Abdalmajeed, Nowar Raad - Final Thesis - Redacted

Pedestrian Level of Service (PLOS) Measurement for
Footpaths in Commercial Streets
Author
Abdalmajeed, Nowar Raad
Published
2020-03-25
Thesis Type
Thesis (PhD Doctorate)
School
School of Environment and Sc
DOI
https://doi.org/10.25904/1912/2482
Copyright Statement
The author owns the copyright in this thesis, unless stated otherwise.
Downloaded from
http://hdl.handle.net/10072/392887
Griffith Research Online

https://research-repository.griffith.edu.au
Pedestrian Level of Service (PLOS) Measurement for Footpaths in
Commercial Streets
Nowar Raad
Bachelor of Civil Engineering (University of Babylon)

Master of Municipalities Management (University of Baghdad)
The School of Environment and Science, Cities Research Institute

Griffith University
Submitted in fulfilment of the requirements of the degree of

Doctor of Philosophy (PhD)
Supervisors
Assoc Matthew Burke
Dr. Jenny Cui
August, 2019
Abstract
Pedestrian level of service (PLOS) is a common method for assessing the quality of
pedestrian facilities. This measure of walking conditions aids in developing street design and
helps in evaluating and prioritising changes to footpaths when retrofitting and in designing
new streets that promote sustainable transport. The development of pedestrian level of service
(PLOS) measures has received considerable academic and practitioner attention, as transport
agencies seek to encourage walking. Engineers and planners determine a link’s performance
using measures codified as levels of service (LOS), which provide guidance on acceptable or
desirable standards. A range of approaches is used in determining pedestrian levels of service
(PLOS) incorporating a wide variety of factors and causing much debate as to what should or
should not be used. Overall, the methodology to evaluate pedestrian level of service mainly
involves determining the various factors which influence pedestrians in terms of perceived
safety and comfort.
This study had a set of inter-related objectives: a) to identify the various factors affecting
PLOS on different pedestrian facilities using systematic methods; b) to select the appropriate
factors for the Australian context that mainly influence the PLOS on footpaths in commercial
areas; c) to examine the importance level of a wide range of factors affecting PLOS, and their
relationship to each other; d) to develop two PLOS models, the first based on our new
approach labeled ‘pedestrian disgruntlement’, and the second based on a satisfaction
approach; and e) to develop a comprehensive PLOS model using an audit tool method to
measure the overall PLOS value. Overall, the main aim of this research is to develop
empirical and systematic PLOS measurements appropriate for the Australian context that
reflect the perceived safety or comfort of pedestrians along footpath segments in commercial
streets.
A mixed-methods design is applied in the research in collecting and analysing data, including:
i) a systematic quantitative literature review using the PRISMA technique examining more
than 600 academic papers; ii) use of the Delphi technique for two rounds, including an online
survey with 36 experts, after which 20 of the same experts participated in a follow-up
‘walkshop’; iii) an on–street pedestrian intercept questionnaire survey of 312 participants,
gathering pedestrians' perceptions of the factors affecting PLOS; and iv) a point system
method by allocating a four point scale for each item, starting from point zero representing the
i
worst condition scenario (i.e., very unsafe/uncomfortable) to point 3, representing the best
condition scenario (i.e., very safe/comfortable).
The thesis results are presented as three published or submitted peer reviewed papers. In
Chapter 4, the review surveyed PLOS models from the peer reviewed literature using a
systematic quantitative literature review method based on the protocol developed by the
Preferred Reporting Items for Systematic Review Recommendations (PRISMA). 58 of the
600 studies surveyed have been produced since the first study examining developing PLOS
models was established in 1970. Later approaches use a much wider range of factors but there
is very little consistency across the studies surveyed. Collectively these factors can be
grouped into themes of comfort, safety and mobility. However, many of the factors being
included have not been empirically studied and almost none of the tools in use have been
tested for such matters as inter-rater reliability. In Chapter 5, a novel, robust method was
developed for determining and measuring the factors that affect PLOS using a satisfaction
approach, and a point system method for evaluating the full range of pedestrians’ opinions on
footpaths in commercial streets. The overall PLOS score shows that the condition of
measuring segments was better than moderate, which reflects pedestrians’ perceptions. In
Chapter 6, the importance of a wide range of factors affecting PLOS was examined to explore
how this relationship may differ for different ages and genders of pedestrians in different
walking environments. The gap analysis approach was used. This has previously been used in
measuring consumer perceptions of Quality of Service (QOS) in the service industries. A new
method was developed to combine measures of pedestrian satisfaction with each factor and
the perceived importance of that factor, creating a novel measure labelled ‘pedestrian
disgruntlement’ to measure PLOS. The results demonstrate how the technique may be used to
improve PLOS measurement and to allow comparison of service to satisfy particular
population sub-groups.
In Chapter 7 a reliable audit tool was developed which, depending on pedestrians’
perceptions, reflects their perceived safety or comfort along footpath segments in commercial
areas. This tool was objectively used to collect and measure factors influencing pedestrian
LOS. More specifically, this study might be considered a guideline for evaluating pedestrian
level of services for such streets in Australian cities in a systemic, reliable and empirical
manner. In this chapter, a guideline manual was also presented, providing specific
information relating to the evaluation of pedestrian level of service (PLOS) on footpaths in
ii
commercial streets in Brisbane using the Pedestrian Level of Service Audit Tool (PLOSAT).
This document provides instructions on how to organise and conduct an audit, the standards
for pedestrian facilities and information about how to use the forms when conducting an
onsite audit.
A major methodological contribution of the study is in adopting a reliable, empirical and
comprehensive method to collect data, measure factors affecting PLOS and evaluate PLOS
for footpaths along commercial streets in Australian cities. The methodological approaches
adopted include a systematic literature review, the Delphi process and a pedestrian intercept
survey, plus inter-rater reliability testing. Few, if any, previous audit tools for real-world
application have used such rigorous approaches in tool development. The applied
contributions of the thesis are in developing an audit tool for PLOS that is systemic, reliable
and empirical, developed with and for practitioners and end-users.
The outcomes of the thesis could be used to improve pedestrian facilities using PLOS
measurements to ensure that different pedestrian facilities are balanced with vehicular
facilities and other land uses. The empirical and systematic PLOS measurements reflect the
perceived safety or comfort of pedestrians along footpath segments. These measurements can
be used objectively to collect and measure factors influencing pedestrian level of service.
PLOS scores can be considered as a basis for selecting or developing pedestrian facilities and
identifying how well a facility or service operates from a user’s perspective. This study is
considered the first attempt in Australia to produce a new PLOS model based on revealed
pedestrians’ perceptions. It could be used as a guide for future studies, including real-world
application by planners, urban designers and transport engineers.
There are some limitations to this study which have implications for further research.
Limitations with the sample and the research methods include the following main issues i) the
results obtained from the Delphi process do not take into account the perceptions of a broader
range of experts; ii) although the pedestrian survey sample was sizeable, obtaining 312
participants for such a lengthy survey, it is not feasible to split the resulting dataset down to
key sub-populations; iii) this study’s outputs, and PLOSAT itself, are solely limited to
commercial streets in the Australian setting. The tool could be used in a few other select
locations, but it is highly specific to this Australian context and does not take into account
features of the path environment that are found elsewhere (i.e., North American cities’
problems with snow).
iii
Statement of Originality
This work has not previously been submitted for a degree or diploma in any university. To the
best of my knowledge and belief, the thesis contains no material previously published or
written by another person except where due reference is made in the thesis itself.
The research reported upon this dissertation was conducted with ethical clearance for human
research as approved by the Griffith University Human Research Ethics Committee (GU
Protocol Number 2016/840) under the supervision of Associate Professor Matthew Burke,
and Dr Jenny Cui.
_________________________________ (26/08/2019)
Nowar Raad
iv
Table of Contents
Abstract ....................................................................................................................................... i
Statement of Originality ............................................................................................................ iv
Table of Contents ....................................................................................................................... v
List of Tables ............................................................................................................................ vii
List of Figures .......................................................................................................................... vii
Acknowledgments ................................................................................................................... viii
Acknowledgment of papers included in this thesis ................................................................... xi
1 Introduction ............................................................................................................................. 1
1.1 Research overview ....................................................................................................... 1
1.2 Research question ........................................................................................................ 3
1.2.1 Primary question 1 ............................................................................................... 3
1.3 Structure of thesis ........................................................................................................ 5
2 Narrative literature review ...................................................................................................... 6
2.1 Background .................................................................................................................. 6
2.2 The study of pedestrian behaviour............................................................................... 8
2.2.1 Urban environment characteristics and physical activity ..................................... 9
2.2.2 Impact of characteristics of physical environment on pedestrians' behaviour ... 10
2.2.3 Impact of urban design qualities on pedestrians’ behaviour .............................. 11
2.3 Complete streets ........................................................................................................ 13
2.4 Planning for pedestrians ............................................................................................ 14
2.5 Concept of pedestrian level of service ....................................................................... 15
2.5.1 Pedestrian level of service (PLOS) evaluation ................................................... 17
2.5.2 Capacity based studies ....................................................................................... 17
2.5.3 Roadway characteristics based studies ............................................................... 20
2.5.4 Capacity roadway characteristics based studies (combination approaches) ...... 22
2.5.5 Conceptually framing previous research approaches ......................................... 26
2.6 Evaluation of environmental audit instruments ......................................................... 31
2.6.1 Previous attempts to develop and test audit tools .............................................. 31
2.6.2 Conceptually framing previous audit tool studies .............................................. 38
2.7 Research gap .............................................................................................................. 43
2.8 Research objective ..................................................................................................... 44
2.9 Research question ...................................................................................................... 44
3 Research approach and methodology .................................................................................... 47
3.1 Research paradigm .................................................................................................... 47
3.2 Case study approach .................................................................................................. 49
3.2.1 Description of the study area .............................................................................. 49
v
3.3 Research design ......................................................................................................... 51
3.4 Research methodology .............................................................................................. 58
3.5 Ethics ......................................................................................................................... 62
4 What are the most important factors for pedestrian level of service? A systematic review of
the literature.............................................................................................................................. 63
4.1 Research summary..................................................................................................... 63
4.2 Research questions addressed in this paper and contributions .................................. 64
4.2.1 Research question ............................................................................................... 64
Primary questions: ............................................................................................................ 64
4.2.2 Summary of contributions .................................................................................. 65
4.3 Statement of contribution to the co-authored published paper .................................. 66
5 Identifying and measuring the most important factors for pedestrian level of service
measurement: results from a Delphi study and a pedestrian perception survey in Brisbane,
Australia ................................................................................................................................... 96
5.1 Research summary..................................................................................................... 96
5.2 Research questions addressed in this paper and contributions .................................. 97
5.2.1 Research question ............................................................................................... 97
5.2.2 Summary of contributions .................................................................................. 97
5.3 Statement of contribution to the co-authored published paper .................................. 98
6 Developing pedestrian level of service (PLOS) model for footpaths in commercial streets
using pedestrian disgruntlement approach ............................................................................. 131
6.1 Research summery................................................................................................... 131
6.2 Research questions addressed in this paper and contributions ................................ 133
6.2.1 Research question ............................................................................................. 133
6.2.2 Summary of contributions ................................................................................ 133
6.3 Statement of contribution to the co-authored published paper ................................ 135
7 Developing and testing pedestrian level of service (PLOS) audit tool for footpaths in
commercial streets .................................................................................................................. 168
7.1 Chapter summery..................................................................................................... 168
7.2 Research questions addressed in this paper and contributions ................................ 169
7.2.1 Research question ............................................................................................. 169
7.2.2 Summary of contributions ................................................................................ 169
8 Discussion ........................................................................................................................... 229
8.1 Research summary................................................................................................... 229
8.2 Response to research questions ............................................................................... 229
8.3 Contributions ........................................................................................................... 232
8.3.1 Conceptual/ theoretical contributions............................................................... 232
8.3.2 Methodological contributions .......................................................................... 232
8.3.3 Applied/practical contributions ........................................................................ 233
8.4 Limitations and future research directions .............................................................. 234
Bibliography ........................................................................................................................... 237
vi
Appendix A: Griffith University thesis preparation policy .................................................... 249
Appendix B: Recruiting email to experts ............................................................................... 255
Appendix C: Expert online survey questionnaire for selecting the most important factors
affecting PLOS ....................................................................................................................... 257
Appendix D: The poster of the advertising of a walking workshop on the PedBikeTrans
Facebook page ........................................................................................................................ 271
Appendix E: Information sheet of experts ............................................................................. 272
Appendix F: Consent forms for experts ................................................................................. 275
Appendix G: Walkshop PrizeDraw ........................................................................................ 276
Appendix H: Walking course sheet ........................................................................................ 278
Appendix I: Workshop sheet .................................................................................................. 283
Appendix J: Information sheet of pedestrian perceptions survey .......................................... 285
Appendix K: Pedestrian Prize Draw ...................................................................................... 289
Appendix L: Pedestrian perception questionnaire ................................................................. 291
Appendix M: Pedestrian level of service audit tool (PLOSAT): Paper form ........................ 300
Appendix N: Pedestrian level of service audit tool (PLOSAT): Google form ...................... 303
Appendix O: The weight assessment form (WAF) ................................................................ 311
Appendix P: Final report form of PLOS ................................................................................ 312
Appendix Q: Summary of current situation, recommendations and remedial action plan
required for audited footpath .................................................................................................. 313
Appendix R: Quick guidelines for using pedestrian level of service audit tool for footpaths in
commercial streets in the field work ...................................................................................... 315
List of Tables
Table 1 Perceptual qualities (source: Ewing & Handy, 2009) ................................................ 12
Table 2 Summary of prominent studies for pedestrian level of service ................................... 27
Table 3 Summary of the key characteristics of the most used audit tools ............................... 39
Table 4 The basic beliefs of received and alternative inquiry paradigms (Guba & Lincoln,
1994)......................................................................................................................................... 48
Table 5 Research tasks ............................................................................................................. 59
List of Figures
Figure 1 Queen street (source: Brisbane City Council, 2014) ................................................... 7
Figure 2 Albert street, Brisbane City (source: Brisbane City Council, 2014) ........................... 7
Figure 3 Hierarchy of walking needs model (source: Alfonzo, 2005) ....................................... 9
Figure 4 Complete streets ......................................................................................................... 14
Figure 5 Common approaches for evaluating PLOS (source: Asadi-Shekari et al. 2013)....... 17
Figure 6 The analysis methods of PLOS (Source: Asadi-Shekari et al. 2013) ........................ 25
Figure 7 Conceptual framework of study by Ewing et al. (2006) ............................................ 35
Figure 8 Map of Brisbane Inner City (source: Google map). .................................................. 51
vii
Acknowledgments
My four years as a PhD student concludes with the writing of this thesis. This period has been
a great learning experience for me, both as a researcher and as a person. I am happy to be able
to use this space to thank all the people without whom this project would never have been
possible. I have had the great fortune to be able to work and collaborate with many very
inspiring, competent, pleasant people, each of whom has contributed to the research in their
own particular way. Although it is my name on the cover, my supportive supervisors, friends
and family have all helped me to achieve my dream and for that I want to give them special
thanks.
First I would like to express my sincere thanks to my incredible supervisory team. Special
mention goes to my enthusiastic principal supervisor, A/Prof Matthew Burke, for his
knowledge, time, encouragement and meticulous support during my PhD journey. His
constructive feedback and reviews have provided a sound basis and great motivation for me in
developing and advancing my research. Matt created an invaluable space for me to do this
research and to develop myself as a researcher in the best possible way. I greatly appreciate
the freedom he has given me to find my own path and the guidance and the support he has
offered when needed. Matt, you have always been there for me and kept me in check, and
over the years, you have become a friend as well. My PhD has been an amazing experience
and I deeply thank you, not only for your tremendous academic support, but also for giving
me so many wonderful opportunities.
My deepest gratitude also goes to my associate supervisor, Dr Jenny Cui, for her constant
support and the sharing of her advice which has enriched my research. Similarly, profound
gratitude goes to Dr. Tooran Alizadeh, who was my initial associate supervisor before moving
to the University of Sydney in 2017. Her fantastic insights and the knowledge she shared have
enriched my academic knowledge and skills, and helped shape my research at the initial stage
of my candidature.
This thesis would not have been possible without the inspiration and support of a number of
wonderful individuals and associations. First, thanks to all the experts at the Pedestrian and
Bicycle Transport Institute of Australasia (PedBikeTrans), who kindly gave their time and
shared their views during the rounds of the Delphi process in this research. Special mention
viii
goes to Kylie Nixon and Aaron Donges from the PedBikeTrans association for their efforts in
recruiting and organising the experts’ meetings; without their contribution I wouldn’t have
been able to complete my study. My thanks and appreciation go to all 312 participants who
were part of the pedestrian intercept survey. I also extend my deepest gratitude to all my
volunteer friends from the Cities Research Institute, who helped me and participated in the
reliability test of the proposed audit tool during the last stage of my study. Their efforts made
it possible to finish my field work in a timely manner.
I would also like to express my sincere acknowledgement to Dr Robert J. Schneider and Dr

