XML and
Web services. Prentice Hall.
Evans, J. D. (2003). OGC Web Coverage Service (WCS) Specification, Version 1.0
(OpenGIS Project Doc. No. OGC 03-065r6).
Gershenfeld, N., Krikorian, R., & Cohen, D. (2004). The Internet of things. Scientific
American, 291(4), 76-81.
Gokhale, A., Kumar, B., & Sahuguet, A. (2002). Reinventing the wheel? CORBA
vs. Web services. Presented at the Eleventh International World Wide Web
Conference, Honolulu, HI.
Harrison, M. (2003). EPC information service — data model and queries, Auto ID
center. White Paper. Retrieved from http://www.autoidlabs.org.
Hazas, M., Scott, J., & Krumm, J. (2004, February). Location-aware computing
comes of age. IEEE Computer, 37(2), 95-97.
Hodes, T. D. (2003). Discovery and adaptation for location-based services. PhD
dissertation, University of California at Berkeley, Computer Science Depart-
ment.
Ibach, P., & Horbank, M. (2004). Highly-location-based services in mobile environ-
ments. In Proceedings of the International Service Ability Symposium 2004,
Munich, Germany.
Ibach, P., Hübner, T., & Schweigert, M. (2004a). MagicMap — Kooperative Posi-
tionsbestimmung über WLAN. In Chaos Communication Congress, Berlin,
Germany.
Ibach, P., Stantchev, V., Lederer, F., Weiß, A., Herbst, T., & Kunze, T. (2005a).
WLAN-based asset tracking for warehouse management. In Proceedings of
the IADIS International Conference E-Commerce, Porto, Portugal.
Ibach, P., Tamm, G., & Horbank, M. (2005b). Dynamic value Webs in mobile envi-
ronments using adaptive location-based services. In Proceedings of the Hawaii
International Conference on System Sciences (HICSS-38), HI.
LaMarca, A., Chawathe, Y., Consolvo, S., Hightower, J., Smith, I., Scott, J., et al.
(2005). In Place Lab: Device Positioning Using Radio Beacons in the Wild,
Proceedings of the International Conferene on Pervasive Computing, Munich,
Germany.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Ibach, Malek, & Tamm

Linthicum, D. S. (2003). Next generation application integration: From simple

information to Web services. Addison-Wesley.
Malek, M. (2003). The NOMADS republic. In Proceedings of International Con-
ferences on Advances in Infrastructure for Electronic Business, Education,
Science, Medicine and Mobile Technologies on the Internet, Scuola Superiore
G. Reiss Romoli (SSGRR), Telecom Italia, L’Aquila, Italy.
Malek, M. (2004). Introduction to NOMADS. Computing Frontiers, 2004, Ischia,
Italy.
Malone, T. W., Yates, J., & Benjamin R. I. (1997). Electronic markets and electronic
hierarchies. Communications of the ACM, 30(6), 484-497.
McKee, L., & Kottman, C. (2003). OGC inside the OpenGIS specification. Open-
GIS.
Milanovic, N., & Malek, M. (2005). Architectural support for automatic service
composition. In Proceedings of the IEEE International Conference on Services
Computing (SCC 2005), Orlando, FL.
Milanovic, N., Richling, J., & Malek, M. (2004). Lightweight services for embedded
systems. Presented at the IEEE Workshop on Software Technologies for Em-
bedded and Ubiquitous Computing Systems (WSTFEUS), Vienna, Austria.
Müller, P., Stich, C., & Zeidler, C. (2001). Components @ work: Component tech-
nology for embedded systems. In Proceedings of the 27th Euromicro Confer-
ence, Warsaw, Poland.
Peng, Z.-R., & Tsou, M.-H. (2003). Internet GIS — distributed geographic informa-
tion services for the Internet and wireless networks. John Wiley & Sons.
Picot, A. (2004). Zehn Eigenschaften der Internet-Ökonomie. Retrieved from
http://www.competence-site.de
Pinto, H., Boas, N. V., & José, R. (2003). Using a private UDDI for publishing loca-
tion-based information to mobile users. In Proceedings of the ICCC/IFIP 7th
International Conference on Electronic Publishing (ElPub 2003), Guimarães,
Portugal.
Rao, B., & Minakakis, L. (2003). Evolution of mobile location-based services.
Communications of the ACM, 46(12), 61-65.
Reed, C. (2005). Integrating geospatial standards and standards strategies into busi-
ness process. Open GIS Consortium (OGC) White Paper. Retrieved from
http://www.opengis.org
Reichenbacher, T. (2004). Mobile cartography — adaptive visualisation of geo-
graphic information on mobile devices. Dissertation submitted at the Institute of
Photogrammetry und Cartography, Technical University, Munich, Germany.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Towards a Global Real-Tme Enterprse 0

Schiller, J., & Voisard, A. (2004). Location based services. Morgan Kaufmann
Schwan, K., Poellabauer, C., Eisenhauer, G., Pande, S., & Pu, C. (2002). Infofabric:
Adaptive services in distributed embedded systems. In Proceedings of the IEEE
Workshop on Large Scale Real-Time and Embedded Systems (in conjunction
with RTSS 2002), Austin, TX.
Sonnet, R. (2005). OGC Web Map Context Documents, Version 1.1 (OpenGIS
Project Doc. No. OGC 05-005).
Tamm, G., & Wuensche, M. (2003). Strategies to reduce information asymmetry
in Web service market. In Proceedings of the 11th European Conference on
Information Systems (ECIS), Napoli, Italy.
UDDI. (2005). Universal description, discovery, and integration of Web services.
Retrieved from www.uddi.org
Vretanos, P. A. (2005). OGC Web Feature Service Implementation Specification,
Version 1.1 (OpenGIS Project Doc. No. OGC 04-094).
Vogels, W. (2002). Technology challenges for the global real-time enterprise. In
Proceedings of the International Workshop on Future Directions in Distributed
Computing, Bertinoro, Italy. Retrieved from http://www.cs.cornell.edu/vo-
gels/papers/fudicobook.pdf
Whiteside, A. (2005). OGC Web Services Common Specification, Version 1.0
(OpenGIS Project Doc. No. OGC 05-008).

Endnotes
1
Wi-Fi Positioning System, www.skyhookwiereless.com; Wireless Location
Appliance 2700, www.cisco.com; Wi-Fi Workplace, www.newburynetworks.
com
2
Here, a connection switchover, for example, from UMTS to WLAN connec-
tion, will probably occur. However, this is processed analogously and is not
addressed in the Figure.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

Chapter.XIV

Mining.Critical.
Infrastructure.
Information.from.
Municipality.Data.Sets:
A.Knowledge-Driven.Approach.
and.Its.Applications
Wllam J. Tolone, Unversty of North Carolna at Charlotte, USA
We-Nng Xang, Unversty of North Carolna at Charlotte, USA
Anta Raja, Unversty of North Carolna at Charlotte, USA
Davd Wlson, Unversty of North Carolna at Charlotte, USA
Qanhong Tang, Unversty of North Carolna at Charlotte, USA
Ken McWllams, Unversty of North Carolna at Charlotte, USA
Robert K. McNally, Unversty of North Carolna at Charlotte, USA

Abstract

An essential task in critical infrastructure protection is the assessment of critical

infrastructure vulnerabilities. The use of scenario sets is widely regarded as the
best form for such assessments. Unfortunately, the construction of scenario sets is
hindered by a lack in the public domain of critical infrastructure information as

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

such information is commonly confidential, proprietary, or business sensitive. At the

same time, there is a wealth of municipal data in the public domain that is pertinent
to critical infrastructures. However, to date, there are no reported studies on how to
extract only the most relevant CI information from these municipal sources, nor does
a methodology exist that guides the practice of CI information mining on municipal
data sets. This problem is particularly challenging as these data sets are typically
voluminous, heterogeneous, and even entrapping. In this chapter, we propose a
knowledge-driven methodology that facilitates the extraction of CI information from
public domain, that is, open source, municipal data sets. Under this methodology,
pieces of deep, though usually tacit, knowledge acquired from CI domain experts
are employed as keys to decipher the massive sets of municipal data and extract
the relevant CI information. The proposed methodology was tested successfully on
a municipality in the Southeastern United States. The methodology is considered
a viable choice for CIP professionals in their efforts to gather CI information for
scenario composition and vulnerability assessment.

Introduction

A critical infrastructure (CI) is an array of assets and systems that, if disrupted, would
threaten national security, economy, public health and safety, and way of life. These
include, but are not limited to, utilities, medical facilities, public transportation,
telecommunication networks, landmarks, buildings, and public spaces. In recent
years, unfortunately, critical infrastructures have become symbolic targets as well
as the mass casualty opportunities for terrorist attacks (Bolz, Dudonis, & Schulz,
2002). For instance, the World Trade Center is a symbol of America’s capitalism
and economic influence, the Pentagon is a symbol of America’s military strength,
and the railway station in Madrid represents a node in a geo-political network. Many
critical infrastructures promote the congregation of people, which increases their
attractiveness to terrorist acts. Because of the dual identity of critical infrastructures
and the high level of vulnerability they bear, critical infrastructure protection (CIP)
has topped the list of priorities in the practice of homeland security planning in the
United States (Terner, Sutton, Hebert, Bailey, Gilbert, & Jacqz, 2004; Thieman, 2004).
Since the tragic events of September 11, 2001, CIP drills have become an integral
part of every counter-terrorism exercise across the country (Thieman, 2004).
An essential task in critical infrastructure protection (CIP) planning is the assess-
ment of CI vulnerability with respect to the threat of potential terrorist attacks. For
such a task, a set of scenarios is widely regarded in both academic and professional
communities to be the best form for such assessments (Garrick, 2002). Unlike
predictions which project critical infrastructure (CI) vulnerability with probability,
a scenario set bounds the range of vulnerabilities by connecting initiating event,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

or initial conditions, to undesired end states (different levels of damage) with the
sequence of events linking the two (Garrick, 2002). Functionally, a scenario set is
both a bridge that connects the process of CIP analysis and modeling with that of
CIP planning, and a cognitive apparatus that stretches people’s thinking and broad-
ens their views in the practice of CIP. This dual function entitles a scenario set to
be a favored member of a family of instruments for CI vulnerability assessment
and CIP planning. It also explains the popularity of scenario sets in CIP drills and
counter-terrorism exercises; they serve as a foundation for emergency response
maneuvers (Thiemann, 2004).
Despite its recognized advantages, however, the use of scenarios in CIP and homeland
security planning is hindered by the difficulties in meeting its informational needs.
To compose scenarios, a scenarist requires certain categories of information, each
corresponding to one of the five components of a scenario (Xiang & Clarke, 2003).
These five components are: (1) alternatives — the range of potential actions, or the
spectrum of incidents/events; (2) consequences — the immediate and cumulative
effects (physical, ecological, economical, and social) that each alternative would
have on an area’s security, economy, public health and safety, and way of life; (3)
causations — the causal bonds between alternatives and consequences; (4) timeframes
— the periods of time between occurrence of the alternatives and the sequence of
the consequences; (5) geographical footprints — the place-oriented blueprints of
alternatives, and the anticipated marks of their ramifications on an area’s geography.
The last component — hardly unique — is so pivotal to CIP scenario composition
and utilization that it distinguishes CIP scenarios from their counterparts in business,
industry, and at times, the military.
Satisfying these informational requirements for composing CIP scenarios is a difficult
task. A major difficulty comes from the fact that an estimated 85 percent of all CI in
the United States is owned and managed by the private sector. Consequently, much
of the information related to the security of critical infrastructures and essential to
homeland security planning and/or emergency management is not within the public
domain, as such information is commonly confidential, proprietary, or business
sensitive (Terner et al., 2004). The Protected Critical Infrastructure Information
(PCII) Program, launched by the U.S. Department of Homeland Security (DHS) in
February, 2004, under provisions of the Critical Infrastructure Information Act of
2002 (CII Act), enables the private sector to submit infrastructure data voluntarily to
the federal government. The federal government assures that competitive data will
be protected from public disclosure until and unless a PCII determination is made
that the information does not meet PCII requirements.1 Because of this voluntary
nature, the success of this milestone development depends on collaborations from
the private sector. Its efficacy remains to be seen.
On the other hand, there is a great wealth of open-source data at the municipal level
that is pertinent to critical infrastructures. These data are commonly geographic
and both collected and maintained by various municipal departments to serve the

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

general civil needs of a community. Examples include, but are not limited to, digital
orthophotos, 2 planimetric data sets, 3 tax parcel data, 4 and street centerline data.
Because of their implicit connections to critical infrastructures, and owing to some
of the important qualities they possess, high definition or precision (at a geographic
scale of 1:2,400), refinement or completeness, and general availability, these data
sets appear to be an attractive source of knowledge that await the practice of CI
information mining. 5 However, to date, there are no reported studies on how to ex-
tract only the most relevant CI information from these municipal sources, nor does
a methodology exist that guides the practice of CI information mining on municipal
data sets. This problem is particularly challenging as these data sets are typically
voluminous, heterogeneous, and even entrapping.
In this chapter, we propose a knowledge-driven methodology that facilitates the
extraction of CI information from public domain, that is, open source, municipal
data sets. Under this methodology, pieces of deep, though usually tacit, knowledge
are first acquired from a group of CI domain experts. These pieces of knowledge are
then employed as keys to decipher the massive sets of municipal data and extract
relevant CI information. The benefits to extracting CI information for a municipality
extend beyond the facilitation of vulnerability assessments for homeland security.
Such information is also particularly relevant to emergency management activities
including disaster preparedness, response, and recovery activities. Geographic
information systems (GIS) are key enablers to this methodology.
The proposed methodology was tested successfully on a municipality in the South-
eastern United States, and can potentially be applied to other municipalities and
regions. It is considered a viable choice for CIP professionals in their efforts to gather
CI information for scenario composition in support of vulnerability assessment and
emergency management for several reasons. First, the methodology produces credible
and useful results. Second, the methodology is generic and may be applied to most,
if not all, municipalities across the United States as the required resources (data,
computation, and knowledge acquisition, including domain experts) are generally,
publicly available. Third, tested on four CI domains in our study, the methodology
is flexible and thus may be adapted to any CI domains in any municipalities or
regions within the United States. Last, the methodology generates results that can
be readily refined and updated.
The remainder of the chapter is organized as follows. First, the methodology is
outlined. Next, details of the methodology are illustrated through a case study.
Case study outcomes are then described. Finally, the merits of the methodology are
discussed, and future research directions are identified.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

A. Knowledge. Driven. Methodology.................

for. CI. Information. Mining

The methodology proposed here gathers CI information through an intertwined

process of knowledge acquisition, rendering, and validation (Figure 1). After an
initial search, a knowledge worker, the person who solicits, interprets, and renders a
domain expert’s knowledge, prepares a questionnaire and sends it to a domain expert
or, in some cases, a group of domain experts, for answers. In addition, the knowl-
edge worker may schedule both structured and unstructured interviews with these
experts. The knowledge worker then geographically renders the elicited knowledge
on the generally available open-source information. The renderings are typically in
the form of digital maps, attribute tables, and rule bases. The renderings are then
reviewed by the expert(s) to assure that the knowledge worker’s interpretations are
accurate, or at least do not “violently contradict any strong held feelings” of the
expert(s) (Keeney & Raiffa, 1976, p. 271). Using the expert’s feedback, the maps,
attribute tables, and rule bases are refined. The corrected renderings are presented
to the domain expert(s) for another round of validation. This interactive process
continues until the domain expert(s) are satisfied with the results.
The remainder of the section provides a detailed account of this methodology. In
some instances the initial search may generate sufficient knowledge such that a first
rendering may occur prior to contacting domain experts. This initial rendering then
serves as input to domain expert discussions.

Initial.Search

The process of CI information mining begins with an initial search. This search
involves a set of three tasks: (1) a survey of all the publicly-available information

Figure 1. A methodology for critical infrastructure information mining

Intal Search

Knowledge Acquston

Knowledge Renderng

Knowledge Valdaton

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

related to the domain; (2) the identification of individual domain experts for each
CI component; and (3) the development of a list of questions for the interviews of
the domain experts.
Publicly-available information usually resides in open source locations such as
generally-accessible databases, journals, books, newspapers, the Internet, and even
word-of-mouth. Much of the information from these sources is non-confidential,
insensitive to business interests, and bearing little insights into the key issues related
to CIP planning and management. However, both the information and the search
process are necessary and valuable for CI information mining. They help knowledge
workers become familiar with the problem domain; and they enable knowledge
workers to identify individual domain experts and develop questionnaires in order
to proceed to the next stage of the information mining process.
It should be noted that the thoroughness of the initial search contributes significantly
to the success of the subsequent stage of knowledge acquisition. First of all, it
makes the interviews, structured and unstructured, more productive while reducing
the time commitment of domain experts, who usually have limited resources and
in many cases must justify the hours spent with the knowledge workers. Secondly,
domain experts are more likely to be forthcoming when knowledge workers are
well prepared; although, this forthcoming nature of domain experts is independent
of whether the knowledge worker utilizes a questionnaire during the interview.

Knowledge.Acquisition

The task of knowledge acquisition deals with two levels of CI knowledge, surface
and procedural. Surface knowledge is the most common information offered when
seeking information on a subject. Surface knowledge is also referred to as declarative
knowledge (Turban & Aronson, 2001). Such knowledge includes facts and figures
that can usually be found easily at the beginning of knowledge acquisition, and in
some cases, during the stage of initial search. Examples include, but are not limited
to, schematic diagrams of a CI system hierarchy, iconographical representations of
CI roles, and functional relationships among objects within a CI layer. The CI pro-
prietors usually are willing to give this generalized information to interested parties
as it satisfies most inquiries and raises little corporate security concerns.
Procedural knowledge (Turban & Aronson, 2001), on the other hand, is not easily
accessible and typically not disseminated to outside seekers for security concerns.
Procedural knowledge includes information such as safety protocols, problem
diagnostic procedures, and emergency response plans. Furthermore, the spatial
element of both surface and procedural knowledge, that is, information about the
geographic footprints of CIs, either as individual layers or as components of a public
utility system (e.g., a traffic control system), is usually regarded as highly sensitive.
Consequently, information about locations, the geographic networks of CI objects,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

knowledge about proximity, and the spatial interactions among CI objects are not
readily available.
Various methods have been suggested for knowledge acquisition. These include,
but are not limited to, the method of “familiar tasks” (Duda & Shortliffe, 1983;
Mittal & Dym, 1985; Stefik et al.,1982), structured and unstructured interviews
(Hoffman, 1987; Weiss & Kulikowski, 1984), limited information tasks (Hoffman,
1987), constrained processing tasks (Hoffman, 1984, 1987; Klein, 1987), and the
method of “tough cases” (Doyle, 1984; Hoffman, 1987). The method of “familiar
tasks” is an observational method involving domain experts performing typical
tasks. Structured and unstructured interviews are examples of question/answer
methods for knowledge elicitation. The method of limited information tasks is
similar to the method of “familiar tasks” except that the information presented to
a domain expert is restricted in some way in an attempt to expose how an expert
derives a solution. The method of constrained processing tasks is an observational
method where the time a domain expert is allowed to reach a conclusion is limited.
Finally, the method of “tough cases” involves presenting a domain expert with an
unfamiliar case and requiring the expert to describe the protocol that he/she is using
while deriving a solution.
Among these methods, interviews (both structured and unstructured) are the most
common and widely accepted approach to knowledge elicitation (Hoffman, 1987;
Millet & Harker, 1990; Xiang & Whitley, 1994). Although interviews are usually
more time-consuming and less efficient than other methods (such as limited infor-
mation tasks and constrained processing tasks), domain experts generally feel more
comfortable with interviews and, therefore, are less hesitant to offer opinions (Ibid.).
For example, during an interview conducted through our case study, a domain expert
offered a detailed physical description of a critical node within the telecommunica-
tion network, providing details on square footage, parking area, number of stories,
and surrounding land use. This invaluable tacit knowledge would have remained
undiscovered without such an interview.
At this stage of CI information mining, a combination of structured and unstructured
interviews is used to acquire deep procedural knowledge, following an approach
developed by Xiang and Whitley (1994). The unstructured interview consists of an
informal discussion with domain experts, where the knowledge worker conducts
an unscripted conversation and slowly builds his/her understanding of the infra-
structure. Interviews with these individuals usually last about an hour and consist
of an informal question and answer session. The structured, more formal interview
is usually conducted afterwards, where the knowledge worker submits a list of
detailed questions to solicit domain experts’ explicit inputs. A structured interview
often further refines the knowledge acquisition process, providing greater clarifica-
tion to the knowledge extracted during the unstructured interviews.
Ideally, a group of experts from each domain should be interviewed to minimize
individual biases. Under this multi-expert approach, the knowledge acquisition
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

process is conducted either in an interactive group setting or through individual

interviews with domain experts. A group setting is considerably more time efficient
because a knowledge worker can solicit input from multiple domain experts in a
single session. However, there are several drawbacks associated with this method,
including domination by one individual, lack of focus, and limited participation
by all members (Moore, 1987). With individual interviews, on the other hand, the
knowledge worker will need to consolidate all of the information obtained from the
domain experts. This process can be cumbersome, and time-consuming, especially
when the information is contradictory (Roth & Wood, 1990). In practice, therefore,
the ideal multi-expert approach may yield to a sub-optimal single-expert approach
for pragmatic reasons.
Another way to de-bias elicited knowledge is to use a shared information approach
that encourages communications among the knowledge workers. During the interview
process, each knowledge worker is assigned to a specific domain, working individu-
ally with domain experts. Under this shared information approach, the knowledge
workers share with one another the knowledge they solicited from various domain
experts through informal discussions at regular intervals. These “cross-domain”
discussions can provoke questions that could prove useful in follow-up interviews
with the respective domain experts. In our case study, for instance, a final, formal
presentation of knowledge acquisition results proved to be a useful communication
session for both knowledge workers and domain experts. The knowledge workers
benefited from the additional informal knowledge validation provided by domain
experts as well as from the information obtained from co-worker presentations.
It also provided an excellent opportunity for domain experts of the various utility
sectors to discuss their interdependencies.