Laura Sandt from the Pedestrian Committee at the Transportation Research Board (TRB) for
granting the 2018 Outstanding Paper Award for my paper which was presented at the 97th
Annual Meeting of TRB in the USA. Similarly, my deep thanks goes to the United Nations
University Institute for the Advanced Study of Sustainability and Young Researchers’ School
(YRS) for granting me a two-week scholarship to attend the 2018 ProSPER.Net Young
Research School Programme. I extend my special gratitude to the Griffith Graduate Research
School (GGRS) for their financial support in approving a GGRS - IEIS Conference Travel
Grant to attend the 39th Australasian Transport Research Forum (ATRF) in New Zealand in
2017. Special thanks also go to the Queensland Department of Transport and Main Roads to
include a suggested change and reference to this research in their feedback to a review of the
AUSTROADS Guide to Traffic Management (GTM) series and Guide to Road Design (GRD)
series in March 2019 in which it is likely early research findings of this thesis will be included
in changes to the key guidance for state and local authorities for street design in Australasia.
My warmest gratitude goes to all my colleagues at the Cities Research Institute and the Iraqi
Student Association of Griffith University, who made my PhD journey a joyful experience. I
want to thank Anne Krupa and Michelle Lovelle from the Cities Research Institute for their
tremendous administrative assistance and planning the morning tea events to enliven the
workplace and create a sense of camaraderie among students and staff. Similarly, I am
especially grateful to Sarah Kirkland from the Financial Aid and Scholarships Office and
Mark Taylor from the International Student Advisor Research Office at Griffith University for
their helpful advice which supported me throughout this journey. In addition, I greatly
appreciate the academic editing undertaken by Gillian Warry, who professionally edited my
ix
thesis. I also wish to acknowledge all the journal and conference editors and reviewers for
providing the constructive feedback which has improved my publications.
This work would not have been possible without the financial support of the Higher
Committee for Education Development in Iraq (HCED)/Prime Minister’s Office. I am
especially indebted to Professor Zuhair Humadi, previous General Manager of the HCED
program, the late Professor Abdul Hakim Al-Rawi, Mr. Zeyad T. Flayyeh, and Mr. Ali Al-
Mamoori for their efforts in maintaining and supporting the HCED programme. This has
opened a pathway for me and for many other Iraqi students to pursue and complete a
postgraduate degree at the best universities around the world. I would also like to express my
huge appreciation to all HCED staff in different departments and positions for their effort and
support throughout this four-year journey.
Nobody has been more important to me in the pursuit of this dream than my family.
Therefore, I express my heartfelt gratitude to my parents, Suad and Raad, for their prayers,
everlasting love, unparalleled support and great patience. I am deeply grateful to my older
brother Ammar for always being there for me as a friend and for his continuous support. To
my younger brother Yasir and to my only sister Yousr, many thanks for your continuous love
and support. I would like to thank my wonderful uncle, Professor Saad Ibrahim, who
encouraged, supported and guided me throughout this journey. You are the ultimate role
models. To all of my other family members and friends in Iraq, thank you for your support
along the way.
And most importantly, I owe a huge debt of thanks to my lovely life-partner and loving and
supportive wife, Amani, my adorable son, Hamoody, and my coming baby (Zain or Julia),
who together provide me with unending inspiration. Many thanks Amani for all your moral
and emotional support, understanding, encouragement, and for your unending patience
throughout this hard journey.
Last but by no means least, I dedicate this milestone to my country Iraq, and to the souls that
sacrificed their blood to protect the people and the land of this great country.
Nowar Raad
Brisbane, August 2019
x
Acknowledgment of papers included in this thesis
All papers included are co-authored

Section 9.1 of the Griffith University Code for the Responsible Conduct of Research
(“Criteria for Authorship”), in accordance with Section 5 of the Australian Code for the
Responsible Conduct of Research, states:
To be named as an author, a researcher must have made a substantial scholarly contribution to

the creative or scholarly work that constitutes the research output, and be able to take public
responsibility for at least that part of the work they contributed. Attribution of authorship
depends to some extent on the discipline and publisher policies, but in all cases, authorship
must be based on substantial contributions in a combination of one or more of:
 Conception and design of the research project.
 Analysis and interpretation of research data.
 Drafting or making significant parts of the creative or scholarly work or critically

revising it so as to contribute significantly to the final output.
Section 9.3 of the Griffith University Code (“Responsibilities of Researchers”), in accordance

with Section 5 of the Australian Code, states:
Researchers are expected to:
 Offer authorship to all people, including research trainees, who meet the criteria for
authorship listed above, but only those people.
 Accept or decline offers of authorship promptly in writing.
 Include in the list of authors only those who have accepted authorship.
 Appoint one author to be the executive author to record authorship and manage
correspondence about the work with the publisher and other interested parties.
 Acknowledge all those who have contributed to the research, facilities or materials but
who do not qualify as authors, such as research assistants, technical staff, and advisors
on cultural or community knowledge. Obtain written consent to name individuals.
xi
Included in this thesis are papers in Chapters 4, 5, and 6 which are co-authored with other
researchers. My contribution to each co-authored paper is outlined at the beginning of the
relevant chapter. Appropriate acknowledgements of those who contributed to the research but
did not qualify as authors are included in each paper.
The bibliographic details (if published or accepted for publication)/ status (if prepared or
submitted for publication) for these papers including all authors, are:
Chapter 4:
First presented as:
 Raad, N., & Burke. M. (2018). What Are the Most Important Factors for Pedestrian
Level-of-Service Estimation? A Systematic Review of the Literature. Presented at the
97th Annual Meeting of Transportation Research Board (TRB), January 7-11, 2018,
Washington, D.C., USA.
 First submission on 01/08/2017. Revision submitted on 15/11/2017.
 In 2018, this paper also won an outstanding award presented by the committee on
pedestrian/TRB.
A revised version was then published as:
 Raad, N., & Burke, M. (2018). What Are the Most Important Factors for Pedestrian
Level-of-Service Estimation? A Systematic Review of the Literature. Transportation
Research Record, 0361198118790623.
 Paper was published on 15/03/2018
(Signed) _________________________________ (Date) 26 August, 2018
Name of Student and Corresponding Author: Nowar Raad
(Countersigned) ___________________________ (Date) 26 August, 2018
Supervisor: A/Prof Matthew Burke
xii
Chapter 5:
First presented as:
 Raad, N., & Burke. M. (2019). Identifying the most important factors for pedestrian
level of service measurement: Results from a Delphi study in Brisbane, Australia.
Presented at the 98th Annual Meeting of Transportation Research Board (TRB),
January 13−17, 2019, Washington, D.C., USA.
 First submission on 01/08/2018. Revision submitted on 15/11/2018.
A revised version was then submitted to the journal of “Case Studies on Transport Policy” as:
 Raad, N., & Burke. M., Identifying and measuring the most important factors for
pedestrian level of service measurement: Results from a Delphi study and pedestrian
perception survey in Brisbane, Australia. Case Studies on Transport Policy
 First submission on 26/08/2019
xiii
Chapter 6:
First presented as:
 Raad, N., & Burke. M. (2019). Investigating relationship between the importance of
the factors affecting PLOS and level of satisfaction on footpaths. Presented at the 15th
World Conference on Transport Research (WCTR), May 26-31 2019, Mumbai, India.
 First submission on 15/08/2018.
A revised version was then submitted to the journal of “Travel Behaviour and Society” as:
 Raad, N., & Burke. M., Developing Pedestrian Level of Service (PLOS) model for
footpaths in commercial streets using pedestrian disgruntlement. Travel Behaviour
and Society.
 First Submission on 26/ 08 /2019
xiv
1 Introduction
1.1 Research overview
The focus of this study is the development of a reliable pedestrian level of service (PLOS)
method for measuring factors affecting PLOS for footpaths along commercial streets, and the
production of a new weighted model which takes pedestrians’ revealed perceptions into
account. More specifically, this study develops an approach and method to evaluate and
eventually audit PLOS in Australian cities in a systemic, reliable and empirical manner.
Urban designers, planners, architects and engineers are facing real challenges in responding to
society’s growing concerns in terms of the environment and the quality of urban space
(Soligo, Irwin, Williams & Schuyler, 1998). Streets are an important component of open
public space, and they perform functional and social roles for their users (Koohsari,
Karakiewicz & Kaczynski, 2013). Many researchers have pointed out how different situations
and various densities can influence people’s behaviour in public spaces, although within these
spaces, people can move freely in any direction (Kerridge, Hine & Wigan, 2001). Therefore,
the streets that pedestrians use should be developed in such a way that they create a friendly
environment, and provide high levels of comfort, efficiency, convenience and security. They
should meet the needs of users and provide a convenient environment for pedestrians, while at
the same time, discouraging dependence on vehicles (Sisiopiku & Akin, 2003). This study
evaluates whether a high or low incidence of pedestrian travel in mixed-use, various-density
environments is due to site designing and planning characteristics, and specifically to the
presence of safe and comfortable pedestrian facilities. This study therefore focuses on
footpaths in commercial streets. Commercial streets were chosen for developing a new
footpath PLOS model because of the importance of these places in the built environment as
mixed-use and high-density environments. Nowadays, commercial streets are not only a
destination for shopping, they are also becoming a combination of tourist attractions and
recreation places. Given appropriate land use conditions, pedestrian facility improvement
initiatives in commercial areas can support more sustainable transport and positively
influence the mode choice of pedestrians’ travel.
Since the 1990s, most investigations have paid attention to level of service (LOS) research in
order to meet the needs of vehicle-oriented design without considering the needs of others
who actually use the streets. Level of service is defined as “a quantitative stratification of a
performance measure or measures that represent quality of service’’ (L. Kang, Y. G. Xiong &
1
F. L. Mannering, 2013). The Highway Capacity Manual (HCM) has developed a six-level
scale for PLOS that ranges from A, representing the ‘best’ to F, representing the ‘worst’ level
of service, as with measures of vehicle LOS on highways (L. Kang et al., 2013). In contrast,
insufficient studies have been conducted which address the needs of cyclists and pedestrians,
all of whom have a diverse variety of abilities and ages (Asadi-Shekari, Moeinaddini & Zaly
Shah, 2013). This is changing as a result of the increasing interest in sustainable transport, and
today this approach is aligned with the promotion of sustainability and livability through
giving more attention to pedestrians. Particular performance measures have been at the heart
of much of the planning and design of footpaths, and none has been made more important
than pedestrian level of service. Gallin (2001) has defined pedestrian level of service (PLOS)
as “an overall measure of walking condition on route, path and facility. PLOS reflects users'
perception in terms of sense of comfort and safety” (Gallin, 2001).
This study’s examination of walking conditions aids in the development of street design. This
research will help in evaluating and prioritising the need for footpath retrofit construction on
existing streets, and the design of new streets which will promote the transition to more
sustainable transport. Overall, the main aim of this research is to develop empirical and
systematic PLOS measurements appropriate for the Australian context that reflect the
perceived safety or comfort of pedestrians along footpath segments in commercial streets.
These measurements can be used objectively to collect and measure factors influencing
pedestrian level of service. Although some previous investigators have identified several
factors significantly correlating with PLOS in their models, the relative importance of each
factor and how it apparently influences different walking behaviour in a given context has
been a weakness in previous research efforts. Often, investigators’ personal judgments have
been central to decisions to include, exclude or weight certain factors. Across all these
different approaches and methods, there is wide variety in the choice of factors suggested for
PLOS evaluations. In this study, we sought to introduce a new method to establish a more
reliable and comprehensive PLOS measure through identifying the factors affecting PLOS by
adopting a Delphi technique with a panel of experts, followed by a survey of ordinary,
everyday pedestrians. A Delphi technique with a panel of experts was adopted, followed by a
survey of ordinary, everyday pedestrians. The Delphi technique is a process of deriving group
consensus for any topic when the required information is not available using a systematic
approach (Pikora et al., 2003). The main features of the survey in the Delphi technique are an
2
anonymous, written, multi-stage survey process and controlled feedback (von der Gracht,
2012). Good practice application of the technique requires sound selection of the expert panel,
finalizing the number of participants, carefully formulating the questionnaires, selecting
strong consensus formation processes and criteria; and statistically processing the answers
(Sousa, Sequeira & Ferré-Grau). The Delphi technique was used to help refine a list of factors
identified in a recently published systematic quantitative literature review (Raad & Burke,
2018). The outputs were then used as inputs in a survey of pedestrians. This approach offers a
much more robust means of obtaining PLOS measures.
The outcomes of the thesis could also be used to improve pedestrian facilities using PLOS
measurements to ensure that different pedestrian facilities are balanced with vehicular
facilities and other land uses. The proposed methodology to evaluate pedestrian level of
service mainly involves determining the various factors which influence pedestrians in terms
of perceived safety and comfort, and could be used as a guide for future studies and planners,
designers and transportation engineers.
1.2 Research question
To fulfill the aims of this research, it is necessary that the following primary and procedural
questions be addressed:
1.2.1 Primary question 1

What were the previous approaches, methodologies and factors used for evaluating the
pedestrian level of service in different pedestrian facilities?
Procedural questions:
 What were the previous approaches used in calculating the level of service for
different pedestrian facilities?
 What were the previous factors for inclusion in the level of service calculation for
 What methods were used regarding evaluation of PLOS?
 Where is the research on PLOS? Where was the research undertaken? Who has
undertaken the research and where was it published?
 Which studies have applied reliability tests for data collection methods?
 What evaluation techniques are there for assessing PLOS?
 What methods have been used to identify factors affecting PLOS?
3
 What are the most important factors found in the literature which should be included
in an assessment of PLOS?

What do experts think should be included in PLOS models appropriate for Australian cities?
 What are the possible factors for inclusion in the pedestrian level of service
calculation for footpaths in commercial streets?
 How can an extensive list of the factors identified be refined to select those factors
that should be included in an examination of PLOS appropriate for Australian cities?
 How can a PLOS model be appropriately created?
 How can an overall PLOS and individual PLOS be calculated for each factor based on
the level of pedestrian satisfaction regarding the factors affecting PLOS in a specific
environment?

What do pedestrians think is the importance level of a wide range of factors affecting PLOS?
 How can the importance level of a wide range of factors affecting PLOS be examined
in relation to each other; and how can the differences in this relationship for varied
ages and genders of pedestrians in different walking environments be explored?
 How can a new PLOS model based on a pedestrian disgruntlement approach be
developed and how can differing age groups and genders be explored?