Knowledge.Rendering

Knowledge rendering is a process in which knowledge workers interpret the knowledge

elicited from domain experts and express this understanding using an appropriate
representation. More specifically, in the case of CI information mining, knowledge
rendering refers to the exercise through which knowledge workers develop a spatial
representation of the infrastructure layer based on the knowledge acquired from the
domain expert(s). Geographic information systems (GIS) are a key enabler to this
step of the proposed methodology. In our case study, for instance, a domain expert
supplied the knowledge worker with a detailed description of the physical charac-
teristics of a critical node in a telecommunication network (for example, parking lot
size, building height, etc.). To render this piece of knowledge spatially, the knowledge
worker first selected all the service provider’s properties from a publicly available
tax parcel data file. The worker then superimposed these selected land parcels onto
digital orthophotos of the study area, and removed all the parcels that did not match

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

the physical characteristics. The worker finalized the rendering by saving all the
remaining parcels on a new map. The map is referred to as a proxy or surrogate
data layer in that it is not a map of the real data per se. Instead, it is a geographic
rendering of the domain expert’s knowledge on the open-source municipal data
(tax parcels and digital orthophotos, in this case) as interpreted by the knowledge
worker. In addition to maps, the rendering of domain expert’s knowledge can take
other forms, such as iconographic representations, flowcharts, and tables.

Knowledge.Validation

During the final phase of the CI information mining process, the knowledge worker
submits the newly-rendered proxy data layer to the domain expert for validation.
A series of structured or unstructured interviews are scheduled, where the domain
expert is asked to verify that the spatial renderings are consistent with the knowledge
that he/she provided in the previous interview(s). The proxy data layers submit-
ted to the domain experts are usually presented as digital maps. These layers are
continuously updated and refined during the process based on feedback from the
domain experts. This phase is complete when the domain experts are satisfied with
the knowledge rendering as presented and have verified that the knowledge worker
did not misrepresent any of the knowledge.

Implementation. of. the. Methodology:.A. Case. Study

The above-described methodology was initially developed to mine information

from four critical infrastructures in a municipality in the Southeastern United States.
These infrastructures include the electric power grid, telecommunication, natural
gas distribution, and transportation networks. As an illustration of the application
of the methodology, this section provides a detailed description of mining exercise
for information about the natural gas supply system.
A pipeline network supplies natural gas to residential and commercial customers
throughout the metropolitan area. Commercial pipelines transmit large volumes of
natural gas at a high capacity of 420 PSI (pounds per square inch, 1 PSI = 6895
Pascals).6 Residential pipelines distribute small volumes at a capacity of 70 PSI or
lower. While this case study involves only one gas distribution domain expert, the
techniques were successfully applied to groups of domain experts, as described
in the previous section, during an extended case study that also included electric
power, telecommunications, and transportation infrastructures. Discussion in the
following section, however, is limited to the focused case study.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

Initial.Search

Because of the security concerns with gas pipeline data, it is not surprising that the
initial search for publicly-available data yielded only modest results. Among the
most relevant are the pipeline system’s general structure and a digital map of the
major transmission pipelines from Texas to New Jersey. In addition, the knowledge
worker collected information about the domain experts who are affiliated with the
natural gas provider in the region, and the information on the land parcels designated
as easement for natural gas providers.

Knowledge.Acquisition

Through a series of structured and unstructured interviews with a gas distribution

domain expert, the knowledge worker obtained two important pieces of knowledge
about the locations of the local pipeline network and the physical characteristics of
the regulator stations, the nodes on the network. During a visit to the company, the
knowledge worker was also given an image file, a screen shot from a GIS data file,
of the local pipeline network. From the image file, it was clear that the easement
areas on the property parcel data set are major pipelines in the municipality area.

Knowledge.Rendering

The process of knowledge rendering consisted of three steps: construction of the

regional pipelines (higher than 70 PSI); identification of the regulator stations, and
creation of a local pipeline network (70 PSI and lower). This process proceeded
under the guidance of the knowledge acquired from the domain expert.
In constructing the regional pipelines, the knowledge worker first superimposed a
map layer of the easement areas on top of the digital orthophotos, and found some
interesting physical characteristics of these easement areas: flat, open, linear, around
15 feet in width, either a little higher or lower than the surrounding ground surface.
With the image file of the local pipeline network provided by the natural gas company
as a reference, the knowledge worker then created a proxy of the regional pipelines
by connecting those easement areas.
To identify the locations of the regulator stations, the knowledge worker first selected
from a tax parcel database all the land parcels that are owned by the gas company
and put them on a new map. The knowledge worker then superimposed the map on
the digital orthophotos to examine, within each land parcel owned by the natural gas
company, whether the physical characteristics of the regulator stations described by

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

the domain expert were found. Typically, a regulator appears to be a small structure
on a digital orthophoto with an electrical generator in the back. Another charac-
teristic is that regulators are usually located at regular intervals along the pipeline.
The knowledge worker then placed a node on the identified regulator to represent
its location. Finally, all the nodes were saved on a new map layer.
The creation of local distribution pipeline network (70 PSI and lower) was guided
by rules acquired from the domain expert. These rules are (1) 90% of the pipes
follow the collector and neighborhood streets, and roughly 10% of the streets in
a municipal street centerline file do not coincide with pipelines; (2) pipelines do
not follow major highways — interstates and arterials; and (3) newly-developed
neighborhoods that do not have natural gas supplies are shown on the image file
that the company provided as empty areas. Applying these rules to the street cen-
terline files of the study area, the knowledge worker produced a map of the proxy
distribution pipeline network.

Knowledge.Validation

The completed renderings of the gas pipeline network and regulator stations proxy
data layers were submitted to the domain expert for validation. Again, because of
security concerns, the domain expert did not explicitly confirm whether the render-
ings were “right” or “wrong.” He instead directed the knowledge worker’s attention
to places where erroneous assumptions in the pipeline placement had occurred.
Based on this feedback, the knowledge worker made adjustments on the pipeline
alignments on the digital maps, and checked on the tax parcel database to assure
that the updates were thorough and complete.
It should be noted that during the process of knowledge acquisition, rendering, and
validation, spatial techniques equipped in GIS and open-source data sets, especially
digital orthophotos, played a significant role. By showing the geographic objects
in their actual forms, rather than cartographic symbols (points, lines, areas, or pix-
els) (Lillesand & Kiefer, 2000), digital orthophotos provide an advanced tool that
was far more effective and efficient than the traditional maps and aerial photos for
knowledge rendering and validation. The use of GIS enables the knowledge work-
ers to manage the complex task of rendering an entire utility network. It helps track
various features such as transmission poles, transmission stations, and gas pipelines
all within a same spatial domain. In this fully integrated system, all of the features
are related by location, and multiple data layers can be dynamically linked for visual
analysis and for map production (Harder, 1999).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

Products. of. the. Information. Mining. Process

The surface and procedural information gathered through the application of the
described methodology can be categorized into two interrelated groups: informa-
tion about the functionality of a CI system, and information about the geography
of CI system structures. This section presents a description of some case study

Figure 2. A functional hierarchy of the natural gas distribution system

Man Transmsson Ppelne Level 

0 psg ppelne

Level 
 psg ppelnes

Level 
0 psg ppelnes

Level 
Connecton to User

Figure 3. A geographic rendering of the natural gas distribution system

Gas regulator Legend

0 Psg natural gas transmsson ppe lne
 Psg natural gas ppe lne
0 Psg natural gas ppe lne

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

outcomes as related to the natural gas distribution system for the case study region.
The presentation of these outcomes is limited due to the sensitive nature of the
information gathered.
Figure 2 shows the hierarchy of the natural gas pipeline network. This information
is principally surface information about natural gas distribution. Level 1 represents
the main national gas pipeline (e.g., the West/East pipeline running from Texas to
New Jersey). The 175 PSI lines at Level 2 transmit natural gas to the municipal
area. The Level 3 network (70 PSI or lower) distributes natural gas to the local
service areas.
A geographic representation of the natural gas distribution network in the study area
is provided in Figure 3. This information is principally procedural as it is specific
to the region of interest. The thick line in the lower right corner of the map repre-
sents the main national gas pipeline. The medium weight line models the regional
pipeline that encircles the municipal boundary. Gas regulators connect the region
pipeline to local service pipelines.

Conclusion

We contend that the proposed methodology is a viable choice for CIP professionals
in their efforts to gather CI information for scenario composition and vulnerability
assessment for several reasons. First, the methodology produces credible and useful
results, as demonstrated through the case study (the CI information gathered was
credible in the eyes of domain experts). Second, the results are also useful for CIP
scenarists in composing scenarios. The CI information from this case study has
successfully supported a CI simulation project funded by a federal agency (Tolone
et al., 2004). Third, the methodology is executable for most, if not all, municipali-
ties across the United States as the required resources for data, computation, and
knowledge acquisition, including domain experts, are generally available. Fourth,
the costs associated with methodology implementation are modest, most of which
are related to the time and expenses of knowledge workers. Fifth, the methodology
is adaptable and can be applied to any CI domain, although only four domains of
CI information were involved in this case study. Finally, not only can methodology
results be readily refined and updated, but they can also be incorporated into a real
CI database should it become available under the Protected Critical Infrastructure
Information (PCII) Program.
The principal limitation of the proposed methodology is that it depends upon ac-
cess to domain experts. Municipalities, however, typically have relationships with
location utility providers, which can facilitate domain expert identification and ease
access issues. The fidelity of methodologies outcomes is directly tied to the level

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

of access to domain experts and the willingness of such experts to be forthcoming

in their participation.
The case study application of the proposed methodology highlighted two important
lessons learned. As previously noted, the geospatial renderings of gathered informa-
tion were critical to effective participation by the domain experts in the knowledge
validation step. Without such renderings, the knowledge validation step would be
significantly hampered. Geographic information systems are a key technology to
generating such renderings. In addition, knowledge validation using domain ex-
perts must be conducted systematically to be effective. Improved techniques and
methodologies for verification and validation of infrastructure models constitute
the primary direction for future research.

Acknowledgments

The authors are grateful to the following individuals at the University of North
Carolina at Charlotte for their contributions to the project: Bei-Tseng Chu, Mirsad
Hadzikadic, Vikram Sharma, Deepak Yavagal, Robin Gandhi, Katie Templeton, Gus-
tavo Borel Menezes, Stuart Phelps, Jocelyn Young, Paul Smith, and Huili Hao.

References

Benbasat, I., & Dhaliwal, J. S. (1989). A framework for the validation of knowledge
acquisition. Knowledge Acquisition, 1, 215-233.
Bolz, F., Dudonis, K., & Schulz, D. (2002). The counterterrorism handbook: Tactics,
procedures, and techniques (2nd ed.). Boca Raton, FL: CRC Press.
Casazza, J., & Delea, F. (2003). Understanding electrical power systems. New
York: John Wiley & Sons.
Doyle, J. (1984). Expert systems without computers or theory and trust in artificial
intelligence. AI Magazine, 5(2), 59-63.
Duda, R. O., & Shortliffe, E. H. (1983). Expert Systems Research. Science, 220,
261-276.
Frawley, W., Piatetsky-Shapiro, G., & Matheus, C. (1992). Knowledge discovery
in databases: An overview. AI Magazine, 13(3), 57-70.
Garrick, J. (2002). Perspectives on the use of risk assessment to address terrorism.
Risk Analysis, 22(3), 421-423.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Tolone, Xang, Raja, Wlson, Tang, McWllams, & McNally

Harder, C. (1999). Enterprise GIS for energy companies. Redlands, CA: ESRI
Press, Inc.
Hoffman, R. R. (1984). Methodological preliminaries to the development of an
expert system for aerial photo interpretation (Tech. Rep. No. TRETL-0342).
Fort Belvoir, VA: Engineer Topographic Laboratories.
Hoffman, R. R. (1987). The problem of extracting the knowledge of experts from the
perspective of experimental psychology. AI Applications in Natural Resources
Management, 1(2), 35-48.
Keeney, R. L., & Raiffa, H. (1976). Decision with multiple objectives: Preferences
and value tradeoffs. New York: John Wiley & Sons.
Klein, G. A. (1987). Applications of analogic reasoning. Metaphor and Symbolic
Activity, 2, 201-218.
Lillesand, T., & Kiefer, R. (2000). Remote sensing and image interpretation. New
York: John Wiley and Sons, Inc.
Millet, J., & Harker, P. T. (1990). Globally effective questioning in the analytic
hierarchy process. European Journal of Operational Research, 48, 88-97.
Mittal, S., & Dym, C. L. (1985). Knowledge acquisition from multiple experts. AI
Magazine, 6(2), 32-36.
Moore, C. M. (1987). Group techniques for idea building. Newbury Park, CA:
Sage Publications, Inc.
Roth, M., & Wood, C. (1990, October 31-November 2). A Delphi approach to
knowledge acquisition from single and multiple experts. In Proceedings of the
1990 ACM SIGBDP Conference on Trends and Directions in Expert Systems,
Orlando, FL (pp. 301-324).
Stefik, M., Aikins, A., Balzer, R., Benoit, J., Birnbaum, L., Hayes-Roth, F., &
Sacerdoti, E. (1982). The organization of expert systems: A tutorial. Artificial
Intelligence, 18, 135-173.
Terner, M., Sutton, R., Hebert, B., Bailey, J., Gilbert, H., & Jacqz, C. (2004, March 1).
Protecting critical infrastructure. GeoIntelligence. Retrieved from http://www.
geointelmag.com/geointelligence/article/articleDetail.jsp?id=90043, 1-2.
Thieman, K. J. (2004). The use of scenarios and geographic information systems and
technology (GIS &T) in counter-terrorism exercises and planning. Master’s
thesis, Department of Geography and Earth Sciences, University of North
Carolina at Charlotte.
Tolone, W. J., Wilson, D., Raja, A., Xiang, W. N., Hao, H., Phelps, S., & Johnson,
E. W. (2004). Critical infrastructure integration modeling and simulation. In
Proceedings of the 2nd Symposium on Intelligence and Security Informatics
(ISI 2004) (LNCS 3073). Berlin: Springer-Verlag.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Mnng Crtcal Infrastructure Informaton from Muncpalty Data Sets 

Turban, E., & Aronson, J. E. (2001). Decision support systems and intelligent sys-
tems. Upper Saddle River, NJ: Prentice Hall.
Weiss, S., & Kulikowski, C. (1984). A practical guide to designing expert systems.
Totowa, NJ: Rowman and Allanheld.
Xiang, W. -N., & Clarke, K. C. (2003). The use of scenarios in land use planning.
Environment and Planning B: Planning and Design, 30, 885-909.
Xiang, W. -N., & Whitley, D. L. (1994). Weighting land suitability factors by the
PLUS method. Environment and Planning B: Planning and Design, 21, 273-
304.

Endnotes
1
http://www.dhs.gov/dhspublic/interapp/editorial/editorial_0404.xml
2
Digital orthophotos are digital versions of aerial photographs that are constructed
to rectify image displacement due to changes in aircraft tilt and topographic
relief.
3
Planimetric data sets are the digital version of planimetric maps that present
only the horizontal positions for features represented. They are distinguished
from topographic maps by the omission of relief in measurable form. A plani-
metric map intended for special use may present only those features essential
to the purpose to be served. Features that are usually shown on a municipal
planimetric map include streets, utility lines, building footprints, tree lines,
water bodies, and contour lines.
4
Tax parcel data delineate land parcel boundaries and identify those people or
institutions that bear the tax burden for each parcel.
5
Data mining (also known as knowledge discovery in databases—KDD) is
defined as “The nontrivial extraction of implicit, previously unknown, and
potentially useful information from data” (Frawley et al., 1992, p. 58).
6
In figures and maps, PSI is referred to as “psig” — “pounds per square inch
pipe.”

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

Chapter.XV

Cognitive.Mapping.and.
GIS.for.Community-Based.
Resource Identification
Lyn Kathlene, Colorado State Unversty, USA

Abstract

This chapter describes and analyzes the effectiveness of two methodological

techniques, cognitive mapping and geographical information systems (GIS),
for identifying social service resources. It also examines the processes used to
integrate hand-drawn map information into geocoded data points and provides
recommendations for improving efficiency and precision. As a first step to inte-
grate Jefferson County social service delivery into community-based child welfare
“systems of care” (SOC), both formal and informal services had to be identified.
Cognitive mapping, a process by which participants draw visual representations
of geographical areas, was conducted with 247 participants in Jefferson County,
Colorado. Over 3,500 resources were identified and entered into a GIS to analyze
the availability, capacity, and distribution of social services in the county and within

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 327

communities. Identification of community resources via cognitive mapping and GIS

analysis provide: (1) a comprehensive database of existing services; (2) a basis to
build communication networks and cooperation among government and community
providers; (3) the ability to create an efficient system that avoids duplication of ef-
forts; (4) an understanding of the geographical distribution of resources; (5) the
identification of resources lacking in the county and specific communities; and (6)
knowledge differences among diverse participant groups.

Introduction

In December, 2003, the Colorado Institute of Public Policy (CIPP) at Colorado State
University was contracted by Jefferson County, Colorado, Division of Human Ser-
vices, to conduct a resource identification analysis. The project was one component
in the first year of a five-year Health and Human Services — Children’s Bureau
grant to create a “systems of care” (SOC) in child welfare social service delivery.1
Jefferson County, Colorado, was one of eight pilot sites awarded an SOC grant.
The CIPP component was to identify services available at the community-level and
discover services that were lacking.
SOC is a major paradigm shift in social service delivery. It removes the locus of
authority away from one individual, the social service worker, and replaces it with a
group of service providers, family, and community members to develop collectively
a comprehensive plan to move the child and family out of crisis. The provision of
services are to be coordinated, community-based, culturally competent and indi-
vidualized (Stroul, 1986).
To integrate Jefferson County social service delivery into community-based compre-
hensive child welfare SOC, both formal and informal services had to be identified.
Informal services are of particular interest since these are likely the least well-known
(there was no official directory) and serve populations at a community SOC level
(rather than county-wide). For definition purposes, informal services were identi-
fied for participants as private or not-for-profit programs, including services such
as church soup kitchens, non-profit agencies providing transportation services for
the elderly, and in-home daycare providers not registered with the county. Formal
services are public programs at the state, county, and local level, such as Jefferson
County Mental Health Services, Title XX daycare providers, public schools, public
transportation, and park and recreation programs.
To identify existing resources at the community level, cognitive mapping, a process
by which participants draw visual representations of geographical areas, was con-
ducted with 247 participants in Jefferson County, Colorado. Participant groups in
the mapping included social service, non-profit, and faith-based providers, social

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

service clients (Temporary Assistance for Needy Families (TANF) recipients, youths,
foster care providers, and adoptive parents), residents and ethnic/racial enclaves
(Latino, Eastern European, Native American, and African American). In addition,
all resources listed in the Jefferson County resource guides were included in the
resource database. Over 3,800 unique resources were identified and entered into a
GIS — ArcMap, a component of ArcView — to analyze the availability, capacity,
and distribution of social services in the county and within communities. Census
data was overlaid to identify high-need areas and ethnic enclaves.
Here, a novel application of GIS for designing improved social service delivery
systems is described. The chapter also discusses complications involved in working
with human service agencies, and reconsiders the processes developed to merge
cognitive mapping information into ArcMap.