How could a more optimal pedestrian level of service (PLOS) audit tool for footpaths in
commercial streets be developed which would be appropriate for Australian cities?
 How can a PLOS model be appropriately created for the Australian context?
 How can that draft PLOS audit tool be developed?
 Would that tool be reliable?
 How would it be validated?
4
 How would the proposed audit tool be used?
 Would that tool be easy to use and practical in the field?
1.3 Structure of thesis

This thesis consists of seven chapters including an introduction, a literature review, research
methodology, five results chapters and a discussion. The first chapter provides a short
introduction to this research, and outlines the overarching aim and research questions,
expected outcomes and contribution of this study. The second chapter is the literature review
which begins with an explanation of the importance of streets as a major public space, and
focuses on the concept of PLOS. The second chapter also broadly summarises the evaluation
of PLOS, evaluates environmental audit instruments and reviews the previous studies on user
perceptions of PLOS for different streets. Chapter Three presents the methodology used in
this study, including the research paradigm, the study area and the research design. It provides
information about the methodology applied for the data collection and analysis. The results
chapters are in the form of published and in review manuscripts, formatted to meet the
requirements of the peer reviewed conferences and academic journals to which they have
been submitted. To keep the consistency of the thesis format, the results chapters are not in
the form of published and in review manuscripts. There are literature review materials
contained in each results chapter, in accordance with the requirements of conference and
journal manuscripts. There is therefore some repetition of material, including the description
of the study area, the data collection process and the reference lists. The thesis was prepared
in accordance with Griffith University’s policy of including research papers in a thesis. For
reference, this policy is provided in Appendix A.
5
2 Narrative literature review
2.1 Background
This narrative review, which helps explain the development of PLOS approaches, is
supplemented by a systematic quantitative literature review that cogently explores the many
different factors used to measure PLOS (Chapter 4). Here the focus is on broad approaches.
Streets are an important part of open public space in urban places. They provide spaces for
functional, social, and leisure activities, and people have begun to see them more broadly as
social spaces, rather than as channels for movement (Mehta, 2007). Streets provide a mixture
of activity hubs, such as shopping centres, office buildings, automobile dealerships, car parks,
some occasional residential buildings, and vacant spaces, which can be defined as major city
streets or commercial strips (Loukaitou-Sideris, 1997). It is critical that designers and
planners know what people want, and how they think and act in the space in order to shape
and design the urban environment. In their studies, researchers have remarked on the fact that
the street is a social space, rather than just being a channel for movement (Mehta, 2007;
Zadeh & Sulaiman, 2010). Jacobs (1993) argues that streets are places where people socialise
and participate in the wider community; as well, they are used for functional purposes to
facilitate the movement of people and goods and to provide access to adjacent property.
Furthermore, for users’ economic growth, physical health and a sense of community are
associated with a positive relationship between streets and physical environment
characteristics through the creation of a sense of familiarity and belonging to the community
(Zadeh & Sulaiman, 2010). According to Moughtin (2003), streets are a significant element in
defining the quality of the urban environment. He explained that, for specialists, urban
environment design requires an accumulation of technical knowledge to cater for social,
economic, political and religious requirements (Moughtin, 2003). A great street provides a
comfortable, safe environment, as Jacobs (1993) argues. Jacobs expands on this, explaining
that a great street should provide comfortable places for people to walk and sit, attractive
buildings integral with various functions and well-maintained high quality spaces.
6
Figure 1 Queen street (source: Brisbane City Council, 2014)
Much research is being devoted to considering how streets can be best developed in this way.
Most efforts towards the improvement of public spaces have focused on streets, and attempts
to make them more dynamic are ongoing (Mehta, 2006). Koohsari, Karakiewicz and
Kaczynski (2013) argue that the nearness and attractiveness of public open spaces, an
understanding of the built environment, and street configurations that allow people to walk
within open public spaces are associated with the ambient features of the built environment.
These features include keeping pedestrians safe from crime, traffic volumes and aesthetics.
Koohsari et al.’s (2013) research shows that people’s perceptions of the built environment and
street configuration are the main urban design issues that need to be addressed in order to
promote walking in open public spaces. Elements of pedestrian oriented urban design at
street and site levels influence pedestrian trips by making the street environment more
convenient (Barker, 2012).
Figure 2 Albert street, Brisbane City (source: Brisbane City Council, 2014)
7
Architects and urban planners should ensure effective interaction between public space and
public life and social interaction to create enjoyable and comfortable architecture for people.
Public space can be defined as a main part of the built environment (Jacobs, 1993). Various
physical and social advantages can be obtained through open public spaces, such as streets for
different levels of people and society (Koohsari et al., 2013).
Therefore, it is necessary to review previous studies to show practices and methodologies that
have been used to assess the LOS of streets for pedestrians and to obtain a clear picture of the
street evaluation models that consider PLOS. Moreover, assessment processes and designs
require tools and methods that can assess and promote responsible design, and the
introduction of a reliable method with which to evaluate footpaths for pedestrians is essential.
2.2 The study of pedestrian behaviour

The study of pedestrian behaviour is a significant field in transport planning, and is very complex
because it is influenced by environmental design and urban style (Guo et al., 2014). The impact of
the built environment on pedestrians in a space has been the subject of much research during the
past decade. Various aspects of the built environment which influence walking have been found,
leading to the creation of numerous tools and methods for measuring walkability (Saelens &
Handy, 2008). However, it has been argued that there are many factors that influence pedestrian
behaviour in the built environment which are yet to be discovered (Praveen K. Maghelal & Cara
Jean Capp, 2011). For example, Cao, Handy and Mokhtarian (2006) explored pedestrian trips,
including strolling travel (primarily for exercise) and utilitarian travel (primarily for transport and
running errands), and found that pedestrians are influenced by built environment factors and
residents’ self-selection into neighbourhoods. As in other studies, each type of walking has been
linked with different factors affecting walkability (Forsyth, Hearst, Oakes & Schmitz, 2008).
Some scholars have argued that the social factor of people in open public spaces such as streets is
more significant than the physical factors of the environment (Mehta, 2006). Many researchers
have pointed out how different situations and various pedestrian densities may influence the
behavioural relationship of people in the urban space (Kerridge et al., 2001). Mehta (2008)
examined the impact of environmental characteristics, including physical characteristics, land-use
and social qualities on people’s walking behaviour on main streets. He found an integrated
relationship between environmental characteristics, people’s behaviour and perceptions of
walking on main streets, and the importance of social characteristics in promoting walking.
Alfonzo (2005) developed the Hierarchy of Walking Needs Model, which fits a social ecological
framework to provide a conceptual framework for understanding how such factors may work
8
together to affect walking behaviour and the walking decision making process (Alfonzo, 2005).
The Hierarchy of Walking Needs Model hierarchically organises five levels of need and presents
them as precursors within the walking decision making process, as shown in Figure 4. The five
levels of need include feasibility, accessibility, safety, comfort and pleasurability.
Figure 3 Hierarchy of walking needs model (source: Alfonzo, 2005)
2.2.1 Urban environment characteristics and physical activity

Urban environment characteristics including residential density, land use mix, street
connectivity, aesthetics, and safety are closely associated with physical activity, and hence,
they influence the use of active transportation such as walking and cycling (Saelens, Sallis,
Black & Chen, 2003). As past reviews have demonstrated, there is evidence from the public
health and the transportation and planning disciplines that attributes of the built environment,
such as access to facilities and land-use mix, are correlated with people’s physical activity
(Troped et al., 2006).
Research has found that land use is linked to walking and that walking trips tend to remain
simple and direct, despite the potential for complex path choices (Saelens, Sallis & Frank,
2003). Accessibility is an important factor in creating a walkable place; people must be able
to access businesses and activities along the street with ease (Jacobs, 1993). Accessibility is
influenced in many ways, Cervero and Duncan (2003) show that the mix of land use is
9
positively associated with choice of walking; Hess et al. (1999) found that the completeness
of footpath systems has a positive association with pedestrian volumes. Frank et al (2003)
found that the inclusion of off-street paths dedicated to pedestrians and bicyclists as part of a
transportation network can also encourage walking (Frank, Engelke & Schmid, 2003).
However, physical activity in small areas needs to be supported by streetscape urban design
through policies that improve the features of streets, such as improved street lighting,
infrastructure projects, ensuring footpath continuity, improving the aesthetics of the street
area and separation from vehicular traffic (Heath et al., 2006). Therefore, it is critical to study
the correlation between the built environment and aspects of urban design which include
physical activity behaviours, especially after the growing realisation that health outcomes are
influenced by the built environment (Badland, Opit, Witten, Kearns & Mavoa, 2010; Badland
& Schofield, 2005).
2.2.2 Impact of characteristics of physical environment on pedestrians' behaviour

Since the early to mid-1990s there has been increasing interest in studies examining
relationships between characteristics of the physical environment and participation in physical
activity (Troped et al., 2006). Physical environments play an important role in influencing
participation in physical activity, in that those environments with facilities that are relevant to
physical activity, such as pavements and parks, may make it easier for people to be physically
active (Pikora et al., 2002). Studies now reflect an increasing awareness of the role of
environmental factors enabling physical activity; in particular, of active transport through
walking (Ewing, Meakins, Hamidi & Nelson, 2014).
The street itself is a key focus for many such studies. Social behaviour and quality of life are
influenced by the physical elements of a streetscape. Jones (2012) defines streetscape as "the
combination of the physical elements installed within and along the street right-of-way that
affects its usability, functionality, appearance, and identity" (Jones, 2012). Hence, any
modifications to the streetscape may influence active transportation, walkability, bikeability
and social interaction (Jones, 2012). Mehta (2007) pointed out that some physical elements of
a streetscape may directly affect the behaviour of users. He examined the behavioural
responses of people to the environmental quality in three commercial neighbourhood streets.
He gave a definition of a lively street with the presence of a number of people who work in
different activities, especially those activities that are social in nature. Mehta (ibid) found that
people are equally apprehensive about the social, land use, and physical aspects of the street.
10
Zadeh and Sulaiman (2010) studied the impact of culture on the physical characteristics of a
street in the city of Mashhad, Iran. His research showed that culture influences the aspects of
physical and behavioural activities in streets that are significant in catering to people’s needs,
activating public life and increasing the attractiveness of the street (Zadeh & Sulaiman, 2010).
Mehta and Bosson (2010) measured the urban design qualities of businesses on main streets
that differentiate third places. Third places, as defined by Ray Oldenburg, are places where
people regularly come to visit with their friends, neighbours, co-workers and even strangers,
more than in their homes or workplaces (Oldenburg, 2001). Mehta and Bosson (2010)
suggested four physical characteristics, including personalisation of the street front by
businesses, permeability of businesses to the street, seating provided by businesses and shelter
provided by businesses on the street space, all of which are largely under the control of
business owners. These physical characteristics support human use and social interaction and
they may be visible from the street. The outcome of their research showed that third places are
relatively high in both personalisation and permeability to the street, while seating and shelter
provisions are considered the most significant urban design characteristics and make the main
street more sociable (Mehta & Bosson, 2010). Other researchers have focused more on
pedestrian environments within and across streets. As one example, Hamed (2001) developed
separate models, including duration and count models, in order to understand and analyse the
behaviour of pedestrians at pedestrian crossings on divided and undivided streets. He noted
that behaviour differs at crossings with respect to waiting time when pedestrians pass from
one side of the street to the other (Hamed, 2001).
However, physical features individually may not tell us much about the experience of walking
down a particular street. Specifically, they do not capture people’s overall perceptions of the
street environment, perceptions that may have complex or subtle relationships to physical
features (Ewing & Handy, 2009). The urban design literature points to numerous perceptual
qualities that may affect the walking environment (Ewing, 1996).
2.2.3 Impact of urban design qualities on pedestrians’ behaviour