Cognitive.Mapping:.Origin.and.Uses

Cognitive mapping has not been comprehensively integrated with GIS to the degree
discussed in this project, although elementary integration has been done in some
previous projects (Fulton, Horan, & Serrano, 1997; Kathlene, 1997; Kathlene &
Horan, 1998; Horan, Serrano, & McMurran, 2001).2 The potential usefulness to
the human services sectors through projects that use the combined methodologies
is substantial. To better understand this “fit,” a brief review of cognitive mapping
follows.
Cognitive mapping did not originate from research on humans. Rather, the term
“cognitive map” was originally used to describe the mental representations that rats
develop as they navigate the same maze multiple times (Tolman, 1948). Quickly,
researchers became interested in using the information from cognitive maps in the
human context, and the resulting collection of methods became known as cognitive
mapping. Later, the term expanded to include mental depictions of more abstract
entities, like ideas or chains of events. Since its inception, cognitive mapping has
been used as an approach to a number of real-world issues.
Cognitive mapping’s first practical application was in the field of urban planning
when Kevin Lynch (1960) found that certain places or elements in a city generated
a positive emotional reaction among its residents. This type of research, he felt,
could be used to better design cities so as to make them more memorable. Jack
Nasar (1988) extended Lynch’s study by including a model of how city attributes
influenced the affective responses of residents and visitors.
Cognitive mapping is also used to evaluate mental representations of smaller areas.
Mapping of specific neighborhoods (Quaiser-Pohl, Lehmann, & Eid, 2004; Uzzell,
Pol, & Badenas, 2002), college campuses (Hardwick, Wooldridge, & Rinalducci,
1983; Holahan & Dobrowolny, 1978; Sholl, 1987), and buildings (Moeser, 1988;
O’Laughlin & Brubaker, 1998; O’Neill, 1991) adds to knowledge about how spatial
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 329

abilities develop and are utilized, the skills related to map creation, and the mapping
abilities of various groups. Numerous overlying maps of these smaller areas have
been used to program autonomous mobile robots (Yoshino, 1991).
Almost anyone can successfully complete a mapping exercise. Studies have found
the quality of female and male’s maps are similar (Evans, 1980; Magana & Nor-
man, 1980; O’Laughlin & Brubaker, 1998); nearly all ages can successfully par-
ticipate (Quaiser-Pohl, Lehmann, & Eid, 2004); and map quality is not influenced
by drawing skill (Evans, 1980; Hardwick, Wooldridge, & Rinalducci, 1983). The
only criteria that affects map quality is familiarity with the target area, where qual-
ity increases with familiarity (Evans, 1980; Fridgen, 1987; Unger & Wandersman,
1985). Familiarity, however, can distort maps. For example, locations of importance
to the participant are often drawn larger or more centrally than other map elements
(Holahan & Dobrowolny, 1978; Kathlene, 1997). Finally, at the coding stage, the
use of multiple coders without knowledge of the study hypothesis is commonly
used (Daniels & Johnson, 2002); yet, studies have found nearly perfect agreement
among multiple coders (Quaiser-Pohl, Lehmann, & Eid, 2004).
For this project, the above issues did not present problems. Each of the partici-
pants was directed to draw the area they self-identified as their most recognizable;
therefore, they had high levels of familiarity. The maps required particularly little
drawing skill as participants were only asked to draw boundaries (for which they
are provided a ruler), mark their home with an “X,” and identify landmarks with
squares. Since the objective in the study was to identify resources of importance
to the participant; having the maps drawn to scale was irrelevant because the exact
geographic coordinates for each location are determined during the coding process.
Coding the maps involved recording objective location information, thereby remov-
ing the problem of coder subjectivity. In short, information gleaned from the maps
was not dependent upon the drawing skills of the participants.

Adaptation. of. Mapping. Methods...................

for. Resource Identification

Each use of mapping described above required a unique adaptation of the method.
Participants in these studies were asked to sketch maps, arrange pictures of pieces
of paper according to their location or importance, locate points on a map, circle
areas on a map, recognize features on an aerial photograph, or label features on a
blank map. The literature effectively suggests that as many methods of mapping
exist as purposes for using it.
In this project, participants attending a mapping workshop were first provided back-
ground on the overall goals of the SOC project as well as their role as a mapper in

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
330 Kathlene

Figure 1. Example of hand drawn cognitive map

achieving those goals. The hope was that understanding the importance of the project
would encourage conscientious participation. Then, participants were provided with
a large blank sheet of paper and asked to indicate the street or physical boundaries
within which they most commonly travel. Street maps of the area were available to
help them in this and later tasks. Workshops consisted of three to sixteen people,
and at each session they were encouraged to work cooperatively with their fellow
workshop participants. Following a visualization exercise to assist them in access-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 331

ing non-verbal memories, they were asked to fill in details such as streets, natural
features, and resources they commonly use. It was explained that these maps were
not meant to be artistic works; simple boxes and lines drawn with their rulers would
suffice to represent objects and locations. Once the maps were complete, partici-
pants were asked to label various types of resources with colors corresponding to
commercial businesses, community resources, government agencies, community
strengths, and community weaknesses. They also completed a survey asking for
detailed description of the important resources and features on their maps and to
provide any additional information regarding resources lacking in their area. Each
map was then entered into a GIS database. Figure 1 is an example of a cognitive
map drawn for the Jefferson County Project.

GIS.Analysis. Technique

The GIS for this cognitive mapping project provides aggregated spatial analysis of
the maps. GIS combines the thousands of data points collected from the participant-
drawn maps, which greatly enhances the value of cognitive mapping information. In
addition, each distinct physical address has dozens of descriptors that identify not
only who provided the information but also secondary data such as the participants’
perceptions of a particular resource or service (Longley, Goodchild, Maguire, &
Rhind, 2001). The data is accessible in Excel-type tables that can be easily queried
and displayed on a digital street map (Longley, Goodchild, Maguire, & Rhind,
2001). This analysis allows for preliminary interpretation of the data set, such as
the dispersion or clustering of resources in geographic areas (Heagerty & Lele,
1998; Reich & Davis, 2003). GIS can incorporate external data sources, such as
census data, which can be layered with the cognitive mapping data. Integrating the
cognitive mapping data with GIS software creates nearly limitless ways to analyze
the data and allows for easily-interpreted visual results.

Technical.Capacity.for.the.Project

For this project the particular GIS capabilities were:

• Visual representation of addresses and areas via data points, data lines, and
data shapes
• The ability to layer data points for sorting, categorizing, and selection pur-
poses

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

• Accompanying tables of descriptive variables for each data layer with the abil-
ity to search and sort individual data points based on particular variable(s)
• The ability to export the data to conduct advanced spatial analysis

The cognitive mapping process was greatly enhanced, especially for analysis
purposes, by using GIS to create unique, specialized parameters (for a theoretical
model, see Jordan, Raubal, Gartrell, & Egenhofer, 1998). The potential outcomes
of such a union include:

• The capability to export a large quantity of data in an Excel-type format that

can be searched and sorted based on any given number of criteria
• The ability to organize and sort data in a spatial manner (i.e. in the form of
maps)
• The opportunity to create an easily-updated database for organizing additional
data gathered by the cognitive mapping process and through other resource
identification processes

Structure.of.the.GIS.for.the.Project

The structure of the GIS system for this project was based on three principles:

1. The structure needed to be flexible and allow for the maximum amount of data
to be stored and easily accessed
2. The structure needed to be relatively simple such that an inexperienced GIS
user can understand the sorting, searching, and selecting potential of the sys-
tem
3. The GIS system needed to be able to organize data from a variety of sources
and be easily updated over time

The first step to organizing the data was to create basic social service or resource
categories (in GIS terms these categories are referred to as layers). Each layer
needed to be broad enough to encompass a variety of unique data points (identified
resources) but narrow enough to create a good classification system for a begin-
ning search and find procedure. These basic layers were the essential framework
of the GIS system and the search capabilities of the system that providers could
use; therefore, they had to match either the basic social service field breakdown of
resource types and/or be intuitively organized.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 333

Ultimately, 24 layers were created for the project including Housing, Mental Health
Services, Transportation, Health Care, Emergency Services, Domestic Violence, and
Victim Services. Within each layer were key fields with similar categories across
all the layers. For example, the data field “Resource Type” exists in each of the 24
layers and contains only one of five values: Governmental, Community Non-Profit,
Faith-Based, Private, or Unknown. This allowed queries and linkages to be made
across layers. As an illustration, a user could search for all the community non-profit
sponsored resources or services in two or more of the layers and combine these
results on one map.
Search or selection functions work on a layer by layer level in ArcMap. Within the
Housing layer there were attributes such as Resource Name, Resource Address,
and Resource Hours of Availability. These attributes are easily sorted. For example,
there should only be one Lakewood Housing Authority (LHA), but if there are two
locations for LHA, they can be distinguished from one another in the data table
due to the different values entered in the Resource Address cell. However, not all
searches are done based on a known name or location; therefore, a more extensive
list of attributes must exist to properly describe each resource. Some examples of
other attribute identifiers were Resource Type (government-sponsored, private,
faith-based, or community non-profit) or Funding Source. Another type of attribute
is a binary descriptor such as Used by Respondent Yes/No. This would apply to data
points identified by clients and useful in various types of qualitative analysis. Below
is a simplified example of a layer table structure and associated attributes.
It is easy to see in Table 1 the large number of possible searches based on only
seven attributes. For example, a provider might want to find all government-spon-
sored housing resources. The search would use “Resource Type=Government.” The
first two rows would be selected from the search. If housing resources provided
by community non-profits is of interest, the search would be written as “Respon-
dent Category=Client and Resource Type=Community Non-Profit and Used by
Respondent=Yes.” In the table above, this would select only the last row.

Table 1. Housing layer attribute table

Survey Respondent Resource Resource Hours of Funding Used by

ID Category Name Type Availability Type Respondent
0105 Client LHA Government 9am-5pm Government No

0431 Provider LHA Government 9am-5pm Government N/A

0302 Community Lakeside Private N/A N/A Yes

Member Apartments
0746 Client Allison Care Community N/A Private Yes
Center Non-Profit

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

Each row represents one unique data point on the GIS map and therefore represents
one unique resource. Resources can be identified and placed on the map multiple
times, as in the above example, with the first and second row both being the Re-
source Type “LHA.” What distinguishes the two data points, however, is not the
location on the map but the multiple attributes in the table listed above. Here, LHA
was identified once by a provider and once by a client, information that indicates
common knowledge. Alternatively, if only clients identify the Allison Care Center, a
knowledge gap in Jefferson County Human Services providers and resource manuals
is revealed, which can be easily remedied. Creating a comprehensive database with
separate rows of multiple criteria allows for flexibility in the analysis of resources
ranging from a simple resource search engine to a complex spatial analysis tool to
numerically describe the layout of resources and services in Jefferson County.

GIS.Analysis. of. Community. Resources

To do the GIS analysis, the resources identified by the mapping and focus group
participants were verified using the 2004 Denver Metro phone book (www.dexonline.
com) and/or MapQuest (www.mapquest.com). The addresses were entered into an
Excel spreadsheet, the type of resource and service was coded into one or more of
the 24 types of resources, and the participant group and demographics were entered.3
The spreadsheet was imported into ArcMap and geo-coded to place the identified
resources with known addresses on a common geographical coordinate system.
The 24 layers of resource and service types were created by: (1) working with
participants in four pilot sessions to understand how they conceptualized services;
(2) presenting the pilot workshop information to Jefferson County Human Service
employees to add categories that aligned with their current resource guides and
experiences; and (3) aligning with the Aires taxonomy used by Colorado 2-1-1
system.4 The resulting twenty-four layers allowed for some resources/services to
be coded into more than one layer, thereby increasing the ability of information
seekers to find the desired resources. For example, a service dedicated to providing
transportation to elderly individuals would be included in Aging and Adult/Senior
Services as well as Transportation Services. The dual coding in this project and used
by the 2-1-1 system produces a comprehensive searchable database that can reach
a specific resource through multiple avenues.
At the onset of the project, the research team decided that a method for determining
when the mapping process had reached a saturation level of redundant information
was needed. A novel method, a “repeat rate,” was created. The repeat rate was set
at 80% based on the time and cost involved in identifying additional resources. The
repeat rate estimates the projected number of new points to be expected from each

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 335

additional map. So, for example, on the first map every identified resource is unique
and a new point in the database, the repeat rate is zero. By the 100th map, if 50% of
the points collected were previously identified on any of the previous 99 maps, the
non-repeated points for that map would be 50%. At some point, theoretically, no
new information will be gleaned from additional maps, resulting in a 100% repeat
rate. In practical terms, full saturation cannot be reached; but more importantly, a
trade-off must be made between additional information and cost to acquire the in-
formation. In this project, there was an average of 20 resource points per map. Each
map took approximately four hours to process (from address coding the points to
data entry into ArcMap). An 80% threshold repeat rate was chosen as a break-even
level. At this rate, only one additional new resource would be identified per one
hour of coding time. Less than one additional new resource per hour of work was
deemed an inefficient use of resources.

Selected. Results

While the cognitive mapping process identified formal and informal resources,
without the ability to do aggregated and stratified analyses the collected data would
only have provided a list of resources known by the mappers. To create a comprehen-
sive SOC child welfare system requires detailed knowledge about the geographical
distribution of the density of community resources and the types available in the
community. Additionally, to understand and correct the community-level information
gaps among providers, community members and clients, a geographical analysis
stratified by resource categories and information sources (e.g., county resource
guides, social service workers, community providers, faith-based providers, and
clients) was needed. GIS is able to provide such geographical analysis. Importantly,
data presented in visual form facilitated information dissemination among a wide
variety of stakeholders.
Seven Jefferson County cities were the focus of the mapping project. Clients and
community resident groups were drawn from Lakewood, Arvada, Wheat Ridge,
Golden, Littleton, and the mountain communities of Conifer and Evergreen. A
total of 42 workshops were held at locations in each of these communities. Special
sessions were held with members of the dominant ethnic groups, identified using
census data imported into ArcMap. The groups included Native American, Russian,
and Latino community residents,5 with Latinos being the largest ethnic population
in Jefferson County. To determine the areas in Jefferson County with the highest
growth rate of Latinos, 1990 and 2000 census data was imported into ArcMap
and the percent change was calculated. The highest concentrations of the Latino
population were found in the central eastern area of Jefferson County (Lakewood
and Arvada) on the border of Denver County. Figure 2 shows the Latino growth

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

Figure 2. Latino population growth in Jefferson County, 1990-2000

patterns from 1990 to 2000. Numbers on the map are the percentage of the area’s
Latino population in the 2000 Census.
Current and past users of Jefferson County social services were identified by Jeffer-
son County. Since the database was confidential, Jefferson County would not allow
CIPP to pull a stratified sample. In addition, over half the client records were miss-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 337

Figure 3. Client home locations and range boundaries

ing complete address information. Because of these difficulties, all complete client
files with zip codes in the study area were mailed invitations by Jefferson County
to participate in the study. Fortunately, the clients, foster, and adoptive parents who
participated were distributed throughout the study area. The following map shows
the home location cross-streets of the participants and the extrapolated hypothetical
boundaries of their identified communities, ranging from one-third mile to one mile

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

Figure 4. All Jefferson County, Colorado, resources identified by participants and

resource guides

around their residence. The buffer area was based on actual community boundaries
drawn on the maps, which ranged from larger than one mile to a handful that were
smaller than one-third mile. Figure 3 indicates that the mapping participants who
were current or previous users of Jefferson County social services had collective
knowledge of nearly all of the study area (see Appendix A for demographic infor-
mation of the participants).
Figure 4 shows the 3,845 resources identified and mapped. The number of unique
points is 1,819 after removing the repeated information. Identified resources are
distributed across all the Jefferson County study areas with a scattering throughout
Denver County and a dense clustering along the main arterial east-west street,
Colfax.
Next are selected categories of resources to illustrate how GIS was used to inform
SOC planning efforts.

• Childcare: One hundred and eleven different childcare providers were iden-
tified. As Figure 5 shows, there is very little overlap in knowledge between

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 339

Figure 5. Childcare resources identified by participants vs. Jefferson County re-

source guides

the Jefferson County resource guides and the participants in the mapping
workshops. This could indicate: (1) clients and providers are not familiar
with the childcare resources which Jefferson County has identified; and (2)
Jefferson County is not effectively disseminating this information. Many of
the childcare resources identified in the Jefferson County resource guides
are before-school and after-school care programs. In contrast, the cognitive
mapping participants identified pre-school childcare facilities rather than
before and after-school programs. Based on this analysis, the current guides
are lacking in pre-school childcare resources. It is likely this information is
housed with a specific organization rather than listed within Jeffco’s resource
guides; however, a consolidated database will facilitate resource integration,
information referral, and client access.
• Education:.There were 366 education resources identified. This is the cat-
egory in the Jefferson County guides with the most resources. As would be
expected, the Jefferson County Public Schools Resource Guide has all the
primary and secondary public schools, as well as a few private schools in the
county. Figure 6 shows there is very little additional information provided by

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kathlene

Figure 6. Education resources: Cognitive mapping vs. resource guides

the mapping sessions with regard to education resources in Jefferson County;

however, the mapping identified a significant number of resources in Denver
that are known and used by Jefferson County residents. The Jefferson County
education resources not identified in the resource guides were typically services
such as breast-feeding classes located at hospitals, ESL classes available at
community centers and/or libraries, and other educational services available
at libraries, these important educational services are missing in Jefferson
County’s guides.
• Health.care.resources:.Of the 190 health care resources, 140 were identified
by the cognitive mapping process with only one-third of the services identi-
fied by both participants and the Jefferson County resource guides. As seen
in the Figure 7, however, the Jefferson County resource guides also have
information about several health care services in Denver. Most of the Denver
and Jefferson County services were also identified by the cognitive mapping
process. Information from the cognitive mapping process increases the resource
base by two-thirds, adding significantly to health care knowledge of Jefferson
County.

Three of the 24 resource categories have been provided above for sake of brev-
ity. However, there were several trends that appeared in almost all of the resource

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 341

Figure 7. Health care resources: Cognitive mapping vs. resource guides

categories. First, the Jefferson County resource guides failed to provide adequate
information. Except in education and childcare, the resource guides fell far short
of the number of resources identified by the mapping participants. While there are
many childcare resources in the Jefferson County information guides, there is little
overlap between the childcare listed in the Jefferson County guides and the childcare
identified by the cognitive mapping. The GIS maps effectively demonstrate such
knowledge gaps.
Second, there are a significant number of resources in Denver County (east of Jef-
ferson County) that providers and clients identify. Reasonable accessibility to Denver
County, as well as lack of availability of the resources in Jefferson County, likely
accounts for this trend. Building a community-based SOC will require Jefferson
County to find ways to offer some of these services locally, a challenge that will
require developing community partnerships to overcome the financial constraints
which the County faces.
Third, opposite of the previous trend, Jefferson County resource guides provide
mainly Denver locations for some types of resources, even though the same re-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

sources exist in numerous places in Jefferson County. Available resources closer to

Jefferson County residents are a fundamental component of SOC and, in this trend,
require only disseminating the information effectively, which is a low-cost method
to improve community-based service delivery.
Finally, there is a large disparity in knowledge between clients and providers. With
the exception of 3 of the 24 categories, Education, Recreation, and Commercial
Resources, the providers and clients did not overlap significantly in knowledge about
resources. Providers know more about traditional resources such as other agencies
or governmentally-supported social services, while clients know about resources
of a less traditional nature, such as churches, motels, and parks where teenagers
gathered to socialize and engage in recreational sports activities. Although these
informal resources are not referral services that providers typically pass along to
clients, they are important community-based resources to share with clients. In
creating a community-based SOC, providers need to be aware of the alternative
methods clients use to meet their needs. In some instances, this new information
will lead to the creation of government/community partnerships to more effectively
and efficiently deliver services. In other circumstances, the additional knowledge of
resources will provide clients with options and/or fill gaps in needs that traditional
government and community providers cannot meet.

Lessons. Learned

Several problems directly and indirectly related to the GIS component of the project
became apparent and required adjustments to the procedures or accommodations
to the expected output. These include research procedures that are incompatible
with social service agencies’ capacity, issues of client confidentiality, repeat rates,
incomplete and/or inaccurate databases for coding resource locations, coding pro-
tocols, and mapping accuracy.
First, as has been found before, many county and local agencies lack leadership that
understands the value of GIS in policy decision-making (Greene, 2000; Nedovic-
Budic, 1996; Ventura, 1995; Worrall & Bond, 1997). Hence, many agencies lack
the technical ability to employ GIS and, consequently, also lack the understanding
to work effectively and efficiently with the researchers. Furthermore, because social
service agencies typically do not have a GIS analyst on staff, data and map files
have limited usefulness beyond the initial analysis as presented in the final report.
Finally, human service agencies have organizational procedures that create signifi-
cant barriers in implementing research projects, barriers that need to be addressed
in the project planning stages (Ventura, 1995). Jefferson County Human Services
suffered from all three impediments and was exacerbated by the high turnover of

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 343

the staff. In the first year, two-thirds of the project staff left. By the middle of the
second year, only one person out of nine key project staff remained. Those who
left included the project manager and the principal investigator, both of who had
been replaced twice. Within 18 months, none of the people who conceptualized
and wrote the HHS grant were involved in the project. Institutional memory was
wiped clean and new staff was unfamiliar and wary of many components laid out
in the grant proposal, including the untraditional resource identification method.
Higher administrative support for the innovative project waned, and “business as
usual” reasserted itself as the dominant paradigm. It became clear that the resource
database developed through the mapping process would not be updated on a regular
basis and, perhaps, not disseminated throughout the organization if left to Jefferson
County. The CIPP sought out a more stable organization to house the resource data,
Colorado 2-1-1, with the permission of the first project manager.
Second, human service agencies as well as educational institutions cannot share
client/student data. This presents a significant research barrier when the project
requires participation of these populations. Ideally, individuals within the orga-
nizations would have both the access to the data and sophistication to manipulate
the data in accordance with standard research protocols. This is unlikely to be the
case in institutions which are financially strapped and lack the vision or political
will to invest in trained personnel and needed research tools. To ameliorate these
conditions, project planning must include agreed-upon protocols for effectively and
efficiently handling confidential data.
Third, unique to this project was the creation of a “repeat rate” to set a standard for
data density. The 80% repeat rate was selected for efficiency of resources, based on
an extrapolation of the average number of points per map and time needed to code
and enter the data for each map. Unknown was how many participants/maps were
needed to reach the 80% repeat rate in each of the 24 categories. Initially, the CIPP
recommended target was 450 participants. This number was revised downward by
Jefferson County Human Services to a maximum of 250 participants. From the 247
actual participants, the 80% repeat rate was reached in only two of the 24 resource
categories. The average repeat rate was 55% across all categories, indicating that
more than 250 participants were needed to reach 80%. Whether 450 participants
were ultimately required is unknown. More importantly, did the lower repeat rate
significantly affect the quality of the project? Certainly, fewer resources were identi-
fied at the 55% rate; but 1,480 resources not in Jefferson County resource guides
were identified; not an insignificant contribution to building a more comprehensive
social services.
Fourth, in the process of coding the maps and sorting the data to find repeated ad-
dresses or groupings by type of provider, and so forth, it was discovered that precise
alphanumeric coding was critical. With the large number of data fields (attributes)
assigned to each participant, there were inconsistencies in some of the categories.
The data cleaning was more extensive than anticipated. Future projects should utilize

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Kathlene

numeric coding in attributes to the fullest extent possible and develop strict alpha-
numeric standards for addresses, organizational names, and other alpha fields.
Finally, to find resource addresses, MapQuest and the Denver metro area phone
book were used. MapQuest was the most efficient method but had the most errors,
as discovered when the address was imported into ArcMap. A cross-check with the
phone books corrected most of these errors.
Nine percent of the mapping points were unidentifiable due to a combination of
missing information in MapQuest and the phone book, and poor location informa-
tion on the hand drawn maps. The latter accounted for a greater proportion of the
unidentified points, especially resources such as neighborhood parks and unnamed
resources such as “soup kitchen.” Rather than rely solely on participants naming
the nearest cross streets to such resources, the closest known commercial entity
should be identified. This redundancy will reduce missing data due to participant
error in naming streets.

Future. Trends

While this project was limited to identifying resources, spatial patterns of resource
locations, and knowledge gaps, the collected data can be mined further. More specific
uses can be created, such as a searchable Web-based provider resource database
and the identification of physical and/or service areas with inadequate resources
in relation to socio-economic deprivation areas. The latter allows providers to
demonstrate specific needs, important for several reasons, including the pursuit of
future programmatic funding. These specific uses are described in greater detail as
follows:

•. Provider.resource.database: In the future, the Web-based database can be

converted into a tool for social service providers to identify available resources
and the most accessible locations for clients (Worrall & Bond, 1997). The end
user (a case-worker) would be able to search for particular resources based on
any number of criteria or a combination of criteria. For example, one might enter
necessary criteria such as Rental Assistance Housing Resource located within
three miles of a given location that also caters to Spanish-speaking clientele.
After these attributes or criteria are entered into the appropriate locations on
the Webpage, a list of all the resources or providers that fit the criteria could
be retrieved, similar to the business name search feature available through a
site such as MapQuest. Finally, digital maps could be generated with driving
directions for the case-worker to print out for the client. It is also possible to
map the public transportation routes to services.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 345

•. Needs. assessments: The database can be used to conduct comprehensive,

quantifiable, and defensible needs assessments. A social service provider ad-
ministrator or grant writer could search the data described above in conjunction
with Census data and the County’s client locations to reveal areas of need or
areas of excess (Bond & Devine, 1991; Worrall & Bond, 1997).6 A strategic
plan could be developed to determine where a new office or access point for
a particular resource should be located to serve the greatest number of clients.
This type of spatial analysis based on quantifiable numbers and distances can
be used to justify a particular course of action either for internal/external ac-
countability or to acquire funding for various projects aimed at community
resource and social service distribution.

Acknowledgments

The author would like to thank April Smith, Department of Psychology, Colorado
State University, and Mary Tye, Department of Psychology, Colorado State University,
for running the workshops and coding the data; David Wallick, Colorado Institute
of Public Policy, Colorado State University, for conducting the GIS analysis; and
Juliana Hissrich for providing administrative support to the project.

Conclusion

Cognitive mapping combined with GIS analysis is a powerful method for identify-
ing community resources by providing: (1) a comprehensive database of existing
services; (2) a basis to build communication networks and cooperation among gov-
ernment and community providers; (3) the ability to create an efficient system that
avoids duplication of efforts; (4) an understanding of the geographical distribution
of resources; (5) the identification of resources lacking in the county and specific
communities; and (6) knowledge differences among diverse participant groups. The
addition of 1,480 resource locations within the seven study areas (only a portion of
Jefferson County) nearly tripled the number of resources and services listed in the
Jefferson County guides.
Ultimately, service delivery in SOC is about building partnerships across the multiple
services and bringing in new, even sometimes untraditional, community partners.
Family involvement is the key in this collaborative arrangement. Similar to untra-
ditional community partners and resources, families as partners do not fit easily
within current social service delivery structures, values, and beliefs. Recognizing,

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
346 Kathlene

valuing, and partnering with resource providers identified by clients and community
members is one important step toward shifting practices. Cognitive mapping with
GIS provides a tool for taking the first critical steps.

References

Bond, D., & Devine, P. (1991). The role of geographic information systems in survey
analysis. The Statistician, 40, 209-215.
Daniels, K., & Johnson, G. (2002). On trees and triviality traps: Locating the debate
on the contribution of cognitive mapping to organizational research. Organiza-
tion Studies, 23(1), 73-81.
Evans, G. W. (1980). Environmental cognition. Psychological Bulletin, 88(2),
259-287.
Fridgen, J. D. (1987). Use of cognitive maps to determine perceived tourism regions.
Leisure Sciences, 9(2), 101-117.
Fulton, W., Horan, T., & Serrano, K. (1997). Putting it all together: Using the ISTEA
framework to synthesize transportation and broader community goals. Clare-
mont Graduate University, University Research Institute, Claremont, CA.
Greene, R. W. (2000). GIS in public policy: Using geographical information for
more effective government. Redlands, CA: ESRI Press.
Hardwick, D. A., Wooldridge, S. C., & Rinalducci, E. J. (1983). Selection of land-
marks as a correlate of cognitive map organization. Psychological Reports,
53(3), 807-813.
Heagerty, P. J., & Lele, S. R. (1998). A composite likelihood approach to binary spatial
data. Journal of the American Statistical Association, 93(443), 1099-1111.
Hjortso, C. N., Christensen, S. M., & Tarp, P. (2005). Rapid stakeholder and conflict
assessment for natural resource management using cognitive mapping: The
case of Damdoi Forest Enterprise, Vietnam. Agriculture and Human Values,
22, 149-167.
Hobbs, B. F., Ludsin, S.A., Knight, R. L., Ryan, P.A., Biberhofer, J., & Ciborowski, J. J. H.
(2002). Fuzzy cognitive mapping as a tool to define management objectives
for complex ecosystems. Ecological Applications, 12, 1548-1565.
Holahan, C. J., & Dobrowolny, M. B. (1978). Cognitive and behavioral correlates of
the spatial environment: An interactional analysis. Environment and Behavior,
10(3), 317-333.
Jordan, T., Raubal, M., Gartrell, B., & Egenhofer, M. J. (1998, July). An affor-
dance-based model of place in GIS. In Eighth International Symposium on

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 347

Spatial Data Handling ’98 Conference Proceedings, Vancouver, BC, Canada

(pp. 98-109).
Kathlene, L. (1997). 29th street greenway corridor citizen survey panel: Results
of mapping exercise, phase 3. Minneapolis, MN: University of Minneapolis,
Humphrey Institute of Public Affairs.
Kathlene, L., & Horan, T. (1998). GIS survey of 29th street corridor, Minneapolis,
MN. Minneapolis, MN: University of Minneapolis, Humphrey Institute of
Public Affairs.
Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D. W. (2001). Geographic
information systems and science. New York: John Wiley and Sons, LTD.
Lynch, K. (1960). The image of the city. Cambridge, MA: MIT Press.
Magana, J. R., & Norman, D. K. (1980). Methodological inquiry into elicitation
procedures: Cognitive mapping and free listing. Perceptual and Motor Skills,
51(3), 931-934.
Horan, T., Serrano, K., & McMurran, G. (2001). GIS for livable communities:
Examiniation of community perceptions of assets, liabilities and transporta-
tion improvements. San Jose, CA: San Jose University, Mineta Transportation
Institute, College of Business.
Moeser, S. D. (1988). Cognitive mapping in a complex building. Environment and
Behavior, 20(1), 21-49.
Nasar, J. L. (1988). The evaluative image of the city. Thousand Oaks, CA: Sage
Publications.
Nedovic-Budic, Z., & Godschalk, D. R. (1996). Human factors in adoption of
geographical information systems: A local government case study. Public
Administration Review, 56, 554-567.
O’Laughlin, E. M., & Brubaker, B. S. (1998). Use of landmarks in cognitive map-
ping: Gender differences in self report versus performance. Personality and
Individual Differences, 24(5), 595-601.
O’Neill, M. J. (1991). Evaluation of a conceptual model of architectural legibility.
Environment and Behavior, 23(3), 259-284.
Quaiser-Pohl, C., Lehmann, W., & Eid, M. (2004). The relationship between spatial
abilities and representations of large-scale space in children — a structural
equation modeling analysis. Personality and Individual Differences, 36(1),
95-107.
Reich, R. M., & Davis, R. (2003). Spatial statistical analysis of natural resources
(Tech. Rep. No. NR512). Fort Collins, CO: Colorado State University.
Sholl, M. J. (1987). Cognitive maps as orienting schemata. Journal of Experimental
Psychology: Learning, Memory, & Cognition, 13(4), 615-628.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
348 Kathlene

Stroul, B. (1996). Profiles of local systems of care. In B. A. Stroul and R. M. Friedman

(Eds.), Systems of care for children’s mental health (pp. 149-176). Baltimore:
Paul H. Brookes Publishing Co.
Tolman, E. C. (1948). Cognitive maps in rats and men. Psychological Review,
55(4), 189-208.
Unger, D. G., & Wandersman, A. (1985). The importance of neighbors: The social,
cognitive, and affective components of neighboring. American Journal of
Community Psychology, 13(2), 139-169.
Uzzell, D., Pol, E., & Badenas, D. (2002). Place identification, social cohesion, and
environmental sustainability. Environment and Behavior, 34(1), 26-53.
Ventura, S. J. (1995). The use of geographical information systems in local govern-
ment. Public Administration Review, 55, 461-467.
Worrall, L., & Bond, D. (1997). Geographical information systems, spatial analysis,
and public policy: The British experience. International Statistical Review,
65, 365-379.
Yoshino, R. (1991). A note on cognitive maps: An optimal spatial knowledge rep-
resentation. Journal of Mathematical Psychology, 35, 371-393.

Endnotes
1
The project was supported by grant #90CA1715/01, CFDA #93.570 from the
Federal Department of Health and Human Services through Jefferson County,
Colorado.
2
The term cognitive mapping is used for a variety of techniques, including
“fuzzy cognitive mapping,” a technique that builds mental maps of perceptions
from focus-group and interviews (Hjortso, Christensen, & Tarp, 2005; Hobbs
et al., 2002). In this project, cognitive mapping means hand-drawn maps of
tangible community resources and locations, a geographical data collection
technique new to GIS.
3
Nine percent of the mapping points could not be accurately located and were
dropped from the analysis. Of the remaining 89%, two possible location er-
rors could occur in transferring the cognitive map information into a database
for ArcMap. First, multiple coders could use different alphanumeric codes,
thereby making the same resource appear as a different resource. To correct
this error, the data was cleaned by conducting sorts on multiple columns in
the excel spreadsheet to reveal unknown duplicates. For example, a search

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Cognitive Mapping and GIS for Community-Based Resource Identification 349

on “Research Name” might find the same resource with inconsistent address
codes. If the address did not match exactly (e.g., one was coded with “St.” and
another coded with “Street,” the coding was corrected to be consistent. Similar
searches were done on other categories such as street address, street name,
and zip code. The data was cleaned accordingly. The second error was from
incorrect addresses in the MapQuest and/or Dex directory. The Dex directory
is the official metropolitan phone and address directory and should have a high
level of reliability; however, the actual reliability rate is unknown. To correct
for possible errors, all identified social services not in the Jefferson County
resource guides (e.g., soup kitchens, English as a Second Language courses,
support groups, etc.) were called to verify the address. It was assumed that
the Jefferson County resource guides had accurate information.
4
All identified resources were provided to Colorado’s 2-1-1 system, which is the
national abbreviated dialing code for free access to health and human services
information and referral (I&R). 2-1-1 is an easy-to-remember and universally-
recognizable number that makes a critical connection between individuals
and families in need and the appropriate community-based organizations and
government agencies. Housing the data with 2-1-1 allows statewide access to
resources and bi-annual updating to keep the information current. Colorado
2-1-1 system is the depository for the resources collected in this project. Web
searchable database of resources can be found at http://211colorado.org/
5
CIPP provided Jefferson County with the ethnic enclave areas based on the
2000 Census. The Asian communities fell outside the project boundaries set
by Jefferson County (see Figure 1) and, unlike Russians, Latinos, and Native
Americans, Jefferson County did not request mapping with the Asian com-
munity.
6
For example, it might be found that 65% of all users of a certain type of resource
(this data would be collected by cognitive mapping alone) live “x” number of
miles away (analysis performed by the GIS system) from a particular needed
or frequently-accessed resource (gathered through cognitive mapping and
other sources).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Kathlene

Appendix

Only forty percent of the participants provided demographic information, which

limits the ability to determine the gender, age, and ethnicity/race of the participants.
However, there is no way to determine the representativeness of the sample on these
traditional demographics since the population characteristics are unknown. Even
among the clients, the demographics are not available because most of the client
records were incomplete. Unlike many social research projects, demographic rep-
resentation is less of a concern. For the identification of resources, a cross-section
of the types of people who use or provide services and the geographical distribution
of their knowledge was most important, of which both criteria were met.

Table 2. Demographics of participants (n=100)

Demographic All participants Providers Clients Community

characteristic (n=100) (n=19) (n=72) Residents (n=9)

Number and percent

85% 90% 82% 100%
female

Average age 34.39 39.75 31.86 44.83

Number and percent

62% 68% 64% 33%
Caucasian
Number and percent
19% 5% 24% 11%
Latino
Number and percent
6% 0% 4% 33%
African American
Number and percent
9% 21% 4% 22%
Native American
Number and percent
4% 5% 3% 0%
Other

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

Chapter.XVI

Spatial.Reasoning.for.
Human-Robot.Teams
Davd J. Bruemmer, Idaho Natonal Laboratory, USA
Douglas A. Few, Idaho Natonal Laboratory, USA
Curts W. Nelsen, Idaho Natonal Laboratory, USA

Abstract

This chapter presents research designed to study and improve an operator’s abil-
ity to navigate or teleoperate a robot that is distant from the operator through the
use of a robot intelligence architecture and a virtual 3D interface. To validate the
use of the robot intelligence architecture and the 3D interface, four user-studies
are presented that compare intelligence modes and interface designs in navigation
and exploration tasks. Results from the user studies suggest that performance is
improved when the robot assumes some of the navigational responsibilities or the
interface presents spatial information as it relates to the pose of the robot in the
remote environment. The authors hope that understanding the roles of intelligence
and interface design when operating a remote robot will lead to improved human-
robot teams that are useful in a variety of tasks.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

Introduction

Robots have been used in a variety of settings where human access is difficult, im-
practical, or dangerous. These settings include search and rescue, space exploration,
toxic site cleanup, reconnaissance, patrols, and many others (Murphy, 2004). Often,
when a robot is used in one of these conditions, the robot is distant from the opera-
tor; this is referred to as teleoperation. Ideally, robots could be a useful member of
a team because they could be used to accomplish tasks that might be too difficult
or impractical for a human to perform.
The potential, however, for humans and robots to work as an effective team is limited
by the lack of an appropriate means for the operator to visualize the remote environ-
ment and how the robot fits within the environment. As an example, several recent
research efforts have investigated the human-robot interaction challenges associated
with real-world operations including search and rescue and remote characterization
of high-radiation environments (Burke, Murphy, Coovert, & Riddle, 2004; Casper
& Murphy, 2003; Murphy, 2004; Yanco, Drury, & Scholtz, 2004a). Across these
disparate domains, researchers have noted that it is difficult for operators to navigate
a remote robot due to difficulty and error in operator understanding of the robot’s
position and/or perspective within the remote environment.
A primary reason for the difficulty in remote robot teleoperation is that for the
overwhelming majority of robotic operations, video remains the primary means of
providing information from the remote environment to the operator (Burke, Mur-
phy, Rogers, Lumelsky, & Scholtz, 2004a). Woods, Tittle, Feil, and Roesler (2004)
describe the process of using video to navigate a robot as attempting to drive while
looking through a “soda straw” because of the limited angular view associated with
the camera (Woods et al., 2004). The limited angular view of the camera presents
problems for robot teleoperation because obstacles outside of the field of view of
the camera still pose navigational threats to the robot even though they are not vis-
ible to the operator.
To alleviate navigational threats to the robot, current research at the Idaho National
Laboratory (INL) is aimed at providing tools that support mixed-initiative control
where humans and robots are able to make decisions and take initiative to accomplish
a task. The goal is to create a set of capabilities that permit robots to be viewed as
trusted teammates rather than passive tools. If this is to happen, the robot as well
as the human must be enabled to reason spatially about the task and environment.
Furthermore, true teamwork requires a shared understanding of the environment and
task between team members in order to understand each others’ intentions (Dennett,
1981). The lack of an effective shared understanding has been a significant impedi-
ment to having humans and intelligent robots work together.
In response to this challenge, the INL has developed a mixed-initiative robot control
architecture that provides a framework for robot intelligence, environment mod-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

eling, and information sharing. In order to support a shared understanding of the

environment and task between robotic and human team members, a virtual three
dimensional 3D interface was developed through collaboration with researchers at
Brigham Young University (BYU). The combination of the virtual 3D interface and
intelligence on the robot can be used to improve the human’s and robot’s ability
to reason spatially about the environment by presenting a shared understanding of
the environment. The technology used to achieve the shared understanding of the
environment is presented next.

System. Design

Through the Office of the Secretary of Defense (OSD) Joint Robotics Program
(JRP), the Space and Naval Warfare Systems Center (SPAWAR) at San Diego and
the INL have worked together to develop, mature, and integrate promising robotics
technologies from throughout the robotics community including components for
perception, communication, behavior, and world modeling. One of the results of
this collaboration is the development of the INL Robot Intelligence Kernel, which is
currently used to unite selected components into a behavior-based architecture that
can be transferred to a variety of fieldable, unmanned ground vehicle systems.
The robot intelligence architecture is the product of an iterative development cycle
where behaviors have been evaluated in the hands of users, modified, and tested
again. In fact, many of the strategies and interface components that originally seemed
elegant from a conceptual standpoint, proved to be frustrating for users. For example,
during a preliminary experiment that evaluated robot intelligence, but provided
minimal spatial reasoning tools to the operator, it was noted that although most
participants felt a high level of control, some participants indicated that they were
confused by the robot behaviors (Marble, Bruemmer, & Few, 2003). In particular,
the automatic initiation of robot behaviors to get the robot out of a narrow hallway
led to operator confusion and a fight for control between the robot and human,
because operators thought the robot could go through the hallway but the robot
sensors indicated that it would not fit. The lack of adequate spatial representation
tools prevented the operator from realizing this fact and, consequently, the human
and robot engaged in a fight for control of the robot’s movements. Findings such as
these serve to motivate improvements to the robot intelligence architecture and the
development of interface components that could support spatial reasoning.
Currently the robot intelligence kernel is divided into four modes of control (Tele,
Safe, Shared, and Autonomous) affording the robot different types of behavior and
levels of autonomy (Marble, Bruemmer, & Few, 2003; Marble, Bruemmer, Few, &
Dudenhoeffer, 2004). The modes of autonomy in the robot intelligence architecture
include:
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

1.. Tele.mode is a fully-manual mode of operation, in which the operator must

manually control all robot movement.
2.. Safe.mode is similar to tele mode, in that robot movement is dependent on
manual control. However, in safe mode, the robot is equipped with a level of
initiative that prevents the operator from colliding with obstacles.
3. In shared.mode, the robot can relieve the operator from the burden of direct
control, using reactive navigation to find a path based on perception of the
environment. Shared mode provides for a dynamic allocation of roles and
responsibilities. The robot accepts varying levels of operator intervention and
supports dialogue through the use of a finite number of scripted suggestions
(e.g., “Path blocked! Continue left or right?”) and other text messages that
appear in a text box within the graphical interface.
4.. Autonomous. mode consists of a series of high-level tasks such as patrol,
search a region, follow a path, or go to a place. In autonomous mode, the
only user intervention occurs on the tasking level; the robot itself manages
all navigational decision-making.