The urban design literature identifies numerous perceptual qualities of the urban environment
that may influence walking behaviour, as shown in Table 1. The visual assessment literature,
which attempts to measure how individuals perceive their environments and better understand
what individuals value in their environments, adds other potentially important qualities
(Ewing, Handy, Brownson, Clemente & Winston, 2006). The visual assessment literature
11
goes beyond the boundaries of urban design to include the fields of architecture, landscape
architecture, park planning and environmental psychology, as perceptual qualities of the
environment figure prominently in these fields as well (Ewing & Handy, 2009). Perception is
the process of attaining awareness or understanding of sensory information. What one
perceives is a result of interplay between past experiences, one’s culture and the interpretation
of the perceived (Ewing & Handy, 2009). Physical features influence the quality of the
walking environment, both directly and indirectly, through the perceptions and sensitivities of
individuals. Urban design qualities are different to qualities such as sense of comfort, sense of
safety and level of interest that reflect how an individual reacts to a place — how they assess
the conditions there, given their own attitudes and preferences (Ewing & Handy, 2009).
Table 1 Perceptual qualities (source: Ewing & Handy, 2009)

Adaptability Distinctiveness Intricacy Richness
Ambiguity Diversity Legibility Sensuousness
Centrality Dominance Linkage Singularity
Clarity Enclosure Meaning Spaciousness
Coherence Expectancy Mystery Territoriality
Compatibility Focality Naturalness Texture
Comfort Formality Novelty Transparency
Complementarity Human scale Openness Unity
Complexity Identifiability Ornateness Upkeep
Continuity Imageability Prospect Variety
Contrast Intelligibility Refuge Visibility
Deflection Interest Regularity Vividness
Depth Intimacy Rhythm
Mehta (2009) studied three neighbourhood commercial streets in Cambridge, Massachusetts,

using behaviour mapping, observations and street user perception through interviews, to
determine which urban design characteristics of the streetscapes support stationary, sustained,
and lingering activities influencing social interaction. Mehta found that physical, land use and
social qualities are critical in gaining an understanding into neighbourhood commercial streets
that support social activities, and that those streets can be ideal locations for stationary,
lingering and social activities (Mehta, 2009).
In today’s rapidly changing environment, there is a need for quality people-friendly places
which serve a diverse range of users through innovative planning and design. Now, streets
play a key role in sustainable life. Therefore, applying the principle of complete streets is an
essential issue in producing quality spaces that may be used for contemporary urban street
12
design. Complete streets are also considered the optimal solution for encouraging good design
where pedestrians, cyclists and public transport have equality of access in and around their
communities and are not disadvantaged relative to motorists.
2.3 Complete streets

One planning movement that has been very influential in commercial street design is the
"Complete Streets" movement. Complete Streets emerged in the United States to ensure that
roads provide safe travelling for all users, not just for vehicles, through imagining and
establishment of transportation facilities. According to the National Complete Streets
Coalition, established in 2005, Complete Streets are “those designed and operated to enable
safe access and travel for all users” (Smith, Reed & Baker, 2010). Complete Streets often
consist of footpaths, bicycle lanes, pedestrian crosswalks, pedestrian refuges, shared-use
paths, designated bus lanes, safe and accessible transit stops, and frequent and safe crossings
for pedestrians. These are considered the typical elements that make up a Complete Street,
and which also influence social behaviour and quality of life (Mehta, 2009). These physical
features provide opportunities that are being used for the activities for which they were
designed—primarily encouraging a pedestrian- and cyclist-friendly environment. The most
significant strategy in use in the transportation field in the United States is a ‘Complete
Streets approach’ in order to make the transportation system accommodate all users, including
vehicle drivers, pedestrians, and bicyclists, as well as those using public transportation. It is
noted that within the communities that adopted Complete Street policies, the population
increased their physical activity levels (Moreland-Russell, Eyler, Barbero, Hipp & Walsh,
2013).
Technological advances, such as computer aided design (CAD), increasingly have helped to
encourage active transportation and social interaction because they enable remote streetscape
design (Badland et al., 2010). Generally, the streetscape has various and multifunctional
aspects in order to cater to the needs of social change, such as transportation needs. Streets are
frequently used as social spaces for meeting and gathering and for engaging in commercial,
cultural and political activities (Bain, Gray & Rodgers, 2012). Multi-functional streetscapes
increase social interaction in the community environment through the access to land uses,
sociability locations and the maintenance of infrastructure (Jones, 2012). The streetscape
should provide opportunities that are not available on an incomplete street, by using the
streetscape to provide a safe environment for pedestrians, bicyclists, motorists, transit users,
13
and travellers of all ages to engage in physical and social activities (Mehta, 2007) and to
move along the street network safely (Smith et al., 2010).
Figure 4 Complete streets

2.4 Planning for pedestrians
Planners are now using a particular set of street design parameters, in part due to complete
streets thinking. During the 20th century, street use was defined by pedestrians. However,
during the 21st century, streets have been functionally separated in terms of travel lanes,
parking lanes, newspaper boxes, pedestrian travel lanes and outdoor seating (Ehrenfeucht,
2006). Siu and Wong (2015), point out that in the early 21st century, streets were only
considered as a key artery of traffic movement, particularly with rapid urbanisation and the
increase in numbers of vehicles, although streets and footpaths have always been considered
as thoroughfares for movement and connection.
As such planners and designers have often designed streets to cater primarily to the needs of
vehicle users with wide travel lanes and narrow footpaths which limit pedestrian accessibility,
and which create other problems such as air pollution, noise, clutter and an increased risk of
accidents (Bain et al., 2012). However, in recent years, the social aspects of streets have been
given more attention, and key issues, including the streets’ functional requirements, have
become more complex (Lillebye, 1996). Lillebye (1996) noted the challenge facing urban
designers as a result of the discrepancy between the importance of the streets as an
architectural element of urban design and their function as a versatile transport artery. He
noted that streets are not always designed according to architectural principles, as far as
dependence on safety and traffic flow are concerned.
14
Planners have changed how they plan streets, using new approaches to work with, and not just
for, communities. In designing effective and friendly facilities for pedestrians it is important
that planners, designers and traffic engineers take into account the revealed needs and
perceptions of actual pedestrians. Another less explored issue is the real-world perceptions of
the people who use the pedestrian facilities. Once this information is revealed, pedestrians’
problems can be addressed and appropriate environments can be created (Sisiopiku & Akin,
2003). During the last three decades, planning processes have shifted, and where once these
processes depended on expert advice, today a participatory approach is preferred. Using
participatory urban planning processes, stakeholders have begun to influence policy decisions,
design processes, investment choices and management decisions affecting their society
(Abdalla, Elariane & El Defrawi, 2015). According to Abdalla et al. (2015), since the 1960s
many researchers and scholars such as Arnstein (1969), Smith (1973), Chamber (1994), and
Fischer (2000) have included communities’ perceptions in their planning processes.
Today, it is necessary for the community’s participation to become a significant tool in the
planning and the design of streets (Skelton, Koplin & Cipolla, 2011). Since the beginning of
the last decade, the concept of the street has begun to be reconsidered, design manuals for
streetscape are changing and, urban planners and designers are developing new improved
criteria for urban transportation planning. As one example, since the 1990s, Melbourne has
transformed its downtown pedestrian realm, responding to the need for the inclusivity of
pedestrians (Siu & Wong, 2015).
2.5 Concept of pedestrian level of service
Particular performance measures have been at the heart of much of the planning and design of
footpaths, and none has been made important than pedestrian level of service.
Recently, transportation engineers and designers have encouraged green transportation, such
as walking and cycling, as a result of the social, economic and environmental challenges
presented by motorised vehicles. Street designs should meet the needs of all users, including
pedestrians and cyclists. Streets are considered to be prominent parts of cities and represent a
key aspect of sustainable transport plans (Zohreh Asadi-Shekari et al., 2013).
Many investigations have used level of service (LOS) for street assessments. LOS can be
defined as an assessment of the quality of service or an overall measure of current conditions
of streets, including facilities, situations, equipment and infrastructure. The level of service is
defined as “a quantitative stratification of a performance measure or measures that represent
15
quality of service” (L. Kang, Y. G. Xiong & F. L. Mannering, 2013). Gallin (2001) has
defined pedestrian level of service (PLOS) as “an overall measure of walking condition on
route, path and facility. PLOS reflects users' perception in terms of sense of comfort and
safety” (Gallin, 2001).
The Highway Capacity Manual (HCM) has developed a six-level scale for PLOS that ranges
from A, representing the ‘best’ to F, representing the ‘worst’ level of service, as with
measures of vehicle LOS on highways (L. Kang et al., 2013). The HCM can be defined as a
guide used to determine the quality and level of service and capacity for given roadway
segments and facilities. The manual was first produced by the U.S.A. Transportation Research
Board in 1950 (Manual, 2000). The importance of the HCM lies in its capacity to provide
appropriate methods for evaluating quality and level of service, and to give the best and most
effective results in specific conditions (Ensley, 2012).
Many studies have been conducted to evaluate streets using LOS in relation to motorised
vehicles; however, similar studies that address the needs of pedestrians have been neglected,
and little research has been conducted to study and analyse the factors that affect the quality
of the walking environment at street level (Landis, Vattikuti, Ottenberg, McLeod &
Guttenplan, 2001). It is important to gather a variety of information about users’ perceptions
and to consider the various requirements of different groups of pedestrians. These
requirements can affect design decisions. The street design for one group may reduce or
enhance other individuals’ capabilities. Therefore, footpaths must be designed to create
accessibility for the majority of users and should also have attractive features for those
individuals who select this type of transportation (Boodlal, 2004). Moreover, qualitative
factors, along with traditional volume and capacity factors, should be taken into account in
analyses of PLOS where the pedestrian experience must not be seen only as a trip from point
A to point B (Jaskiewicz, 2000). However, several attempts have been made by the
transportation planning and engineering community to create new methods for the analysis
and design of the environment, in order to make street and roadway environments more
livable (Landis et al., 2001).
Although the HCM has only determined two variables to evaluate PLOS on footpaths,
including a PLOS score (a linear combination of vehicle exposure and segment configuration)
and average pedestrian space (L. Kang et al., 2013), other factors linked with comfort,
security, safety, and the possible presence of bicycles on the pedestrian facility could affect
16
the results of LOS (Sarkar, 2003). The improvement of pedestrian facilities and operational
characteristics has been given considerable attention in an attempt to decrease congestion, to
provide a significant degree of safety and comfort, and hence achieve the highest level of
quality of life (Lawlor et al., 2003; Lo, 2009).
As will be shown, questions remain about how PLOS should be framed and what factors
should be considered. We turn to such issues in detail now.
2.5.1 Pedestrian level of service (PLOS) evaluation