To investigate the challenges of sharing control of the robot between the robot and
the operator, the experiments reported in this research focus on the middle ground
that falls between teleoperation and full robotic autonomy (i.e., safe mode and shared
mode). Although the experiments restricted each participant to only one level of
control, normal operation would permit the user to switch between all four modes
of autonomy as the task constraints, human needs, and robot capabilities change.
As an example, tele mode could be useful to push open a door or shift a chair out
of the way, whereas autonomous mode could be used to reduce human workload or
in an area where communications to and from the robot are sporadic.
In order to protect the robot from collisions with obstacles in robot control modes
that have some robot autonomy (safe, shared, autonomous), a guarded motion
behavior based on a technique described by Pacis, Everett, Farrington, and Bruem-
mer (2004) is implemented. In response to laser and sonar range sensing of nearby
obstacles, the guarded motion behavior scales down the robot’s velocity using an
event horizon calculation, which measures the maximum speed at which the robot
can safely travel in order to come to a stop approximately two inches from an ob-
stacle. By scaling down the speed in small increments, it is possible to insure that
regardless of the commanded translational or rotational velocity, guarded motion
will stop the robot at a consistent distance from an obstacle. This approach provides
predictability and ensures minimal interference with the operator’s control of the
vehicle. If the robot is being driven near an obstacle rather than directly towards
it, guarded motion will not stop the robot, but may slow its speed according to the
event horizon calculation.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

In order for the robot to be navigated successfully, spatial information of the envi-
ronment must be available. The robot intelligence kernel gathers spatial information
from the environment with a laser range finder. Information from laser scans is
combined into a map of the environment using a technique developed at the Stan-
ford Research Institute (SRI) called consistent pose estimation (CPE) (Gutman &
Konolige. 1999; Konolige, 2004). This map-building algorithm is designed to build
an occupancy-grid based map of the robot’s environment as the robot explores the
environment (Elfes, 1987; Moravec, 1988). The mapping algorithm is particularly
useful because it provides an accurate spatial representation of complex environ-
ments that are previously unknown to the robot or the operator.
Since no single robot platform is appropriate for all tasks, the INL robot intelligence
architecture can port to a variety of robot geometries and sensor suites and is currently
in use as a standard by several research teams throughout the human-robot interac-
tion (HRI) community. Experiments presented later in this paper were performed
with an iRobot “ATRV mini” or an iRobot “ATRV Jr” shown in Figure 1. On each
robot, the intelligence architecture utilizes a variety of sensor information including
inertial sensors, compass, wheel encoders, laser, computer vision, thermal camera,
infrared break beams, tilt sensors, bump sensors, sonar, and ultrasonic sensors.
The default configuration of the interface used to interact with the robot consists of a
single touch screen display containing five re-sizeable windows as shown in Figure
2 (Bruemmer et al., 2005). The upper left-hand window on the screen contains a
video feed from the robot as well as controls for panning, tilting, and zooming the
camera. Frame size, frame rate, and compression settings can be accessed from a
sub-window, but are held constant throughout the experiments reported here.
The upper right-hand window contains sensor status indicators and controls that
allow the operator to monitor and configure the robot’s sensor suite as needed. The

Figure 1. The robots used for Experiments 1 – 4

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

Figure 2. The standard interface

lower right-hand window pertains to movement within the local environment and
provides indications of robot velocity, obstructions, resistance to motion, and feed-
back from contact sensors. The interface indicates blockages that impede motion in
a given direction as red ovals next to the iconographic representation of the robot
wheels (lower right of Figure 2). The current snapshot of the interface indicates that
movement right and left is not possible because of an object close to the wheels on
the left side of the robot. These indicators are designed to inform the operator as to
why the robot has overridden a movement command. Since the visual indications
can sometimes be overlooked, a force feedback joystick is also implemented to resist
movement in the blocked direction. The joystick vibrates if the user continues to
command movement in a direction already indicated as blocked. At the far right of
the window the user can select between different levels of robot autonomy.
The lower central window displays the map of the environment as it is discovered
by the robot and allows the user to initiate a number of waypoint-based autonomous
behaviors such as search region, patrol region, create a path, or go to a place. Addi-
tionally, the map can be moved and zoomed in and out to provide a desired perspec-
tive. The lower left-hand window contains information about the robot’s operational
status such as communication activity, power, and the robot’s pitch and roll.
A virtual three-dimensional (3D) display (Figure 3) was designed to support the
operator’s awareness of the spatial information in the robot’s environment and show
the information related to the robot’s current pose within the environment. The
virtual 3D component has been developed by melding technologies from the INL
(Bruemmer et al., 2005), Brigham Young University (BYU) (Nielsen et al., 2004),
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

Figure 3. The virtual 3D interface

and the Stanford Research Institute (SRI) (Gutman & Konolige, 1999; Konolige,
2004). The 3D display is not based on true 3D range sensing, but rather by extrud-
ing the 2D map built by the robot into a 3D perspective. The map information in
the 3D interface and the standard interface both originate from the map-building
algorithm on the robot. The only difference is the manner in which the information
is presented to the operator.
The map information produces the basis for the 3D representation that includes
obstacles and other semantic entities that are of significance to the operator such as
start location, labels, and waypoints. These items can be inserted by the robot to
indicate percepts and intentions or by the human to identify and classify targets in
the environment. Also, the user is able to add, verify, remove, or annotate semantic
entities displayed within the map. Collaborative construction of the map enhances
each individual team member’s understanding of the environment and provides a
basis for the human-robot team to “communicate” naturally about the environment
through the visualization of relevant spatial information.
In the 3D interface, the operator may also insert translucent still images excerpted
from the robot video, which are overlaid onto the corresponding area of the 3D map
display, providing a means to fuse real video information with the virtual represen-
tation of the environment (Nielsen et al., 2004). By changing the virtual display’s
zoom, pitch, and yaw, it is possible to move the virtual perspective of the robot and
environment from an egocentric perspective (i.e., looking out from the robot), to a
fully exocentric view where the entire environment (map) can be seen at once.
The experiments presented next utilize the robot intelligence kernel as described in
this section to compare the safe and shared modes of robot control along with the use
of the standard interface and the 3D interface in navigation and exploration tasks.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

Experiment.1

The first experiment was intended to: (a) show that the behaviors on board the robot
(e.g. guarded motion and autonomous navigation) were useful in an exploration
task, and (b) to compare the safe and shared modes of autonomy in an exploration
task. It was hypothesized that participants would perform better with the shared
autonomy mode than with the safe autonomy mode.
For this experiment a 20’ × 30’ maze environment was created using conventional
office dividers and cylindrical pylons. Participants controlled the robot from a re-
mote station where the robot environment was not visible. Five objects of interest
(two mannequins, a stuffed dog, a disabled robot, and a small simulated explosive
device) were placed throughout the arena in locations that remained fixed for all
participants. The placement of these items further complicated the navigation task
since operators were told not to drive into or over the objects. Moreover, certain
objects remained hidden except from certain vantage points so the operator was
required to maneuver the robot in order to see all the objects of interest.
Each participant was given 60 seconds to locate as many of the five items in the
search area as possible using the standard interface and either the safe or the shared
autonomy mode. Prior to the experiment, participants were instructed on the use
of the joystick for controlling the robot and the camera on the robot (e.g., pan, tilt,
and zoom), but were given no opportunity to practice controlling the robot until the
experiment began. Operators with the safe autonomy mode were reminded that the
robot would take initiative to avoid collisions but that they (the operators) should
seek to avoid collisions as well. Operators with the shared autonomy mode were
reminded to let the robot do most of the driving, but that if they wanted to redirect
the robot, it would temporarily yield control to their joystick commands.
There were 107 participants drawn as volunteers from attendees of the 2003 INL
annual science and engineering exposition at the Museum of Idaho in Idaho Falls.
The participants consisted of 46 females and 61 males, ranging in age from 3 to
78 years old, with a mean age of 14. Participants were asked demographic questions
including their age and gender, and whether they had experience in remote systems
operation. It was determined by self-report that none of the participants had experience
remotely controlling robots, or had knowledge of or access to the remote environ-
ment. Furthermore, none had prior experience with or knowledge of the interface or
robot control system; therefore, it was determined that all of the participants could
be regarded as novice users. Participants were assigned to either the shared or safe
autonomy modes alternately based on their sequence in participation.
On average, participants who used the robot’s shared autonomy mode found an
average of 2.87 objects while those who used the safe autonomy mode found an
average of 2.35 objects (Bruemmer et al., 2005). Comparisons between different

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

age groups and gender were analyzed, but a significant difference in the number
of items found did not exist based on age or gender. Although this experiment was
not intended to support a careful comparison of age and gender groupings, it does
support the claim that the interface allowed a wide variety of participants to find
objects successfully. Participants were able to find objects successfully in both safe
mode and shared mode, indicating that both the guarded motion used in safe mode
and the autonomous navigation behaviors used in shared mode were usable by par-
ticipants. Across all age and gender groupings, performance was better in shared
mode than in safe mode, providing evidence that the robot’s ability to navigate the
environment can actually exceed the ability of a human operator. The performance
benefit experienced by allowing the robot to navigate suggests the potential to use
robot initiative and autonomy not only as a last resort (i.e., when communication
fails or operator workload increases), but as a basis for collaborative interaction
between a human and a robot.
Taken on its own, this first study demonstrates the utility of robot autonomy, but
leaves many questions to be answered by further experiments. The first experiment
did not look beyond overall performance (as measured by items found) to discern the
reasons for the observed difference in performance between safe mode and shared
mode. In response to this limitation, it was determined that the next experiments
should empirically measure differences in operator workload, operator error, and
operator confusion in order to provide deeper insight. Additionally, this experiment
utilized a relatively small search environment. Areas of the environment required
careful maneuvering, but the task was not designed to reward path planning or
strategy. Future experiments address this question by using larger environments that
require some path-planning and strategy to explore the environment efficiently.
Experiment 1 also raised the question of how useful the streaming video provided
by the interface actually was to users when navigating the robot. In tight spaces
where spatial information is important to prevent collisions, participants often
found the entire visual field filled by an immediate obstacle, thereby diminishing
the usefulness of the video for navigation. Furthermore, video information fails to
illustrate obstacles outside of the current visual field, which makes it difficult for
the operator to remember the location of obstacles on the sides of the robot. One
hypothesis was that in such instances video promoted a false sense of spatial aware-
ness and led to operator confusion. As an example, consider the common scenario
of a robot approaching an open doorway in safe mode. The door frame disappears
from the video feed before the robot has reached the doorway. However, the op-
erator, already viewing video information from the next room, may believe that
the robot is already through the door. To prevent a collision with the doorframe,
the robot may stop and refuse to move forward. Although the robot communicates
that it is blocked in front, the user may be confused by the lack of obstacles in the
visual feed. Put simply, the default interface used in Experiment 1 did not provide

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 Bruemmer, Few, & Nelsen

the operator with an adequate representation of the spatial information around the
robot. Experiment 2 was designed to explore the use of a new interface component
intended to better support an operator’s understanding of the spatial information
around the robot.

Experiment.2

Observations from Experiment 1 suggest that video may not provide an adequate
perspective of the remote environment, nor means for the operator to predict robot
behavior or understand the robot’s intentions. However, humans are visual and pre-
fer pictures and diagrams when attempting to understand or communicate (Pashler,
1990). In order to address the human-robot-interaction (HRI) limitations observed in
Experiment 1, some means were required to support collaborative understanding and
yet take advantage of the functional utility associated with visual representation. In
addition to these human factors, there were also significant engineering reasons for
assessing alternatives to video presentation of the remote environment. In particular,
video demands high-bandwidth, continuous communication, and is therefore ill-suited
for many of the very environments where robots could be most useful. Except for
short ranges, transmission of high-bandwidth video is only possible when line of
sight can be maintained either with a satellite or another radio antenna. For instance,
high-bandwidth video cannot be transmitted through layers of concrete and rebar,
making it inappropriate for urban terrain or urban search and rescue. Likewise,
forest and jungle canopy precludes reliable transmission of video.
In response to these human and engineering factors, collaboration between the INL
and Brigham Young University (BYU) was used to develop a new 3D interface
component that could provide a better perspective of the spatial structure of the
environment around the robot. The improved presentation of the spatial information
may help the operator gain insight into the reason for robot initiative and diminish
the likelihood of operator confusion. The purpose of this experiment is to assess the
effectiveness of the 3D interface in a spatial exploration task where the operators
were to use the robot to construct a map of an environment. The hypothesis was that
the 3D interface without video would support the operator in the spatial exploration
task better than the standard interface with video.
The experiment was performed over a seven-day period within the St. Louis Science
Center in 2004 and utilized 64 visitors who volunteered to take part in the experi-
ment. The majority of participants were high school students from schools in the
St. Louis area. These students were not pre-selected, but rather volunteered to take
part in the study while visiting the Science Center. As before, the experiment was
set up as a remote deployment such that the operator control station was located
several stories above the robot arena so that the operator could not see the robot or
the robot’s environment. The arena was built by the production staff of the Science

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

Figure 4. A partial view of the arena built at the St. Louis Science Center

Center and contained artificial rocks, artificial trees, mannequins, and plywood
dividers to create a maze environment (Figure 4).
Due to the distance and physical occlusions separating the control station from
the actual robot environment, analog video was not possible. Instead, state-of-the-
art video compression was used to digitize the analog video into a motion JPEG
(MJPEG) format and wirelessly transmit from the robot to a nearby access point
connected to the building’s network. The building’s wired network was then used
to transfer the video data two stories up to the operator. Exploiting the wired infra-
structure in place throughout the building made it possible to provide continuous,
reliable video at a high frame rate. The presentation speed and resolution of this
video exceeded that possible through an entirely wireless data link. This configura-
tion ensured that the comparison between video and the 3D map display was not
merely a function of current communication bandwidth constraints, but rather an
investigation of the fundamental differences between an interface based primarily
on viewing raw video and one which presented the environment and obstacles as
they relate to the robot’s pose.
Before the experiment, each participant was given basic instructions on how to use
the interface, and no participant was permitted to drive the robot until the start of
the trial run. Participants only used the safe mode autonomy level in order to sim-
plify the comparison of performance between the two interfaces. Participants were
assigned to alternating display conditions (standard interface with video, standard
interface with the 3D window in place of the video) in order to ensure equal num-
bers of participants in each condition and no participant was allowed to operate the
robot in more than one trial. A time limit of three minutes was set in place to help

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

Figure 5. A near-complete map built up by one of the participants

insure that the measured performance was a function of the interface presentation
rather than a function of operator interest or time spent on the task.
At the beginning of each experiment, the map built by the previous participant was
erased by restarting the map-building algorithm on the robot. Each participant was
then instructed to drive the robot around the environment in order to build as large
a map as possible as quickly as possible. All participants were given access to the
same 2D map component (Figure 5) within which the robot presents the map that it
builds as it explores new territory. Exactly half of the participants used the standard
interface and were able to see both the 2D map and the video module. The other half
of participants used the same interface except that the 3D interface module entirely
occluded the video module.
During each trial, the interface stored a variety of useful information about the
participant’s interactions with the robot. Joystick bandwidth was recorded as the
number of messages sent from the joystick indicating a change of more than 10%
in the position of the stick. This information is used as an indirect measure of work-
load (Clarke, Yen, Kondraske, Khoury, & Maxwell, 1991; Khoury & Kondraske,
1991). The interface also recorded the number of joystick vibrations caused by
human navigational error. The map produced by the robot for each experiment was
also saved in order to assess performance based on coverage of the environment.
This approach provided a reasonable assessment of the operator’s ability to explore
the environment in the time available. Immediately after completing a trial, each
participant was asked to rank on a scale of 1 to 10 how “in control” they felt during
the operation, where 1 signified “The robot did nothing that I wanted it to do” and
10 signified, “The robot did everything I wanted it to do.”
In the three minutes provided, 80% of the participants explored over half of the total
environment. One person, a 3D display participant, was able to build the entire map
in the allotted 3 minutes. As described above, task performance was calculated by

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

comparing the map generated during the exploration task with the complete map of
the task environment. This comparison showed no significant difference between
the use of the video module and the 3D module. Using joystick bandwidth as an
indication of human workload and joystick vibration as a metric for human navi-
gational error, analysis shows that operators using the virtual 3D display worked
less and demonstrated fewer instances of navigational error. On average, the joy-
stick bandwidth for participants using the 3D module was 1,057 messages from
the interface to the robot, compared to 1,229 average messages for operators using
the video module. Further, there were, on average, 11.00 instances of navigational
error with the 3D module and 14.29 instances with the video module as measured
by joystick vibrations (Bruemmer et al., 2005).
In addition to reduced workload and fewer navigational errors, use of the virtual 3D
display slightly increased the operator’s subjective “feeling of control” while operat-
ing the robot. The average feeling of control for the 3D display was 7.219 compared
with an average of 7.059 for the video.
The second experiment provided initial evidence that the virtual 3D perspective of
the robot’s environment supported an operator’s ability to reason spatially about
the task and environment better than streaming video information. Results suggest
that although there was no significant change in performance (as measured by the
percentage of the map discovered), there was reduced operator workload, less
navigational error, and a slightly improved sense of control.
One motivation for the development of the virtual 3D display had been to promote a
shared understanding between the robot and the operator of the task and the robot’s
environment. To assess the effectiveness of the virtual 3D display in this regard,
it is useful to consider that a decrease in joystick vibrations not only represents a
reduction in operator navigational error, but also a reduction in the instances where
the operator failed to understand the reason the robot took initiative to protect itself.
Recall that the joystick vibrates only if the operator commands movement in a di-
rection in which the robot has already recognized an obstacle and taken initiative
to prevent a collision. These results indicate progress towards the goal of providing
a representation that supports spatial reasoning and a shared understanding of the
environment. More broadly, these results provide evidence that it may be possible
to support navigational needs of human operators without using video. This find-
ing provides an important counterpoint to opinion within the field of human-robot
interaction that reliable, continuous video is essential for remote navigation (Baker,
Casey, Keyes, & Yanco, 2004).
From an engineering perspective, this experiment shows that it is possible to have
a robot build a map of the robot’s environment as the environment is explored and
communicate the map back to a remote user fast enough to support real-time robot
navigation by the operator. The significance of this result to the area of remote sys-
tems can be seen most clearly when one considers the reduction in communication
bandwidth made possible by using the 3D map display. Whereas the video alone
Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

requires 3,000,000 bits per second (bps), the total interface bandwidth with the
virtual 3D interface was only 64,000 bps. This bandwidth savings allows control
to extend into new domains using data transmission methods that can be used in
underground bunkers, caves, nuclear reactors, and urban search and rescue sites
where it is often impossible to maintain a video feed.
Despite the fact that the human-robot team can function effectively without video,
there is no reason to disregard the potential benefits of video in those instances when
video is available. Experience with operators and subject area experts from energy,
defense, and emergency management contexts indicate that operators expect and
can exploit video in remarkable ways (Casper & Murphy, 2003; Marble, Bruem-
mer, & Few, 2003; Yanco & Drury, 2004;). Many applications require the human
to play a role in visual search and detection. Although this experiment suggests that
video could be replaced with the 3D representation, the optimal interface will likely
provide a dynamic balance between the video and virtual displays.

Experiment.3

The second experiment showed that the virtual 3D display could support the operator’s
comprehension of the spatial information regarding the task and environment. The
question still remains, however, as to whether the use of the virtual 3D display could
be improved with shared control where the human and robot engage in a dynamic
sharing of roles and responsibilities. A previous usability study by Marble et al.
(2003) showed that shared mode offered the greatest potential for operator confu-
sion and frustration. Consequently, it was hypothesized that shared mode might
provide the greatest potential for the virtual 3D display to reduce navigational error
and operator workload.
One goal of this experiment is to compare the safe and shared autonomy modes
when the 3D interface is used. Another goal is to show that the benefits of sharing
control between the human and operator observed in the first experiment are not
merely due to the high cognitive workload placed on the operator when using the
standard interface, but are related to the robot’s ability to navigate itself through the
environment. The typical assumption found in the literature is that robot autonomy
trails behind human performance, but may be useful when the human’s ability
to spatially reason about the task and environment is encumbered (i.e., operator
workload increases, communications fail, or map and position accuracy begins
to degrade) (Goodrich, Olsen Jr., Crandall, & Plamer, 2001; Nielsen et al., 2004;
Trouvain, Wolf, & Schneider, 2003). It was hoped that this experiment could pro-
vide evidence that the robot’s ability to reason about the environment can improve
performance even when the operator’s ability to reason spatially is unhindered (i.e.,
data link connectivity is maintained, human workload is minimal, and mapping and
localization is reliable).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

The task for this experiment is similar to that of Experiment 1 where the partici-
pants were asked to find as many items of interest as possible. For this experiment,
however, in order to minimize individual human workload, the control task was
separated into specific operator functions, namely navigation, driving, and opera-
tion of a pan, tilt, and zoom camera. Instead of using only individuals, groups of
participants were assigned roles where members had responsibility over one aspect
of the robot control. In addition to minimizing individual human workload, an added
benefit of assigning different roles was that it afforded an opportunity to observe
the exchange of information between team members in different roles. In fact, it
became very clear that operators in different roles require different perspectives. For
example, the navigation or planning role requires an exocentric display where the
operator can see the entire environment while the driving role requires an egocentric
perspective so the operator can visualize the robot’s situation in the environment.
As Scholtz (2002) points out, the roles of human operators do not remain static,
and interfaces should be able to adapt accordingly.
This experiment included 120 volunteers grouped into teams of six members. The
participating teams consisted of one team of teachers, three teams of eighth grade
students, and the remainder of the teams being drawn from local high schools.
Participants were recruited from a solicitation of local schools through the St.
Louis Science Center’s outreach program. Participants knew and selected the other
people in their team prior to participation in the experiment. Age and gender were
not recorded due to the fact that most participants were of similar age and the fact
that gender was mixed for each team.
The experiment was run over seven days at the St. Louis Science Center in 2004.
Teams of participants were assigned to alternating conditions so as to ensure equal
numbers of teams in each condition. No participant was allowed to take part on more
than one team. As in the previous experiment, the robot was located in the lower
level of the Science Center, while the control center was located on the top level.
This experiment used the same environment as was used in Experiment 2 with the
same lighting and placement of obstacles. Three mannequins were placed in loca-
tions in the environment designed to force teams to coordinate their information
in order to discover aspects regarding each particular mannequin’s location. The
mannequins remained in place throughout the entire experiment. An equal number
of teams used the shared and safe modes of autonomy while controlling the robot.
The interface components were divided across three separate stations, each with its
own monitor and input devices. No interface component was visible at more than
one control station. Two participants manned each station resulting in a total of six
people dedicated to robotic system control. The stations were arranged in an arc
such that the participants at each station could communicate easily with the others,
but could not see the other displays.
The first control station was dedicated to the application payload, which in this case
was a pan, tilt, and zoom camera. Using a joystick that allowed operation of the vari-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

ous camera controls, the application payload participants used the visual feedback
from the robot to seek out the three mannequins and to provide navigational advice.
The second control station was dedicated to driving the robot. Participants were
permitted to see the virtual 3D interface along with the local environment window,
the sensor status window, and the robot state window from the standard interface
(Figure 2). Primarily, the operators at the driving station used the virtual 3D display,
but were constrained to an egocentric perspective which precluded a global view
of the environment. The final station was the navigation station where participants
had access to the 2D map being built as the robot traveled through its environment.
This gave them a bird’s eye view of the environment and the robot’s position in it.
Additionally, participants at the navigation station were given a hard-copy of a map
showing the locations of the three mannequins. Having two participants at each sta-
tion was not necessary, but ensured that workload was minimal. Task completion
required the three groups to self-organize in order to arrive at and gain a visual lock
on all three of the mannequins as quickly as possible.
On average, less time was required to find the three mannequins for the teams using
the shared robot autonomy mode. The average completion time for shared mode
teams was 467 seconds compared to an average completion time of 641 seconds
for the safe mode teams. Safe mode teams also demonstrated a greater workload, as
measured by joystick movement, than that of their shared mode counterparts. Safe
mode teams made, on average, 2,744 significant joystick movements compared to
an average of 1,725 significant joystick movements for shared mode teams. Using
joystick vibration as a metric for human navigational error shows that safe mode
teams made 25.1 errors on average compared to 16.8 errors for the shared mode
teams (Bruemmer et al., 2005).
As with the first experiment, participants using the shared mode experienced in-
creased performance efficiency when compared to their safe mode counterparts. The
results from Experiment 3 show that with a representation that supports the human
and robot’s ability to reason spatially, performance can be significantly improved by
sharing control between humans and a robot. Moreover, it shows that reducing the
workload placed on the human driver and increasing the importance of strategy and
intelligence does not diminish the performance benefits of sharing control between
human and robot team members.
Previous research has shown that effective teams utilize a shared mental model of
the task and current situation (Cooke, Salas, Cannon-Bowers, & Stout, 2000; Yen et
al., 2001). Similarly, the findings from Experiment 3 suggest that in order to fully
realize the benefits of sharing control between human and robot team members, it is
advantageous to provide a shared model of the environment. Unlike most interfaces
for remotely controlling a mobile robot, the virtual 3D display presents information
from the robot’s environment from a perspective that helps the operator perceive
and comprehend the spatial information around the robot. Improved comprehension
of the robot’s environment makes it easier for the operator to predict robot behavior

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatal Reasonng for Human-Robot Teams 

and understand occasions of robot initiative (Endsley, 1988).

In many operational scenarios, it is not only possible, but probable that the roles of
driving, navigating, and operating the application payload will be distributed among
multiple human operators. Several researchers have pointed out the high cognitive
burden associated with remote deployment of mobile robots and have argued that
effective control requires multiple human operators (Burke et al., 2004; Casper &
Murphy, 2003; Murphy, 2004). Although detailed analysis of these different roles
(i.e., driver, navigator, payload operator) was beyond the scope of this experiment,
anecdotal observations (recorded during and after the experiment) suggest interest-
ing areas for further investigation. One observation was that just as performance
can be degraded by a fight for control between the driver and robot; there is also
the potential for similar conflicts between human operators primarily because they
visualize the information differently. Their ability to reason spatially about the task
and environment is dependent on the different perspectives associated with their
roles. Further experimentation will be necessary to characterize the reasons for these
choices and quantify their effect on team performance. One explanation found in
the literature is that team success depends on the ability of each team member to
understand the perspective of other members (Yen et al., 2001). If this is true, the
most effective human-robot teams will be those that utilize a collaborative model
of the environment and task. Such research questions provide a fertile ground for
further experimentation into the challenges of providing shared representation, not
only between human and robot, but also between humans.

Experiment.4

Experiments 1-3 showed that the INL control architecture, including the 3D interface
and the robot intelligence kernel could reduce reliance on continuous video, increase
overall task efficiency, and reduce operator error and workload. However, it is un-
clear what role the perspective of the virtual 3D environment had in bringing about
these benefits. It is possible that the benefits due to the 3D perspective are largely
due to the simplification brought through the abstraction process. However, it is
also possible that the main benefit of the 3D display is that it provides a perspective
that illustrates more of the spatial information near the robot and is, therefore, more
useful for navigation and exploration tasks than the video display typically used in
remote robot operation (teleoperation). The purpose of this study is to investigate
the role of perspective in the 3D interface in terms of operator error, workload, and
overall task efficiency.
This experiment included 216 participants drawn at random from attendees of the
INL’s 2004 annual community exposition. The participants consisted of 61 females
and 155 males, ranging in age from 3 to 70 years old, with a mean age of 12. The
robot used for this study was an ATRVmini designed by IRobot. Participants were

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 Bruemmer, Few, & Nelsen

assigned the task of discovering the physical structure of the environment using the
safe autonomy mode on the robot and the 3D interface which was populated by the
map as the robot was navigated through the environment.
To test the role of perspective in the 3D interface, each volunteer was assigned one
of four different perspectives (first person, close, elevated, and far). The first person
perspective places the camera inside the robot, so the view is what it would be if
the participant was sitting in the robot. It is similar to the perspective provided by
the video in the standard interface where the user sees the video from the perspec-
tive of the robot’s camera. The close perspective is zoomed out slightly and uses a
virtual camera position somewhat above and behind the robot such that the front
half of the robot is also visible at the bottom of the screen. The elevated perspec-
tive zooms the map display out and places the camera behind and above the robot
such that more of the map is visible in the interface. The far perspective zooms
out further by placing the virtual camera position directly above the robot. It is far
enough above the robot to allow the entire map to be visible on the screen. This is
often referred to as a “bird’s eye view.” Figure 6 illustrates the different perspec-
tives used for this experiment.
A maze environment was constructed on the first floor of the Museum of Idaho using
cubicle wall dividers. On the second floor of the museum, a control station was set
up that consisted of a laptop and monitor to display the interface and a joystick with
which to control the robot. The participants could see the interface, but did not have

Figure 6. Perspectives of the virtual 3D environment used in Experiment 4; clockwise

from top left: 1st person, close, far, and elevated

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatial Reasoning for Human-Robot Teams 369

the ability to see the actual robot or the maze itself as they drove the robot.
Prior to the experiment, each participant was instructed on the use of the joystick
for controlling the robot. They were then requested to build a complete map of the
maze as quickly as possible without running the robot into obstacles. Participants
were also informed that the robot would prevent collisions, but that they should
drive the robot in order to prevent such instances. Each participant used one of the
four perspectives, which were assigned to volunteers in successive, cyclical order.
Information, including the time required to complete the task, the initiative exercised
by the robot, and the total joystick bandwidth used to guide the robot, was measured
and recorded automatically and stored in a data file on the interface computer. Also,
information on age, gender, and a self-assessment of video game skill (on a scale
of 1 to 10) was recorded for each participant.
The results suggest that the 1st person perspective was by far the most difficult to
use, and the other three perspectives (close, elevated, and far) were similar to each
other in their influence on the operator’s ability to control the robot. In particular,
participants using the 1st person perspective took, on average, 133 seconds to discover
the environment, while the close, elevated, and far perspectives had averages of
95, 96, and 97 seconds respectively. Additionally, participants using the 1st person
perspective had an average joystick bandwidth of 1,345, compared to 764, 724,
and 693 for the close, elevated, and far perspectives respectively. There was not
a significant difference in the number of times the robot took initiative to protect
itself between any of the four different perspectives.
The results presented here suggest that the 1st person perspective within the 3D
display is inferior to the exocentric perspectives that show the robot and how it fits
in relation to its environment. Although perspective is a critical factor in terms of
time and joystick usage, it does not, at least for this study, seem to play a critical
role in terms of operator navigational error (i.e., instances which necessitated robot
initiative to protect itself). It is perhaps not surprising that perspective plays an im-
portant role; but what is surprising is that once the perspective moves from the 1st
person to include the robot, there seems to be little difference between the various
exocentric perspectives used. The close, elevated, and far perspectives all seemed to
be very similar in terms of time, joystick usage, and robot initiative. This suggests
that in comparison to the video module on the standard interface, the operator only
needs a little more spatial information concerning obstacles near the robot in order
to improve navigation significantly.
Additional studies will be necessary to further understand the benefits and limi-
tations associated with different perspectives. Most likely, there will not be one
optimal perspective. Rather, perspective should change based on the task element
(e.g., navigation, search, patrol), the level of robot autonomy (e.g., direct human
control, shared control, autonomous tasking), and the number of robots employed
(Scholtz, 2002).

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
370 Bruemmer, Few, & Nielsen

Conclusion

In this chapter we presented tools that improve a human-robot team performance in

navigation and exploration tasks. The tools include behavior-based intelligence on
the robot and a virtual 3D interface through which the operator views the information
from the robot. The role of intelligence on the robot is to reduce the operator’s need
to understand the spatial environment immediately near the robot by empowering
the robot to move and avoid obstacles without any operator control. The role of the
virtual 3D interface is to improve the operator’s ability to perceive and comprehend
the spatial information around the robot, which enables the operator to issue more
informed commands to the robot.
The reason the virtual 3D interface helps so much in the navigation and exploration
experiments presented is because information is presented as it spatially relates to
the robot. In contrast, the standard interface displays information in a manner that
requires the operator to cognitively interpret the information into a holistic under-
standing of the robot’s environment. This extra cognitive effort may impair the
human’s ability to anticipate or predict how the robot will respond to instructions.
With the 3D interface, since information from the robot is automatically integrated
by the manner of the presentation, the operator has more cognitive resources to
anticipate how the robot will respond to instructions.
In the experiments where the human-robot interaction used the shared control mode
as opposed to the safe control mode, performance also improved because the opera-
tor was not concerned with the low-level navigational control of the robot. Since
the operator plays more of a supervisor role when the robot has navigational intel-
ligence, the operator has more cognitive resources to allocate towards anticipating
and predicting how the robot will respond to the environment.
By allowing the operator to visualize the robot’s environment more clearly and
providing the robot with intelligence to handle elementary aspects of navigation
tasks, we bring the robot and the operator into a more unified frame of reference.
With a unified reference frame the human and the robot move towards a true teaming
paradigm where responsibilities and roles can shift dynamically depending on the
needs of the human, the robot, or the task at hand. Improving the ability of robots
and humans to work together has the potential to increase the applications and situ-
ations where robots can be effectively utilized as a valuable team member.

Acknowledgments

The authors wish to thank the following individuals for their contributions to this
work: From the INL: Miles C. Walton, Julie Marble, PhD, Ron Boring, PhD, and

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatial Reasoning for Human-Robot Teams 371

Dan Henry; from Stanford Research Institute International: Kurt Konolige, PhD;
from Brigham Young University: Mike Goodrich, PhD; and from Washington Uni-
versity in St. Louis: William D. Smart, PhD, and James Garner.

References

Baker, M., Casey, R., Keyes, B., & Yanco, H. A. (2004). Improved interfaces for
human-robot interaction in urban search and rescue. In Proceedings of the
IEEE Conference on Systems, Man, and Cybernetics (pp. 2960-2965). The
Hague, The Netherlands: IEEE Press.
Bruemmer, D. J., Few, D. A., Boring, R., Walton, M., Marble, J. L., Nielsen, C.,
& Garner, J. (2005). Shared understanding for collaborative control. IEEE
Transactions on Systems, Man, and Cybernetics Part-A, 35(4), 494-504.
Burke, J. L., Murphy, R R., Coovert, M. D., & Riddle, D. L. (2004). Moonlight in
Miami: A field study of human-robot interaction in the context of an urban
search and rescue disaster response training exercise. Human-Computer In-
teraction, 19, 85-116.
Burke, J. L., Murphy, R. R., Rogers E., Lumelsky, V. J., & Scholtz, J. (2004a).
Final report for the DARPA/NSF interdisciplinary study on human-robot
interaction. IEEE Transactions on Systems, Man, and Cybernetics Part‑C,
34(2),103-112.
Casper, J., & Murphy, R. R. (2003). Human-robot interactions during the robot-
assisted urban search and rescue response at the World Trade Center. IEEE
Transactions on Systems, Man, and Cybernetics-Part B, 33(3), 367-385.
Clarke, D., Yen, S., Kondraske, G. V., Khoury, G. J., & Maxwell, K. J. (1991).
Telerobotic network workstation for system performance and operator work-
load monitoring (Tech. Rep. No. 91‑013R). National Aeronautical and Space
Administration, Johnson Space Center.
Cooke, N. J., Salas, E., Cannon-Bowers, J. A., & Stout, R. (2000). Measuring team
knowledge. Human Factors, 42, 151‑173.
Dennett, D. (1981). True believers: The intentional strategy and why it works. In
A. F. Heath (Ed.), Scientific explanations: Papers based on Herbert Spencer
lectures given in the University of Oxford, Reprinted in Dennet (Ed.), The
International Stance, MIT, 1987, 13-35.
Elfes, A. (1987). Sonar-based real-world mapping and navigation. IEEE Journal of
Robotics and Automation, 3(3), 249-265.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
372 Bruemmer, Few, & Nielsen

Endsley, M. R. (1988). Design and evaluation for situation awareness enhancement.

In Proceedings of the Human Factors Society 32nd Annual Meeting, 1, 97-101.
Santa Monica, CA: HFES.
Goodrich, M. A., Olsen Jr, D. R., Crandall, J. W., & Plamer, T. J. (2001). Experiments
in adjustable autonomy. In Proceedings of the IJCAI Workshop on Autonomy,
Delegation, and Control: Interaction with Autonomous Agents. Seattle, WA:
Morgan Kaufmann.
Gutman, J. S., & Konolige, K. (1999). Incremental mapping of large cyclic environ-
ments. In Proceedings of the IEEE International Symposium on Computational
Intelligence in Robotics and Automation (CIRA), 318-325. Monterey, CA:
IEEE Press.
Khoury, G. J., & Kondraske, G. V. (1991). Measurement and continuous monitoring
of human workload associated with manual control devices (Tech. Rep. No.
91‑011R). National Aeronautical and Space Administration, Johnson Space
Center.
Konolige, K. (2004). Large-scale map-making. In Proceedings of the American
Association for Artificial Intelligence, 19, 457‑463. San Jose, CA:, AAAI
Press.
Marble, J. L., Bruemmer, D. J., & Few, D. A. (2003). Lessons learned from usabil-
ity tests with a collaborative cognitive workspace for human-robot teams. In
Proceedings of the Institute of Electrical and Electronics Engineers (IEEE)
Conference on Systems, Man, and Cybernetics, 1, 448-453. Washington, DC:
IEEE Press.
Marble, J. L., Bruemmer, D. J., Few, D. A., & Dudenhoeffer, D. D. (2004). Evaluation
of supervisory vs. peer-peer interaction for human-robot teams. In Proceed-
ings of the 37th Annual Hawaii International Conference on Systems Sciences
(50130b). Waikoloa, HI: IEEE Press.
Moravec, H. P. (1988). Sensor fusion in certainty grids for mobile robots. AI Magazine,
9(2), 61-74.
Murphy, R. R. (2004). Human-robot interaction in rescue robotics. Institute of
Electrical and Electronics Engineers (IEEE) Transactions on Systems, Man,
and Cybernetics-Part C, 34(2), 138-153.
Nielsen, C. W., Ricks, B.W., Goodrich, M. A., & Crandall, J. Bruemmer, D.J. (2004).
Enhancing functionality and autonomy in man-portable robots. In Proceedings
Ieee Conference on Systems, Man, and Cybernetics (Vol. 3, pp. 2853-2858).
The Hague, The Netherlands: IEEE Press.
Pacis, E. B., Everett, H. R., Farrington, N., & Bruemmer, D. J. (2004). Enhancing
functionality and autonomy in man-portable robots. In Proceedings of the
SPIE Defense and Security Symposium (Vol. 5422, pp. 355-366). Orlando,
FL: SPIE.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 Spatial Reasoning for Human-Robot Teams 373

Pashler, H. (1990). Coordinate frame for symmetry detection and object recognition.
Journal of Experimental Psychology: Human Perception and Performance,
16, 150‑163.
Ricks, B. W., Nielsen, C. W., & Goodrich, M. A. (2004). Ecological displays for
robot interaction: A new perspective. In Proceedings of the IEEE/RSJ Inter-
national Conference on Intelligent Robots and Systems (IROS) (Vol. 3, pp.
2855-2860). Sendai, Japan: IEEE Press.
Scholtz, J. (2002). Human-robot interactions: Creating synergistic cyber forces. In
Proceedings of the 2002 Workshop on Multi-Robot Systems. In Alan C. Schultz
& Lynn E. Parker (Eds.), Multi-Robot Teams: From Swarms to Intelligent Au-
tomata. Proceedings of the 2002 NRL Workshop on Multi-Robot Systems (Vol.
2, pp. 177-184). Dordrecht, The Netherlands: Kluwer Academic Publishers.
Trouvain, B., Wolf, H. L., & Schneider, F. E. (2003). Impact of autonomy in multi-
robot systems on teleoperation performance. In Proceedings of the 2003
Workshop on Multi-Robot Systems (pp. 253‑264).
Woods, D., Tittle, J., Feil, M., & Roesler, A. (2004). Envisioning human-robot coor-
dination in future operations. Institute of Electrical and Electronics Engineers
(IEEE) Transactions on Systems, Man & Cybernetics: Part‑C: Special Issue
on Human-Robot Interaction, 34(2), 210‑218.
Yanco, H. A., & Drury, J. (2004). Where am I? Acquiring situation awareness us-
ing a remote robot platform. In Proceedings of the Institute of Electrical and
Electronics Engineers (IEEE) Conference on Systems, Man and Cybernetics,
(Vol. 3, pp. 2835‑2840). The Hague, The Netherlands: IEEE Press.
Yanco, H. A., Drury, J. L., & Scholtz, J. (2004a). Beyond usability evaluation:
Analysis of human-robot interaction at a major robotics competition. Journal
of Human-Computer Interaction, 19, 117‑149.
Yen, J., Yin, J., Ioerger, T., Miller, M., Xu, D., & Volz, R. (2001). CAST: Collabora-
tive agents for simulating teamwork. In Proceedings of the 17th International
Joint Conference on Artificial Intelligence (pp. 135‑142). Seattle, WA: Morgan
Kaufmann.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 About the Authors

About.the.Authors

Brian.N..Hilton received a PhD in management information systems from Cla-

remont Graduate University, School of Information Science, USA, in May, 2004.
His current research interests lie in spatial information system development, Inter-
net-based geographical information systems, and the use of emerging technologies
in information system development. He holds an MS in management information
systems (Claremont Graduate University) and a BA in economics (Richard Stockton
College of New Jersey). Dr. Hilton is a research associate at the Claremont Informa-
tion and Technology Institute.