To assess and calculate walking conditions, the most common method utilises the PLOS.
Various approaches were determined by an examination of previous studies in selecting
effective indicators to evaluate PLOS. As shown in Figure 5, two general approaches were
used to evaluate PLOS: the first is a capacity based model and the second is a roadway
characteristic based model. Some previous studies have attempted to use a combination of
models to evaluate PLOS while other studies have relied on a single approach (Zohreh Asadi-
Shekari et al., 2013).
Figure 5 Common approaches for evaluating PLOS (source: Asadi-Shekari et al. 2013)
2.5.2 Capacity based studies

Capacity based modelling studies have relied on the criteria of highway capacity, modified
and improved so as to be suitable for evaluating pedestrian facilities. This approach was used
in the planning of pedestrian facilities, but information regarding acceptability by pedestrians
is limited (Singh & Jain, 2011).
Fruin (1971) proposed the first attempt at a PLOS based on footpath capacity and pedestrian
volume. He described the relationships between basic vehicle flow and pedestrian volumes,
17
average speed and density using the reciprocal of density, and defined it in area modules of
square feet per pedestrian. Fruin suggested six LOS, in a similar manner to vehicular LOS.
The factors that influence the quality and ease of pedestrian traffic flow were explained in
Fruin's study. Fruin’s (1971) factors involve the possibility of progressing at a desired normal
walking speed, the chance of conflicts between pedestrians in the main traffic flow and those
in crossing directions, the chance of passing a slow pedestrian, and the existence or
nonexistence of two-directional traffic. Fruin explored the theory that the size of the average
area available to each single pedestrian in traffic flow affects these factors, which differ with
respect to the environmental features of the physical facilities. Fruin's model evaluated PLOS
using a survey which was conducted at a segment area to study how pedestrians were affected
by space per person (density), speed of movement and traffic flow. He created LOS criteria
for walkways based on speed-density-flow linear relationships (Fruin, 1971).
Pushkarev and Zupan (1975) investigated flow and space standards for walking facilities and
their application. Their definition of level of service for walking is essentially the same as
Fruin's LOS standards. However, they went beyond the range investigated by Fruin, and their
study depended on low densities large occupancies spaces and short-term fluctuation of flow.
They proposed four ranges to define LOS, including open flow, unimpeded flow, dense flow
and jammed flow (Pushkarev & Zupan, 1975).
A study conducted by Polus et al. (1983) also suggested a range of LOS definitions for
pedestrian traffic. The calibrated and evaluated one- and three-regime linear speed-density
regression models were used to analyse the properties and characteristics of pedestrian flow
on footpaths. They established qualitative and quantitative LOS definitions for pedestrians
based on these analyses. The proposed service level ranges involved four levels, including
free flow (A); restricted, impeded, unstable flow (B); dense flow (C) and jammed flow (D),
which are appropriate for the planning and design of pedestrian pathways (Polus, Schofer &
Ushpiz, 1983).
Based on the Fruin’s LOS principal as a design standard, Tanaboriboon and Guyano (1989)
suggested an LOS method for the design of footpaths in Bangkok. A capacity based model
was used in their study, and included pedestrian characteristics such as speed, density and
flow. These characteristics were examined and analysed in order to develop the LOS criteria
for footpaths in Bangkok. Tanaboriboon and Guyano (1989) developed their set of LOS
18
criteria in planning pedestrian footpaths in Bangkok using Fruin’s LOS design standard in
comparison with these proposed LOS criteria. (Tanaboriboon & Guyano, 1989).
Lam, Morrall and Ho (1995) described pedestrian flow characteristics in Hong Kong. They
developed a basis for the relationship between pedestrian flow characteristics, walking
distance, speed, flow, and density for each type of pedestrian link. Their method has been
used to establish pedestrian design standards and simulation models in Hong Kong (Lam,
Morrall & Ho, 1995).
These studies were all carried out with the average pedestrian flow cases on footpaths and did
not take into consideration the influence of pedestrian flow in groups and platoons. In
contrast, Davis and Braaksma discussed the presence of platoons as well as the mix of
pedestrian traffic, which can have a profound impact on LOS. A new set of LOS criteria was
developed, dependent on the incorporation of relative measures of the prevailing unit flow
rates, area modules, and speed. They defined platoon flow as the grouping or bunching of
pedestrians because of internal or external traffic impedance (Davis & Braaksma, 1987).
In addition, Jaskiewicz (2000) went further than just the study of volumes and capacities in
evaluating PLOS. He suggested a method of evaluating PLOS based on trip quality, in order
to encourage walking as a viable alternate form of transportation. Nine specific evaluation
measures were described, including enclosure/definition, complexity of path networks,
building articulations, complexity of spaces, transparencies, overhangs/awnings/varied roof
lines, buffers, shade, trees, and physical components/conditions. A set of safety issues,
volume and capacity considerations, and qualitative design factors contributed to each of
these nine measures. This study found that more attention should be paid to PLOS evaluation
regarding some qualitative variables on roadway environment, highlighting pedestrian
comfort and safety as well as the traditional variables of volume and capacity.
Capacity, volume and speed were also discussed by the Transportation Research Board
(2000) in their evaluation of PLOS. LOS analysis for each factor affecting pedestrian facilities
was provided by the HCM 2000 with the assumption that pedestrians display behaviour
similar to vehicles, for example, travelling in a linear path. Faster speed indicates efficient
flow and, to a degree, more people indicate congested conditions (Manual, 2000). However,
the HCM methodology has been criticised by later researchers, because pedestrian and
vehicular traffic have been dealt with under the same model (Asadi-Shekari, Moeinaddini &
Zaly Shah, 2012). The HCM manual did not take into consideration other universal cultural
19
contexts; it relied simply on American culture (Singh & Jain, 2011). In addition, Mateo-
Babiano and Ieda (2007) considered HCM methodology to be partly inaccurate as a result of
the changes to and complexity of pedestrian movement patterns, such as avoidance of
barriers, choice of alternative routes, resting on a bench, stopping to buy food or to converse,
and so on (Mateo-Babiano & Ieda, 2007).
The methods and models discussed above also have not taken into consideration some key
variables related to the physical environment of the street, such as qualitative dimensions,
facilities and street furniture. A different set of researchers took up this challenge.
2.5.3 Roadway characteristics based studies

Roadway characteristics based methods are constructed on the characteristics and features of
the footpath environment or pedestrian facilities. It is essential that pedestrian perceptions be
used in these methods, and that the comfort level of pedestrians while encountering certain
roadway characteristics is determined.
A milestone in PLOS research was attained by Lautso and Murole (1974). They described the
influence of environmental factors on pedestrian facilities (Muraleetharan, Adachi, Hagiwara
& Kagaya, 2005). Later researchers expanded and improved this approach, including further
factors which have additional impacts on PLOS, as demonstrated in the following studies.
Gregory Benz et al. (1986) developed a new methodology they named the time-space
approach. This methodology was best suited to transportation terminals and other complex
pedestrian spaces; however it could be used for footpaths. The time-space approach is
characterised by pedestrian activities that generate time-space needs. The areas where these
activities take place are time-space zones, and they have limited capacity to meet pedestrian
time-space needs (Benz, 1986). A qualitative model was developed by Sarkar (1993), in order
to evaluate PLOS depending on a set of factors affecting PLOS. The key factors considered
by this qualitative model to assess streets included safety, security, convenience and comfort,
continuity, system coherence, and attractiveness. After studying the levels of activity and use
in the pedestrian environments in Munich and Rome, he suggested six service levels. This
range of service levels for pedestrian environments was classified in a range from the most
pedestrian friendly street, represented by PLOS level A, to incomplete streets, represented by
PLOS level F (Sarkar, 1993). Sarkar's investigation qualitatively defined pedestrian
environment features rather than quantifying them. In Sarkar's method, it is difficult to
20
measure each factor affecting PLOS in a practical field; especially when each factor is related
to the other.
In contrast, a quantitative method to evaluate PLOS in streets was suggested by Khisty
(1994). Khisty's study relied on the same factors proposed by Sarker (1993). The survey
responses were used to determine the most significant factors affecting PLOS. He found that
safety and security (i.e., protection conditions) are the most important factors in his model
(Khisty, 1994). Khisty’s method is considered inexpensive and its use can supplement the
qualitative LOS of the facility on a point scale, but unfortunately it is not considered to be
easy to use because the results of these scales are not always easy to interpret. Also, the
scaling systems may not represent pedestrians’ perceptions.
Dixon (1996) proposed a PLOS evaluation criterion that involved the provision of basic
facilities, conflicts, amenities, motor vehicle LOS, maintenance and travel demand
management, and multimodal provisions. This method, which was used to evaluate bicycle
and PLOS performance measures for a roadway in Gainesville, Florida, used a point system
from one to twenty-one, which resulted in LOS ratings from A to F. The scoring system was
developed with sensitivity to characteristics that may be mutually exclusive or inclusive, in
order to determine all possible combinations of points. The PLOS ratings were defined by the
measures of pedestrian safety features and the level of automobile oriented development
characteristics along the corridor (Dixon, 1996). Although the LOS ratings described the
degree to which facility provisions encouraged pedestrian use, the drawback of this model
was that the weighting of factors affecting users was selected arbitrarily.
A scaling system was developed for PLOS assessment by Miller et al. (2000). A simulation
and data analysis tool was used to calibrate a PLOS scale. A scale method was calibrated
using 3-D visualization (computer aided modelling techniques consisting of still shots and
animations). A conjoint technique was used to collect the factors affecting PLOS at
intersections, and these factors involved security, handicapped access, median openings,
signalisation parameters, pedestrian warning devices, crosswalks and the speed limits of road
crossings (Miller, Bigelow & Garber, 2000). In this approach, the need for pedestrian
crossings was systematically considered as the key issue. This methodology is applicable in
urban areas where pedestrian needs beyond physical capacity are to be explicitly considered.
Based on the previous literature in the 20th century, Henson (2000) listed a fuller
consideration of factors affecting PLOS, which included consideration of at least five broad
21
environmental factors: comfort, convenience, safety, security and economy (Henson, 2000).
By contrast, Sarkar (2003) suggested another PLOS model based only on the key attributes of
comfort. These attributes were used to qualitatively grade the physical factors of footpaths
(adequate footpath, continuous footpath, comfort for vulnerable users, footpath free of
impediments, comfortable walking surface, seating and protection from extreme weather
conditions), the psychological aspects (ability to maintain desired walking speed, ability to
participate in various pedestrian activities) and the physiological factors (noise and pollution).
Sarkar's model involved two separate evaluations, including service levels which give
standards for the overall desirable and undesirable comfort conditions at the macro level, and
quality levels that look at the micro level finer details of pedestrian comfort. Research from
urban design, environmental psychology, landscape architecture and urban planning was used
to develop this model. Although Sarkar’s (2003) model dealt with the micro level conditions
of comfort for pedestrians, it drew from a small number of standards and factors for different
street indicators, hence, this model did not provide a comprehensive evaluation of inclusive
facilities.
2.5.4 Capacity roadway characteristics based studies (combination approaches)