***

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 About the Authors 

Tarun.Abhichandani is a PhD student at the School of Information Systems and

Technology, Claremont Graduate University (CGU), USA. His research interests
include middleware for videoconferencing applications, transit-based e-government
initiatives, and P2P technologies. In the past, he has held various positions while
designing and administering organization-wide networking infrastructure, database
applications, and ERP systems. He holds a master’s degree in management of in-
formation systems (MIS) from CGU and a master’s degree in banking and finance
from Mumbai University, India.

Brian.Bakker works as a geographical information systems specialist for Aerial

Services, Inc., USA, a photogrammetry and mapping company based in Cedar
Falls, Iowa. He is currently involved in developing several custom geographical
information systems applications. Mr. Bakker holds a Bachelor of Arts in manage-
ment information systems, and Master of Arts in geography from the University
of Northern Iowa. During his graduate study, he has developed several Web-based
applications using ArcIMS and ArcObjects.

Lars.Brodersen was educated as chartered surveyor, and graduated with a degree of

cand.geom. from Aalborg University, Denmark. He was employed as a map editor at
the Schweizer Weltatlas at the Swiss Federal Institute of Technology Zürich (ETH
Zürich) from 1981 to 1986. He obtained the title as Dr.sc.techn in cartography from
ETH Zürich in 1986, followed by two years as head of the cartography department
in a private company in Odense, Denmark. From 1987 to 1993, he was the head of
section for surveying and registration at the Greenland Home Rule Agency. From
1993 to 2003, he was employed at the National Survey and Cadastre Denmark as
head of section for product development and later as senior research scientist in
geo-communication. In between two years in the Sultanate of Oman as production
control manager at the National Survey Authority in Oman. Since 2003, he has
been employed as associate professor in geo-communication and geo-information
at Aalborg University, Denmark.

David.J..Bruemmer graduated from Swarthmore College in 1998 with a BA in

computer science and a BA in religion. He is currently a principal research scientist
at the Idaho National Laboratory (INL), USA, where he serves as technical director
for unmanned ground vehicle systems. As a consultant to the Defense Advanced
Research Projects Agency, Mr. Bruemmer worked to coordinate development of
autonomous robotics technologies through several offices and programs. Since ar-
riving at the INL, he has led development of a robot intelligence architecture that
has been ported to a variety of robots for applications including remote character-
ization, mine sweeping operations, military reconnaissance, and search and rescue

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 About the Authors

operations. He has been a guest editor of the Intelligent Systems Magazine and is
the program chair for the 2006 Conference on Human Robot Interaction. His in-
terests include autonomous robot behavior, swarm robotics, mobile manipulation,
and teaming between unmanned air and ground vehicles.

Richard.J..Burkhard is an assistant professor of management information systems

in the College of Business at San Jose State University, USA, and a doctoral candi-
date at the School of Information Systems and Technology at Claremont Graduate
University. Rich’s research focuses on virtual organizations, technology-mediated
group work in business and medical environments, and design theory approaches
to support these efforts.

Liping.Di is a professor of geographic information science and director of Laboratory

for Advanced Information Technology and Standards (LAITS), School of Compu-
tational Sciences, George Mason University, USA. Dr. Di received his PhD degree
in remote sensing/geographical information systems in 1991. Before joining GMU
in 2000, he was a chief scientist at Raytheon ITSS Company. Dr. Di has engaged in
geospatial information science research for over 20 years. He has published more
than 100 papers in the fields and led the development of several federal, national,
and international standards on geospatial interoperability.

David.E..Drew holds the Joseph B. Platt Chair in the management of technology

at the Claremont Graduate University, USA. He recently completed a ten-year
term as dean of the School of Educational Studies. His graduate seminars in both
the education and the executive management PhD programs focus on multivariate
statistical analysis and mathematical modeling. Previously he held senior research
positions at the Rand Corporation, the National Academy of Sciences/National Re-
search Council, and the American Council on Education. Earlier he held a research
faculty position at Harvard University, from which he received his PhD For many
years, the focus of Professor Drew’s research and writing has been the evaluation
of science reform efforts.

Fengxian.Fan, faculty member of Kunming University, People’s Republic of China,

is teaching courses on computer sciences. She is also vice-director of the research
institute on computer-control technology at KU and is responsible for research
projects’ technical issues. Prior to coming to Kunming University, she received her
,master’s degree on mechanical engineering at the National University of Defense
Technology in China. In 2001, she came to University of Hawaii (UH) as a visiting
scholar and joined the research group of the “Pods” project supported by DARPA.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 About the Authors 

In 2004, she received her master’s degree of science on information and computer
sciences at UH.

Douglas.A..Few graduated with a BS in computer science from Keene State College

in 2003. Currently Mr. Few is a principal research scientist with the Robotic and
Human Systems group at the Idaho National Laboratory (INL), USA. Mr. Douglas
Few’s interests include human-robot interactions, mixed autonomy robotic systems,
generic robot control architectures, and system reliability algorithms. Prior to joining
INL, Mr. Few supported robotic research efforts around the world in his position as
software support engineer for iRobot Corporation’s Research Robot Division. Ad-
ditionally, he served as a project manager for CHI Engineering Services Portsmouth,
NH, a small engineering firm catering to the needs of the natural gas industry.

Shan.Gao completed her master’s degree in the Department of Information Sys-

tems and Operations Management at the University of Auckland, New Zealand, in
2002. She is currently working as an information system specialist on a number of
information systems projects. She also holds a BCom in information system and
accountancy as well as a BEng in mechanical engineering. Her practical experience
in information system includes system and database design and development, ERP
system implementation, and enterprise systems integration. Her primary research
interests are in decision support systems, geographical information systems, geo-
graphical information systems-enterprise applications integration, and spatial data
mining. She is a co-author of four conference papers and one journal article.

June.K..Hilton has taught all levels of secondary and postsecondary science and
mathematics and has served as mathematics and science department chair in three
secondary schools. She currently teaches physics and chemistry at Claremont High
School, USA. Dr. Hilton holds teaching credentials in Rhode Island, New Jersey,
and California, and also has national board certification in adolescent/young adult
science and physics. She received her PhD in education from Claremont Graduate
University in December, 2003. Her research centers on the use of technology to
increase student achievement.

Peter.Ibach received his master’s degree in computer science from University of

Dortmund (Germany) in 1994. Then, he worked as a system engineer at the IBM
service center in Cologne, Germany. Since 1995, he has been a research associ-
ate at Computer Science Department of Humboldt University, Berlin, Germany,
and since 2003, he has been a member of the Berlin Research Center on Internet
Economics. His research interest focuses geospatial/location-based services and
adaptive service composition.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 About the Authors

Shriram.Ilavajhala graduated from the University of Northern Iowa with a master’s

degree in computer science and geography. He is currently working as a faculty
research assistant with the Department of Geography at University of Maryland,
College Park, USA. His research interests include Web-based geographical informa-
tion systems, open source geographical information systems, user interface design,
information systems, and application software development. He can be reached at
[email protected]

Gunjan.Kalra is a senior software scientist at Quantum Leap Innovations, USA. She

is involved with research and development of novel approaches in the area of data
retrieval, information integration and dissemination, data mining, and multi-agent
systems including policy management. Gunjan has over five years of experience
in designing and developing agent-based systems and in applying artificial intel-
ligence techniques to real-world problems. Prior to joining Quantum Leap, Gunjan
worked as a research programmer in the Laboratory of Information Technology
at University of Maryland, Baltimore County (UMBC). She received her M.S.
degree in computer science from UMBC and her B.E. in computer engineering
from University of Mumbai (Bombay). Gunjan also spent part of her career with
Mahindra British Telecom as a network programmer on their Sterling 5000 Billing
Mediation System.

Lyn.Kathlene is the director of the Colorado Institute of Public Policy (CIPP) at

Colorado State University, USA, an interdisciplinary research institute that addresses
major policy issues facing the Rocky Mountain West. In addition to the Institute’s
research projects, Dr. Kathlene has worked extensively with communities, facilitated
research-based stakeholder dialogues, directed numerous community action proj-
ects, and published reports and articles on citizen participatory planning, research
methods, and the impact of institutional culture on policymaking. Her research
examines the policy formulation process with a focus on how people and groups
that have typically lacked political power can be effectively heard; how participa-
tory policymaking can change the process and outcomes of public policy; and how
community features affect democratic engagement.

Wei-Shinn.Ku received his bachelor’s degree in computer science from National

Taiwan Normal University, Taipei, Taiwan, in 1999. He also obtained an MS in com-
puter science and an MS in electrical engineering from the University of Southern
California, USA, in 2003 and 2006, respectively. He is currently a PhD student in
the Computer Science Department at the University of Southern California. His
research interests include spatial/temporal data management, geographical informa-
tion systems, and peer-to-peer systems. He can be reached at [email protected].

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 About the Authors 

Miroslaw.Malek is professor and holder of chair in computer architecture and com-

munication at the Department of Computer Science at Humboldt University in Berlin,
Germany. His research interests focus on dependable distributed systems including
service composition and ad hoc networks. He has authored and co-authored over
150 publications, and founded, organized, and co-organized numerous workshops
and conferences. Malek received his PhD in computer science from the Technical
University of Wroclaw in Poland, spent 17 years as professor at the University of
Texas at Austin, and was also, among others, visiting professor at Stanford, and
guest researcher at Bell Laboratories and IBM T. J. Watson Research Center.

Robert.K..McNally is a master’s student in geography at the University of North

Carolina at Charlotte, USA.

Ken.McWilliams is a master’s student in geography at the University of North

Carolina at Charlotte, USA.

Anders.Nielsen is a senior cartographer with the National Survey and Cadastre,

Denmark. He was born in Denmark in 1955. Education: Craftsmanship in cartog-
raphy, topography, surveying, photogrammetry, and remote sensing at the National
Mapping Agency in Copenhagen, The Geodetic Institute. Supplementary courses in
these disciplines at the Technical University of Denmark. Occupation: A large part
of his career has been devoted to the photogrammetric survey of Greenland. Since
the late 1990’s full time occupation with geospatial standardization, primarily in
the defense mapping domain.

Curtis.W.. Nielsen graduated with a B.S, an M.S. in computer science in 1999

and 2003, respectively from Brigham Young University (BYU) where he is also
currently pursuing a PhD in computer science (to be finished in January 2006).
Currently, Mr. Nielsen is a principle research scientist with the Robotic and Hu-
man Systems group at the Idaho National Laboratory (INL), USA. Prior to joining
INL, Mr. Nielsen was a research assistant at BYU, where he developed interface
technologies and performed user studies to improve mobile robot teleoperation. Mr.
Nielsen’s interests include computer graphics, robotics, human-robot interaction,
and search and rescue.

Todd.G..Olson is a real estate and land use attorney, as well as a land development
consultant who has specialized in new urbanist, transit-oriented, and conservation-
oriented development. As part of his practice, he has spent over 16 years applying
principals of conservation biology to actual habitat conservation projects at various

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
0 About the Authors

scales, from a 750-acre private conservation bank to a 2-million-acre multiple-

species conservation plan encompassing thousands of different landowners. Todd
developed and documented the “Habitat Transaction Method” in 1992, establishing
himself as a recognized expert in systems of conservation credits. Recently, he has
led a team in developing the Conservation Studio suite of map-based conservation
planning tools.

Anita.Raja is an assistant professor of software and information systems at the

University of North Carolina at Charlotte, USA. She received a B.S. Honors in
computer science with a minor in mathematics summa cum laude from Temple
University, Philadelphia, USA, in 1996, and a M.S. and PhD in computer science
from the University of Massachusetts, Amherst, USA, in 1998 and 2003, respectively.
Her research interests are in the field of artificial intelligence including design and
control of multi-agent systems, meta-cognition, bounded-rationality, adaptive agent
control, multi-agent learning, and organizational design.

Ramanathan.Sugumaran is an associate professor of geography and GeoTREE

director at the University of Northern Iowa, USA. Dr. Sugumaran has over 14 years
of experience in remote sensing, geographical information systems, global positioning
systems spatial decision support systems (SDSS) applications for natural resources
and environmental planning and management. He is and has been working with
federal, state, local, and tribal government agencies (FSLT) for the past 10 years
and developed several SDSS tool and techniques. He has published/presented more
than 70 articles in journals, conferences (national and international), and books.

Vijayan.Sugumaran is an associate professor of management information systems

in the Department of Decision and Information Sciences at Oakland University,
Rochester, Michigan, USA. His research interests are in the areas of geographic
information systems, decision support systems, ontologies and semantic Web, intel-
ligent agent and multi-agent systems, and knowledge-based systems. Dr. Sugumaran
has published over 90 articles in journals, conferences and books. Dr. Sugumaran
is the editor-in-chief of the International Journal of Intelligent Information Tech-
nologies.

David.Sundaram is a senior lecturer in the Department of Information Systems and

Operations Management, University of Auckland, New Zealand. He has a varied
academic (BE in electronics and communications, PG Dip in industrial engineer-
ing, and PhD in information systems) as well as work (systems analysis and design,
consulting, teaching, and research) background. His primary research interests

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 About the Authors 

include the (1) design and implementation of flexible and evolvable information
and decision systems; (2) process, information, and decision modeling; (3) triple
bottom line modeling and reporting; and (4) enterprise application integration with
a focus on ERP-DSS integration.

Gerrit.Tamm studied industrial engineering and financial management at the Tech-

nical University, Berlin, and at the University of California, Berkeley. Dr. Tamm
finalized his doctoral thesis “Web-Based Services: Supply, Demand, and Matching”
at the Berlin-Brandenburg Graduate School of Distributed Information Systems. He
was the director of the Berlin Research Center on Internet Economics in 2003, and
the manager of the Competence Center “Integrated Information Management” at
University St. Gallen. He was business development manager at Lycos Europe and
Teamtoolz, Inc., San Francisco. Dr. Tamm is the founder of Asperado, Absolvent.
de and Electronic Business Forum. Dr. Tamm is a member of the UDDI Advisory
Board.

Qianhong.Tang is a doctoral student in public policy at the University of North

Carolina at Charlotte, USA.

William.J..Tolone is an associate professor of software and information systems

in the College of Information Technology at the University of North Carolina at
Charlotte, USA. He received his doctoral degree from the University of Illinois at
Urbana-Champaign, USA, in 1996. His areas of specialization are in modeling and
simulation, computer-supported cooperative work, secure collaboration architec-
tures, enterprise integration, information environments, agent-based systems, and
meta-level architectures. Since 2003, Dr. Tolone has been lead researcher on several
awards investigating techniques for multi-infrastructure modeling and simulation.

David.C..Wilson is an assistant professor in the College of Information Technology

at University of North Carolina (UNC) at Charlotte, USA. He received his doctoral
degree from the University of Indiana, USA, in 2001. Dr. Wilson is an expert in
intelligent software systems and a leading name in the field of case-based reasoning.
Dr. Wilson’s research emphasizes the development of intelligent software systems to
bridge the gaps between human information needs and the computational resources
available to meet them. It involves the coordination of artificial intelligence tech-
niques with multimedia, database, Internet, and communications systems to elicit,
enhance, apply, and present relevant task-based knowledge. Dr. Wilson’s current
research encompasses several projects focused on smart spatial systems.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 About the Authors

Judith.Woodhall is the managing director of COMCARE, USA, a non-profit orga-

nization dedicated to advancing emergency response. Currently she is focused on
a number of initiatives to deploy data interoperability solutions for the geographic
targeting and cooperative exchange of emergency incident information among
emergency response entities. Prior to COMCARE, Judith served as president for
a regional management consulting firm, as the chief operating officer and chief
information officer for a financial services organization, as chief information of-
ficer for an executive search and Internet solutions company, and as the head of a
global organization responsible for e-business initiatives. Judith earned a bachelor’s
degree from the University of Rochester and a master’s degree in computer systems
management from the Rochester Institute of Technology.

Wei-Ning.Xiang is a professor of geographic information science at University

of North Carolina at Charlotte, USA. He received his doctoral degree from the
University of California at Berkeley, USA, in 1989. Since he joined the faculty at
UNC Charlotte in 1990, Dr. Xiang has conducted teaching, research, and service
activities in the areas of geographic information science, spatial modeling, spatial
decision support systems, multi-attribute assessment and evaluation, and land use/
environmental planning. He was PI or Co-PI of over 30 funded research projects. His
scholarly publications appeared in International Journal of Geographic Information
Science, Environment and Planning B, Lecture Notes in Computer Science, Journal
of Environmental Management, and Landscape and Urban Planning. He served
on the editorial board of Environment and Planning B. In 2002, he was a visiting
professor and research fellow at National Center for Geographic Information and
Analysis, University of California at Santa Barbara.

Genong.(Eugene).Yu is a post-doctoral research associate at the Laboratory for

Advanced Information Technology and Standards (LAITS), George Mason Univer-
sity, USA, and a non-degree graduate in computer science. He has been trained as a
geographer through education led to degrees of Bachelor of Science (1985, Peking
University, Beijing, China), Master of Science (1991, University of Aberdeen,
Aberdeen, UK), and Doctorate (2004, Indiana State University, Terre Haute, USA)
and practiced as a remote sensing and geographical information system specialist
since 1985. His current research interest is in multi-agent systems, geospatial Web
services, neural networks, and multi-scale mathematics for remotely-sensed data
analysis.

Peisheng.Zhao is a research assistant professor of geographic information science

at the Laboratory for Advanced Information Technology and Standards (LAITS),
School of Computational Sciences, George Mason University, USA. Dr. Zhao re-

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permis-
sion of Idea Group Inc. is prohibited.
 About the Authors 

ceived his PhD degree in cartography/geographical information systems in 2000.

His research has been focusing on intelligent geospatial Web services for integration
and interoperation of distributed geospatial data. He has published more than 20
papers in journals and conference proceedings.

Roger.Zimmermann is currently a research assistant professor with the Computer

Science Department and a research area director with the Integrated Media Systems
Center (IMSC) at the University of Southern California, USA. His research activi-
ties focus on streaming media architectures, peer-to-peer systems, Web services,
and geospatial databases. He has made significant contributions in the areas of geo-
technical information and database integration and mobile location-based services.
Dr. Zimmermann has co-authored a book, a patent, and more than 70 conference
publications, journal articles, and book chapters in the areas of multimedia and
databases. He is on the editorial board of SIGMOD DiSC, the ACM Computers in
Entertainment magazine and the International Journal of Multimedia Tools and
Applications. He is a member of ACM and IEEE.