Pedestrian facilities have also been evaluated by combined methods. Mozer et al. (1994)
suggested a pedestrian, bicycle, auto, transit level of access (P-BAT LOA) model to
determine the LOS for all pedestrians, bicycles and transit. This method used an equation
based on primary and secondary variables. The primary variables involved walk area width-
volume, walk area outside lane buffer, outside lane traffic volume and outside lane motor
vehicle speed. Moreover, the secondary variables included walk area penetration, heavy
vehicle volume and intersection waiting time (Mozer, 1994). Mozer's model had insufficient
assessment tools to evaluate streets for a full range of pedestrians, and the ratings were not
based on users’ opinions, however, the model was an advanced one in its time.
A later PLOS model was introduced by Landis et al. (2001), which focused on determining
the factors that may significantly influence a pedestrian's feeling of safety or comfort on
roadway segments between intersections. These factors involve lateral separation elements
between pedestrians and motor vehicle traffic, motor vehicle traffic volume, the effect of
motor vehicle speed, motor vehicle mix and driveway access frequency and volume. A
stepwise multi-variable regression analysis was used in this model using 1250 observations
from an activity that placed 75 people walking on a roadway course in the Pensacola
22
metropolitan area in Florida (Landis et al., 2001). The gathering of these factors into a
mathematical expression, tested for statistical reliability, provided a measure of the roadway
segment’s PLOS. This approach is similar to the methods used to assess the motor vehicle
LOS in the HCM.
Gallin (2001) developed a unique model for evaluating PLOS based on several factors that
affected LOS. The factors in Gallin’s model were classified into three categories including
physical characteristics, location factors and users’ factors. A point system method was used
to calculate PLOS (Gallin, 2001). The drawback of Gallin’s model is that the weighting of
factors affecting PLOS appeared to be based on personal decisions (i.e., experts’ judgments
only). In addition, the factors included in this model are not representative of all aspects
affecting the pedestrian environment. Therefore, the results derived from this model are, to
some extent, limited.
Muraleetharan et al. (2004) proposed a method to evaluate PLOS of footpaths using another
combined method. The factors affecting PLOS on footpaths and at intersections were
weighted by relative importance by using a conjoint analysis technique. The factors in this
method included width separation, obstructions, flow rate and bicycle events on the footpath,
and space at corners, crossing facilities, turning vehicles and delays at intersections. This
method was useful for evaluating PLOS as well as for determining which factors contribute to
low and high LOS (Muraleetharan, Adachi, Hagiwara & Kagaya, 2004 B; Muraleetharan et
al., 2005; Muraleetharan, Adachi, Hagiwara, Kagaya & Uchida, 2004 A).
In addition, Petritsch et al. (2006) identified total width of crossings at conflict locations and
average volume on the adjacent roadway as a primary factor influencing pedestrian
perceptions of LOS. Petritsch et al. developed a field-calibrated PLOS model for roadway
facilities for pedestrians travelling along urban arterials with footpaths. Data for the model
was obtained by modelling approximately 500 combined real time perceptions (observations)
from pedestrians walking a course along a typical U.S. metropolitan urban area’s streets.
Pearson correlation analyses and stepwise regression modelling were used to analyse
participants’ perceptions of how well urban arterials with footpaths met their needs (Petritsch
et al., 2006). This model focused only on participants’ perceptions and relied on the narrow
range of the factors affecting PLOS, but neglected other aspects.
Byrd and Sisiopiku (2006) compared the more commonly accepted methods of determining
PLOS for footpaths, including the HCM 2000, Landis, Australian, and trip quality methods.
23
The comparison found that, using these methods, it is possible to receive multiple LOS ratings
for the same facility under the same conditions. The study concluded that a combined model
could be developed that synthesises the quantitative and qualitative factors that affect
pedestrian operations (Sisiopiku & Byrd, 2006).
Bian et al. (2007) conducted a footpath intercept survey to measure pedestrian perceptions of
footpath LOS in Nanjing, China. A total of 501 people were interviewed on nine footpath
segments. A questionnaire survey was conducted with participants, who were classified into
three groups based on age, gender and walking experience. They identified lateral separation
from traffic, motor vehicle volume and speed, bicycle volume and speed, pedestrian volume,
obstructions, and driveway frequency as factors influencing PLOS in this Chinese context.
They defined LOS 1 as ‘excellent’, and LOS 2-6 as ‘good’, ‘average’, ‘inferior’, ‘poor’ and
‘terrible’, respectively. A linear regression model was fitted to the data, in order to predict the
mean LOS rating. The numerical score predicted by the model was converted to a letter grade
using the following limits: LOS A <= 1.5, LOS B <= 2.5, LOS C <= 3.5, LOS D <= 4.5, LOS
E <= 5.5 (Bian, Wang, Lu, Ma & Tan, 2007).
The Danish Road Directorate sponsored a study by Jensen (2007) to develop methods for
objectively quantifying pedestrians’ and bicyclists’ stated satisfaction with road sections
between intersections (Jensen, 2007). Jensen developed a cumulative logic model to capture
various factors affecting pedestrian satisfaction on footpaths, taking Danish conditions into
consideration. The models showed that many variables influence pedestrians’ and bicyclists’
satisfaction and LOS on roadway segments. He used video clips to identify variables that
significantly influenced the level of satisfaction. Existing traffic operations, geometric
conditions and other variables affecting pedestrians’ and bicyclists’ satisfaction were chosen
by 407 randomly selected Danish people who were shown video clips from 56 roadway
segments. Roadway segments and video clips were described by 150 variables, which would
then be used to model pedestrians’ and bicyclists’ requirements.
Asadi-Shekari and Zaly-Shah (2011) discussed PLOS using several indicators from various
guidelines which met different pedestrian needs and determined existing streets’ failures. The
analysis examined many pedestrian factors from more than 20 guidelines in developed
countries, in order to promote the accuracy of PLOS in this method. This model is useful for
improving existing streets and it produces a guideline for designing new streets, taking
pedestrians into consideration. However, no consideration of users' perceptions was taken into
24
account and the model was based on the authors’ judgement (Asadi-Shekari & Zaly-Shah,
2011).
Hidayat, Choocharukul and Kishi (2011) proposed an alternative model for evaluating PLOS
for footpath segments in commercial areas in Bangkok and Jakarta with street vendor
activities, by incorporating qualitative and quantitative variables. The proposed model
included pedestrians’ perceptions, pedestrian traffic and pedestrians’ behaviour toward
vendors’ activities as the variables affecting PLOS. Pedestrians’ perceptions were considered
as dependant variables in this model, based on interviews with pedestrians. Regression
analysis was used to produce a PLOS model as a function of pedestrians’ perceptions of
comfort and problems caused by vendors’ activities, pedestrian volume and the number of
pedestrians who interacted with street vendors, along with pedestrian traffic data. This model
appears useful for evaluating footpaths in commercial districts in developing cities (Hidayat,
Choocharukul & Kishi, 2011).
Asadi-Shekari et al. (2013) suggested three key approaches underpinning the abovementioned
research. They categorised these analysis methods into three groups based on the methods
used in calculating the PLOS from previous studies. These groups include point systems,
regression and simulation, and they showed that each method has advantages and
disadvantages, as shown in Figure 6.
Figure 6 The analysis methods of PLOS (Source: Asadi-Shekari et al. 2013)
25
2.5.5 Conceptually framing previous research approaches
This narrative review of past literature explains data collection methods and data analysis
methods and identifies a wide variety of factors which must be considered in order to
adequately capture the pedestrian walking experience, thus to calculate PLOS. Table 2
summarises the most prominent LOS studies. It is apparent from the existing methods
summarized in this table that the majority of studies used surveys, questionnaires, direct
observation and video techniques to collect data and measure factors for LOS models.
26
Table 2 Summary of prominent studies for pedestrian level of service
Reliability
Author of Country of Data Unit of Analysis
Target area Evaluatio Model type of data Factors / Variables Types of factors
study study collection analysis method
n type collection
Linear Space per person (density); speed of movement
Fruin (1971) U.S.A. Survey Segment Area LOS Capacity based model n/a Design and traffic
relationship and traffic flow.
Available space per pedestrian, flow rates,
Pushkarev & Linear
U.S.A. Survey Segment Area LOS Capacity based model n/a pedestrians/ min/ft, pedestrians/min/m, sq Design and traffic
Zupan, 1975 relationship
ft/pedestrian.
Density in pedestrians per square metre
Tanaboriboon Portable video (ped/m’); Area module in square metre per
Linear
and Thailand camera and Segment Footpath LOS Capacity based model n/a pedestrian (m2/ped); Speed in metres/min Design and traffic
relationship
Guyano (1989) survey (m/min); Flow in pedestrians per metre width
per minute (ped/m/min).
A videotape Linear speed‐
Polus, Schofer,
Israel recorder and a Segment Footpath density LOS Capacity based model n/a Characteristics of pedestrian flow. Design and traffic
& Ushpiz, 1983
digital clock regression
Roadway Safety, security, comfort and convenience,
Munich and
Sarkar (1993) Observation Segment Street Point system PLOS Characteristics Based n/a continuity, system coherence, and Environmental
Rome
Model attractiveness.
Roadway Safety, security, comfort and convenience,
Khisty (1994) U.S.A. Observation Segment Street Point system PLOS characteristics based n/a continuity, system coherence, and Environmental
model attractiveness.
Primary variables (walk area width-volume,
walk area-outside lane buffer, outside lane
traffic volume, outside lane motor vehicle
Mozer, 1994 U.S.A. Worksheet Segment Street Point system LOS Combination model n/a Design and traffic
speed); secondary factors (walk area
penetrations, heavy vehicle volumes and
intersection wait-time).
A pedestrian facility type (dominant facility
type, footpath width, off-street/parallel
alternative facility); conflicts ( number of
driveways, reduced turn-conflict
Roadway
Corridor implementations, and crossing widths, Design, traffic and
Dixon (1996) U.S.A. Audit Segment Point system PLOS characteristics based n/a
roadway pedestrian signal delay times, speed of traffic environmental
model
and medians present); travel demand
management; amenities (benches, lighting,
buffers and shade trees); maintenance; motor
vehicle LOS; and multimodal provision A-F.
Enclosure/definition, complexity of path
Jaskiewicz network, building articulation, complexity of
U.S.A. Observation Segment Roadway Point system PLOS Capacity based model n/a Environmental
(2000) spaces, transparency, buffer, shade trees,
overhangs/awnings/varied roof lines, and
27
physical components/condition.
Security; handicapped access; median openings;
Miller, Bigelow, Simulation Roadway Environmental and
signalisation parameters; pedestrian warning
and U.S.A. Survey Crosswalk Intersection and PLOS characteristics based n/a traffic
devices; crosswalks; speed limit of road
Garber (2000) point system model
crossing.
Design factors (path width, surface quality,
obstructions, crossing opportunities and support
facilities); location factors (connectivity, path Design, traffic and
Gallin (2001) Australia Audit Segment Roadway Point system PLOS Combination model n/a
environment and potential for vehicle conflict); environmental
user factors (pedestrian volume, mix of path
users and personal security).
Lateral separation elements between pedestrians
and motor vehicle traffic (presence of footpath,
width of footpath, buffers between footpath and
Stepwise motor vehicle travel lanes, presence of barriers
Landis et al. multi-variable within the buffer area, presence of on-street
U.S.A. Observations Segment Roadway PLOS Combination model n/a Design and traffic
(2001) regression parking, width of outside travel lane, presence
analysis and width of shoulder or bike lane); motor
vehicle traffic volume; effect of (motor vehicle)
speed; motor vehicle mix (i.e., percentage of
trucks); driveway access frequency and volume.
Physical (adequate walkway, continuous
Service footpath, comfortable for vulnerable users,
levels walkway free of impediments, comfortable
Roadway
and walking surface, seating, protection from
Sarkar (2003) U.S.A. Audit Segment Walkway Point system characteristics based n/a Environmental
quality extreme weather conditions); psychological
model
levels (ability to maintain desired walking speed,
ability to participate in various pedestrian
activities); physiological (noise and pollution).
Design, traffic and
environmental
Muraleetharan, Width and lateral separation (buffer area,
Survey (operational and
Adachi, Linear Overall shoulder, on-street parking; obstructions; flow
Japan (field Segment Footpath Combination model n/a geometric
Hagiwara, and relationship PLOS rate; number of bicycle passing and opposing
measurement) characteristics)
Kagaya (2004) events.
Muraleetharan, Stepwise Geometric,