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
  Index

Index

Symbols agent technology 86

ALBS implementation 296
3D environment 368 alternatives 157
3D interface 351 ambiguous match 26
3D interface module 362 annual average daily traffic (AADT) 186
9-1-1 center 65 Apache Xalan parser 123
application agent 101
A
application domains 55
accessibility 135 application domains ontology 55
active components 241 application programming interface (API)
active server pages (ASP) 192 86, 119, 120
adaptation 329 applying standards-based technologies
adaptive location-based services (ALBS) 214
281, 294 architectural layers 67
adaptivity 274 ArcIMS ActiveX Object Connector v. 4.1
administration services 71 192
advanced traveler information systems association of geo-technical and geoenvi-
(ATIS) 44, 51, 52, 54 ronmental specialists (AGS) 111
agency applications 67, 68 asynchronous collaborative discussion
agent-based approach to information inte- environment 268
gration 85 asynchronous collaborative discussion
agent-based, service-oriented framework system 266
97 asynchronous JavaScript and XML
agent-based information integration 93 (AJAX) 185
agent 85, 92, 93, 97 ATIS customer demand factors 52
Agent Communication Language (ACL) ATIS ontology 44, 54
99 authentication/authorization query 71

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
 Index 

Auto-ID Center 291 Colorado 2-1-1 system 334

automatic generation of maps 265 Colorado Institute of Public Policy (CIPP)
autonomous behavior 356 327
autonomous mode 354 color coding scheme 261
autonomy 353, 354, 356, 359, 364, 369 combined scenario 172
COMCARE 64, 79
B commodity communication standards 275
backbone 191 common alerting protocol (CAP) 68
basic organisation 247 common gateway interface (CGI) 264
biodiversity data model 206 common object request broker architecture
Bloom’s Taxonomy 144 (CORBA) 2
Bluetooth 274 community-based 327
borehole 111, 118, 122, 124 community resources 334
borehole data 111 commute trip reduction (CTR) 43
borehole query (BQ) 118, 124 complexity 157
BPEL4WS 296 complex spatial scenario (structural inte-
Brigham Young University (BYU) gration) 173
353, 356 concepts 46
business process 250, 271 concepts and theories ontology 44
business process engineering 250 cone penetration test (CPT) 113, 127
Business Process Execution Language confined animal feeding operation (CAFO)
(BPEL) 19 219, 220, 221, 222, 223, 224,
225, 226, 228, 229, 230, 233,
C 234, 235, 236, 237
conservation criteria developer 209, 215
CAFO decision support system develop- Conservation Studio 207
ment 223 Conservation Studio spatial information
CALPUFF 104 system 215
catalog 2 consistent pose estimation (CPE) 355
catalog service (CAT) 286 consolidation service 99
Catalog Service for Web (CSW) 15, 16 Consortium of Organizations Strong Mo-
categorisations 251 tion Observation Systems (COS-
Cell-ID 274 MOS) 111
CI information mining 314 conventions 287
CIP planning 311 conversion utility 215
CIP professionals 322 core penetration test (CPT) 125
classes 42 Crawler-based geospatial Web service
client-side solutions 22 discovery 17
close perspective 368 create weighted distance map 232
cognitive mapping 328 critical habitat designations 212
collaboration (MoUs) 248 critical infrastructure (CI) 311
collaborative cognitive workspace 372 critical infrastructure protection (CIP) 311,
collaborative discussion 266 322
collection workflow integration system CSC client application 126
(CWIS) 140 CSW interfaces 16

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
  Index

D distributed common object model (DCOM)

2
DARPA Agent Markup Language (DAML) document type definition (DTD) 5, 117
292 drink point laye 165
data 5, 8, 67, 86, 88, 98, 190, 193, 197, DSS objects layer 161
206, 207, 215, 255, 258, 264 dynamic environmental effects model
data accessibility 264 (DEEM) 155
data acquisition interface 207, 215 dynamic value Webs 273, 281, 298
database file (DBF) 190, 193, 197
data interoperability 206 E
data manipulation 255
data model 206 e-commerce tool 215
data processing 86, 88 e-safety network 67
data provisioning (retrieval) service 98 easy-to-use interface 214
data retrieval 88 ebXML information model (ebRIM) 15
data source 88 economic criteria 169
data sources 86 effective use 136
data standards 8, 68 electronic product code (EPC) 278
data transport 67 Emergency Alerting System 69
data validation 88 Emergency Alert System (EAS) 65
data visualization 255, 258 Emergency Data Exchange Language7
DB2 87 (EDXL) 68
decision-making 176 Emergency Medical Services (EMS)
decision support system (DSS) 46, 154, 65, 69
221 emergency provider access directory 69
definition of meaning 244 Emergency Provider Access Directory
Department of Homeland Security (DHS) (EPAD) 63
64, 312 Emerging Technology Toolkit 278
Department of Transportation (DOT) 68 enabling technologies 275
design research 38 endangered species 203
development of a Web-based intelligent endangered species range modeling 212
spatial decision support system 184 enterprise application integration (EAI)
development of conservation banks 213 93, 94, 281
development of habitat conservation plans enterprise information integration (EII)
212, 213 93, 95
DHS Disaster Management eGov Initia- Enterprise Java Beans (EJBs) 72
tive 68 environmental consulting 213
DHS national information exchange model Environmental Decision Support System
(NIEM) 68 (EDSS) 155
digital line graph (DLG) 8 Environmental Protection Agency (EPA)
digital raster graph (DRG) 8 104
directory facilitator (DF) 100 Environmental Systems Research Institute
directory management services 71 (ESRI) 191, 223
directory services 71 EPAD 78
discussion artifact 267 EPAD GIS 73
DisplayXMLNode 127 EPAD GIS services module 73
EPAD N-tier system architecture 70

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
 Index 

EPAD next generation 78 geographic position coordinates 258

EPAD solution set 72 geographic resources analysis support
EPAD system architecture 69 system (GRASS) 258
EPAD system view 70 geographic services ontology 44
EPAD technology components 71 Geography Markup Language (GML)
EPC Discovery Service 280 8, 112, 129, 285, 286, 288, 291
EPC information service (EPCIS) 279 geospatial data 288
EPC Network 273 geospatially enabled directory for emer-
EPC network 291 gency response interoperability 63
ETL 93 geospatial Web services 1, 6, 7, 11, 13, 17,
European article numbers (EAN) 279 21
evaluation of habitat conservation plans geostatistical analysis applied to habitat
213 suitability models 205
event horizon 354 geotechnical 110, 115, 118, 120
exception reports 287 geotechnical file storage Web service 120
eXtensible Markup Language (XML) geotechnical information exchange archi-
68, 184, 277 tecture using Web services 110
extract, transform, and load technology geotechnical information management and
(ETL) 94 exchange (GIME) 110, 115, 116,
122, 124
F German LBS market 273
facilitation services 67 GetCapabilities 287
feature collection 289 GetCapabilities operation 287
Federal Communications Commission GIS-based site suitability decision support
(FCC) 65 system 219
federal land planning 212 GIS analysis 331, 334
File Storage Web Service (FSWS) 115 GIS data 76, 216
filtering 252 GIS query 71
first person perspective 368 global justice XML data model 68
flexible spatial decision 153 global positioning system (GPS) 258, 274
flexible spatial decision support system global real-time enterprise 271, 272
(FSDSS) 157, 160, 161, 163, 164 Globus Toolkit 278
formatting and translation services 99 graphical user interface (GUI) 87, 124
functional hierarchy 321 GRASS 263
GRASS GIS 260, 261
G GRASSLAND 155
GSM 274
gas regulators 322 guarded motion 354, 359
general LBS 290
geo-communication 240, 242, 246 H
geo-information 241
geo-information (data) 248 habitat resources 203
GeoGetXMLFile 127 habitat suitability analyst 208, 215
geographic information system (GIS) Habitat Suitability Index (HSI) 205
63, 68, 71, 76, 134, 137, 154, 184, habitat suitability modeling 203, 205
185, 205, 216, 220, 246, 258, 291, habitat transaction method 209
313, 328, 331, 334 Hamerhead 275

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
  Index

hand drawn cognitive map 330 intelligent agent 200

Help menu 197 intelligent component 191
hierarchical data format for Earth observa- intelligent data management (IDM) 106
tion system (HDFEOS) 8 intelligent transportation systems (ITS) 51
hierarchies 282 intelligent Web-based spatial decision sup-
historical data 88 port system 185
horizontal accuracy 77 inter-company business processes 276
HRI 360 interdependent LBS 291
HTTP 288 interface 5, 267
HTTP protocols 118 interface standards 8
human-centered design 47 internal data format(s) 215
human-computer interaction Design 266 International Organization for Standardiza-
human-computer interface 266 tion 47
human-robot interaction (HRI) 352, 355, Internet-driven standardization 273
360, 363, 371, 373 Internet economy 282
Internet Engineering Task Force (IETF) 5
I Internet 67, 272
Idaho National Laboratory (INL) 352, interoperability 2, 64
353, 355, 356, 358, 367 interpretation system 255
identity rights management services 71 intra-company business processes 276
Illinois River Decision Support System Iowa Department of Transportation
(ILRDSS) 155 (IDOT) 186, 201
implementation domains 164 Iowa State University (ISU) 221
implementation modeler 209, 215 ISDT 43
import services 71 ISO/TC211 4
Incident Command System (ICS) 69 IT-service-catalog 273
incident routing query 75 IT development mechanisms 87
incident sending and routing query 71 J
individual information services (EPCIS)
280 Java 87
information 36, 37, 38, 39, 40, 46, 91, Java 2 Enterprise Edition (J2EE) 72
252, 259, 260, 268, 282, 321 java remote method invocation API (RMI)
information asymmetries 282 2
information economics 282 Java Server Pages (JSP) 72
information integration 91 Java Software Development Toolkit (JDK,
information interpretation system version 1.4.1) 124
259, 260, 268 JBoss application 72
information mining process 321 Jefferson County 336
information system design theory (ISDT) jeopardy determinations 212
36, 37, 38, 39, 40 Joint Robotics Program (JRP) 353
information systems development method-
ologies 46 K
InfoSleuth 93 key goal indicators 284
institutional economics 282 key performance indicators 284
integration 134 keyword queries 106
integration layer 162 killer application 275

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
 Index 

knowledge-driven approach 310 mobile LBS 290

knowledge acquisition 313, 315 models 240, 248, 250
knowledge driven methodology 314 modes of transportation 53
knowledge rendering 317, 319 motion JPEG (MJPEG) 361
knowledge validation 318 Moxi Media™ 73
multi-agent development environment
L (MADE) 99
large databases 86 multi-agent system (MAS) 96, 97, 106
large data environment 86 multi-criteria 154, 157, 176, 225
Latino 336 multi-criteria decision-making (MCDM)
LBS value chains 274 154
left-hand side (LHS) 193 multi-criteria evaluation (MCE) 225
legislation 248, 249 multi-criteria spatial scenarios 176
lightweight services 305 Multi-X Problem 274
Linux cron 264 multistage representation 252
LoadXML 127 MySQL 87
location 169, 272, 275, 278, 306 N
location-based mapping 272
location-based services 272, 275, 278, N-tier architectural design 69
306 National Agriculture Imagery Program
(NAIP) 223
M National Climatic Data Center (NCDC)
main interface of WebISDSS 195 88, 89
management of ontology libraries 56 National Map Accuracy Standard (NMAS)
map agent 102 77
mapping methods 329 National Oceanic and Atmospheric Admin-
map translation agent 102 istration (NOAA) 89
markets 282 National Science Foundation (NSF) 140
material allocation 198 National Weather Service (NWS) 89
material assignment interface of WebIS- natural gas distribution system 321
DSS 199 Natural Heritage Programs 206
materials menu 197 needs assessments 345
metadata 5, 9, 105, 252 network economics 272
methodology 318 new institutional economics 282
Microsoft TerraServer 122 NEXRAD radar image 196
Microsoft Visual Studio .NET framework Next Generation Radar WSR-88D 188
124 NMAS standard 78
middle agents 92 NOAA forecast agent 102
Midwest Transportation Consortium NOAA historical agent 102
(MTC) 201 NOMADS (Networks of Mobile Adaptive
MIKS 93 Dependable Systems) 290
Mississippi Automated System Project NOMADS Campus 272
(MASP) 80 non-federal regional conservation planning
mixed-initiative control 352 212
mobile commerce 306 non-spatial databases 191
normative references 286

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
 0 Index

O orchestration approaches 19
Organization for the Advancement of
object-relational database management Structured Information Standards
system (ORDBMS) 72 (OASIS) 5
Object Management Group (OMG) 2 OWL-based Web service ontology (OWL-
object services layer 161 S) 19
Office of the Secretary of Defense (OSD) OWS implementation specifications 287
353 OXYGENE 128
OGC/ISO 128
OGC Web coverage service 289 P
OGC Web feature service 288
OGC Web map service 287 Pacific Earthquake Engineering Research
OGC Web services common specification Center at Berkeley (PEER) 111
(OWS) 286 passive components 248
online asynchronous collaborative discus- persistence layer 161
sion environment 265 Physical Markup Language (PML)
online communities 266 280, 291
online resource portal 133 planning confined animal Feeding opera-
OnStar2 66 tions in Iowa 219
ontologies 41, 96 planning snow removal using WebISDSS
ontologies for spatial information system 194
development 45 platform-independent structured informa-
ontology 37, 41, 44, 56, 105, 292 tion format 2
ontology-based ISDT 44 Pods sensor networks 258
ontology fundamentals 41 policies 67
Ontology Language for Web (OWL) 17 population growth 336
ontology library system 56 PostGIS9 73
ontology mapping 105 Postgres 87
ontology translation service (OTS) 292 POSTGRES, Version 4.2 72
opaque geospatial services orchestration PostgreSQL 72
19 PostgreSQL database 260, 263
open-source solutions 24 predictive range mapping 206
Open Geospatial Consortium (OGC) presentation layer 162
4, 73, 112, 284 prioritization of land conservancy acquisi-
open geospatial services 286 tions 213
OpenGIS Consortium (OGC) 206 problem identification 168
open grid services architecture (OGSA) problem modelling 169
277 Procedural Rule Language (PRL) 193
open service-oriented computing 276 processes of orchestration 18
open source 71, 79, 214 professional development 135
open source advantage 79 Protected Critical Infrastructure Informa-
open source software 214 tion (PCII) Program 322
operation parameters 287 Protégé 41
operation request 287 provider resource database 344
operator intervention 354 proximate commuting system 43
Oracle 87 proxy data layer 318
public services 247

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
 Index 

Q S
Qualcomm3 66 SAFECOM program 64
quality criteria 169 Safe mode 354
Query & Exchange Web Service (QEWS) SARS (severe acute respiratory syndrome)
115 64
SARS Commission 64
R satellite-based GPS 273
radio frequency identification (RFID) 278 scalable vector graphics (SVG) 114
reactive navigation 354 scenario development 170
read-write (RW) 115 scenario evaluation 176
real-time enterprises 272, 273 scenario integration and instantiation 174
real-time enterprises based on RFID 273 schema 105
real-time weather retrieval component 191 science, technology, engineering, and
rectification 260 mathematics (STEM) 138
relational database management system science education 133
(RDBMS) 113 ScienceMaps 133, 138, 139, 142
repeat rate 334 seamless outdoor positioning 275
repository agents 101 seamless positioning technologies 273
reputation systems 298 search and rescue 352, 364, 371
requirements driven 241 search by index 16
research efforts 91 search by peer-to-peer 17
resource agents 93 search by registry 13
resource definition framework (RDF) 292 second generation 106
resource description framework (RDF) 96 security strategies 251
resource identification 329 semantic issues 25
resources 326, 327 semantic issues on geospatial Web service
response encoding 287 25
response times 93 Semantic Location Network (SEMALON)
retrieval agents 100 271, 290, 291
retrieval methodologies 252 semantic matching 106
RETSINA 93 semantic solutions 27
RFID 273, 274, 290 Semantic Web (SW) 85, 96, 106
rich site summary (RSS) 185 sensor network 255, 256
right-hand side (RHS) 193 server-side solutions 22
robot initiative 359, 360, 363, 369 service-oriented architecture 1
robot intelligence architecture 353 service-oriented architectures 281
ROSRINE (Resolution of Site Response service-oriented computing 276
Issues from the Northridge Earth- service composition 274
quake) 113 service levels 284
Route menu 195 service orchestration 18
Rule Machine Corporation 193 service oriented architecture (SOA) 106,
run-time adaptation 302 273
services 248, 327
shared mode 354

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
  Index

Silene Hawaiiensis 256, 266 standards-based science instruction using

simple non-spatial scenario 170 geographic information systems
simple object access protocol (SOAP) 2, technology 133
15, 68, 114, 115, 118 standards for geospatial web services 5
simple spatial scenario 170 Stanford Research Institute (SRI) 355, 357
site suitability 221 state and local resource regulation 212
slots 42 stationary LBS 290
snow removal asset management system storage 251
(SRAMS) 185, 188 storage service 99
snow removal operations 184, 186, 187 STORM Project 237
software components 118 streptococcus 221
soil classification (CSC) 125, 126 Structured Query Language (SQL) 87
South Carolina Emergency Management structure of geo-communication 244
Division 80 symbolisation strategy 252
Space and Naval Warfare Systems Center system architecture 99
(SPAWAR) 353 systems of care (SOC) 326
Space and Naval Warfare Systems Center system usability scale (SUS) 145
San Diego 353 system use case 251
spatial data engine (SDE) 190, 192
spatial data infrastructure (SDI) T
240, 242, 244, 246 technical capacity 331
spatial data server 215 techno-savvy 133
spatial data transfer standard (SDTS) 8 technology of usability engineering 268
spatial decision-making process 153, Tele mode 354
157, 158 teleoperation 352, 354
spatial decision support system (SDSS) Temporary Assistance for Needy Families
46, 153, 154, 185 (TANF) 328
spatial information system 36, 37, 45, 203 TerraServer 122
spatial layer manager 165 textual forecast data 89
spatial ontologies 291 theories ontology 46
spatial reasoning 351, 353, 363 Thiessen polygons 263
spatial reasoning for human-robot teams topologically integrated geographic encod-
351 ing and referencing (TIGER) 72,
spatial relations 293 76, 78
spatial representation 317 total cost of ownership (TCO) 79
spatial server 191 trace the spread of a disease 177
spatial theme manager 167 tradable conservation credits 203
specification 251 traffic agent 102
specifications 240, 248 traffic translation agent 102
St. Louis Science Center 360, 361, 365 transaction costs 283
standard interface 356 Transitweb 51, 53
standardizing and completion services 98 translation agents 101
Standard Penetration Test (SPT) 111 translucent geospatial Web service orches-
standards 134, 240, 248 tration 19

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
 Index 

transparent geospatial Web services or- W

chestration 19
transport 67 weather data distribution map 259
transportation demand management weather data spatial distribution 259
(TDM) 43 weather forecast information 196
traveler characteristics 52 weather menu interface of WebISDSS 195
tribal land planning 212 weather repository agent 102
trip types 53 weather support to de-icing decision-mak-
ing (WSDDM) 188
U Weather Tool 85, 97, 101, 105
Web-based client 191
U.S. Department of Transportation (US- Web-based service 273
DOT) 188 Web application 4
U.S. Geological Survey (USGS) 113 Web coordinate transformation service
ubiquitous computing 272 (WCTS) 8
unique resource identifiers (URI) 96 Web coverage service (WCS) 7, 206, 286
universal description discovery and inte- Web feature service (WFS)
gration (UDDI) 2, 114, 116, 277 7, 206, 286, 288
universal service-oriented communication Web image classification service (WICS)
306 7
University of Northern Iowa 201 WebISDSS 189, 193
unmanned ground vehicle 353 Web Map Service (WMS 73
updating service 99 Web map service (WMS) 7, 206, 286
usability engineering 267 Web Ontology Language (OWL) 96
use case scenario 299 Web server 191
user driven development 241 Web service aggregator 283
user intervention 354 Web Service Description Language
(WSDL) 2, 15, 114, 115
V Web services dynamic discovery (WS-Dis-
validation service 98 covery) 305
value-chain 242 Web Services Trust Language (WS-Trust)
value model 250 296
vehicular emergency data set (VEDS) 68 Web services (WS) 5, 10, 76, 114,
verification of the geotechnical XML files 240, 241, 296
120 Web Services Description Language
virtual 3D display 357, 363, 364, 366 (WSDL) 296
virtual 3D interface 357 Web service standard stacks 5
Visual Basic 87, 125, 127, 223 well known text (WKT) 74
Visual Basic for Applications (VBA) 125, WFS 287
127, 223 WiMax 274
visualization tools 265 wireless emergency services 275
Visualization Web Service (VWS) 115 wireless positioning systems (WPS) 275
visualized data 264 WLAN 274
Visual Rule Studio 193 WMS 287
voice over Internet protocol (VoIP) 68 World Wide Web Consortium (W3C)
voronoi diagrams 261, 264 5, 95

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc. is prohibited.
  Index

World Wide Web of Geotechnical Engi- XML schemas 287

neers (W3G) 112 XSD (XML schema) 5
WS-Discovery 277 XSLT (XSL Transformations) 115

X Z
XML 273 zero latency 271
XML (eXtensible Markup Language) 2 zoom-to 142
XML borehole files 123

Copyright © 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission of
Idea Group Inc. is prohibited.
Looking for a way to make information science and technology research easy?
Idea Group Inc. Electronic Resources are designed to keep your institution
up-to-date on the latest information science technology trends and research.

Information
Technology Research
at the Click of
a Mouse!

InfoSci-Online

Instant access to thousands of information technology
book chapters, journal articles, teaching cases, and confer-
IGI Electronic
ence proceedings Resources

Multiple search functions

Full-text entries and complete citation information have flexible

Upgrade to InfoSci-Online Premium and add thousands of pricing to
authoritative entries from Idea Group Reference’s hand-
books of research and encyclopedias! help meet the
IGI Full-Text Online Journal Collection needs of any

Instant access to thousands of scholarly journal articles institution.

Full-text entries and complete citation information
IGI Teaching Case Collection

Instant access to hundreds of comprehensive teaching cases

Password-protected access to case instructor files
IGI E-Access www.igi-online.com

Online, full-text access to IGI individual journals,
encyclopedias, or handbooks of research
Additional E-Resources
Sign Up for a

E-Books Free Trial of

Individual Electronic Journal Articles Our Databases!

Individual Electronic Teaching Cases
 g
r o d ucin
Int
IGI Teaching Case Collection
The new IGI Teaching Case Collection is a full-text database contain-
ing hundreds of teaching cases related to the fields of information science,
technology, and management.

Key Features
• Project background
information
• Searches by keywords and
categories
• Abstracts and citation
information
• Full-text copies available for
each case
• All cases are available in PDF
format with instructor files
• Cases are written by IT View each case in full-text, PDF form.
educators, researchers, and Hundreds of cases provide a real-world
professionals worldwide edge in information technology classes or
research!

The Benefits of the IGI Teaching Case Collection

• Frequent updates as new cases are available
• Instant access to all full-text articles saves research time
• No longer necessary to purchase individual cases
• Password-protected case instructor files included in the database

For More Information Visit

www.igi-online.com
Recommend to your librarian today!

A Product Of