Space at corners; crossing facilities; turning
Adachi, multiple operational; traffic
Japan Field survey Crosswalk Intersection PLOS Combination model n/a vehicles; delay at signals; and pedestrian-
Hagiwara, and regression and delay
bicycle interaction.
Kagaya (2005) model characteristics
Pearson
Footpath correlation
Petritsch et al. Field Total width of crossings at conflict locations
U.S.A. Segment and analyses and PLOS Combination model n/a Design and traffic
(2006) observation and average volume on the adjacent roadway.
intersection stepwise
regression
28
Width of dividing strip between motor vehicle
lanes and non-motorized vehicle lane; width of
Intercept
non-motorized vehicle lane; width of
Bian et al. survey Step-wise
China Segment Footpath PLOS Combination model n/a landscaped strip (or device strip); width of Design and traffic
(2007) (questionnaire regression
footpath; motor vehicle volume; non-motorized
survey)
vehicle volume; pedestrian flow rate and
obstruction.
Type of walking area; type of roadside
development or landscape; motor vehicles per
hour in both directions; average motor vehicle
speed (km/h); passed pedestrians per hour on
Respondents,
Cumulative nearest roadside; bicycles and mopeds per hour
video clips, Design, traffic and
Jensen (2007) Denmark Segment Roadway logic B&PLOS Combination model n/a in both directions; width of buffer area between
and environmental
regression walking area and drive lane (m); parked motor
questionnaire
vehicle on road; median dummy; width of
walking area; total width of walking area and
nearest drive lane; drive lane dummy, four or
more drive lanes; tree dummy.
Design, traffic and
Survey and Segment Roadway Footpath width; pedestrian flow rate; outside environmental
NCHRP (2008) U.S.A. video and and Regression PLOS Combination model n/a lane; shoulder width; on-street parking; barrier; (operational and
clip crosswalk intersection buffer width; traffic lanes; traffic speed.. geometric
characteristics)
Width of outside lane; width of shoulder or
bicycle lane; on-street parking effect coefficient;
per cent of segment with occupied on-street
parking; buffer area barrier coefficient; buffer
Stepwise
FDOT (2009) U.S.A. Observation Segment Roadway PLOS Combination model n/a width; footpath presence coefficient; width of Design and traffic
regression
footpath; volume of motorised vehicles in the
peak 15-minute period; total number of
directional through lanes; average running
speed of motorised vehicles traffic.
Right turning vehicle; bicycle volume from the
street parallel to the crosswalk during pedestrian
Intercept
green time; permissive left-turning vehicles and
Bian et al. survey Step-wise
China Crosswalk Intersection PLOS Combination model n/a bicycles approaching from the street parallel to Traffic and delay
(2009) (questionnaire regression
the crosswalk; through bicycle volume on the
survey)
street parallel to the crosswalk; pedestrians’
delay.
Slower traffic speed; buffer and barriers (kerb
Asadi-Shekari and furnishing zone); fewer traffic lane; shorter
Malaysia & Design, traffic and
and Zaly Observation Segment Roadway Point system PLOS Combination model n/a crossing distance (kerb extension); mid-block
Singapore environmental
Shah (2011) crossing; social space (cafe´); landscape and
tree; facility (fire hydrant); furniture (trash
29
receptacle); footpath pavement; marking.
.(crosswalk); pedestrian refuge and median;
corner island; footpath on both sides; advance
stop bar; width of footpath; driveway; lighting;
signing and bollards.
Interview &
Hidayat, Multiple Roadway
Thailand & questionnaire Safety/security, comfort, vendors attraction,
Choocharukul, Segment footpath regression PLOS characteristics based n/a Environmental
Indonesia surveys with movement easiness, and footpath condition.
& Kishi (2011) analysis model
pedestrian
30
2.6 Evaluation of environmental audit instruments
Researchers have developed instruments and protocols which are best evaluated through
direct observation in measuring the actual physical environment (Moudon & Lee, 2003).
Researchers have used audit tools to collect key data on physical features (e.g., street trees,
footpath width) that could not be obtained through other methods such as GIS databases or
aerial photos (Brownson, Hoehner, Day, Forsyth & Sallis, 2009).
The use of audit tools is considered a systematic approach to use in observing the physical
environment, including the factors affecting physical activity (e.g., street pattern, number and
quality of public spaces, footpath quality). Some proposed audit tools were not only
established for research purposes, but they have also been used to support local decision
making by community members. These tools have been designed with less detail than those
designed for research purposes and it is probable that they have not been tested for reliability
(Moudon & Lee, 2003).
Audit tools consist of a paper form, usually with close ended questions (e.g., check boxes,
Likert scales), and sometimes with open ended questions or comments. The collection of data
using an audit tool method needs an in-person observation walk or drive through a
neighbourhood, park, or trail (Ewing et al., 2006). The typical unit of observation which is
used for assessing neighbourhood or community features is a ‘street segment’. Segments are
typically sampled because it is not feasible to audit entire neighbourhoods or areas, with some
exceptions (Lee, Booth, Reese-Smith, Regan & Howard, 2005) and sampling is either random
or purposeful (Brownson et al., 2009).
2.6.1 Previous attempts to develop and test audit tools

Caughy, O’Campo and Patterson (2001) developed an assessment tool to collect information
on neighbourhood characteristics relevant to the health and well-being of families and
children. A brief observational method for urban neighbourhoods was described by collecting
data from a socio-economically diverse group of urban neighbourhoods. A reliability test for
this instrument was conducted by a drive through of two socio-economically diverse
neighbourhoods. Based on the results of the pilot test, the instrument was modified depending
on the teams' recommendations in terms of the addition and elimination of items. The final
instrument included 45 objective items that covered the type and condition of buildings, the
condition of grounds and undeveloped spaces, indications of block uniformity / territoriality,
type of street, presence of graffiti / litter, neighbourhood resources, and the presence and
31
activities of people. Furthermore, this instrument was used to determine the presence and
types of non-residential land uses (Caughy, O’Campo & Patterson, 2001).
The built environment in several London wards was assessed by Weich et al. (2001) using the
Built Environment Site Survey Checklist (BESSC). The BESSC consisted of 27 items of
various aspects including height and age of housing, number of dwellings and type of access,
provision of gardens, use of public space, amount of derelict land, security and accessibility
of local shops and amenities. Eleven housing areas were chosen to describe the built
environment by two raters, and only the third area in the ward was selected to measure inter-
rater reliability using kappa coefficients. Only fifteen of twenty-five (60%) of the items were
found to have moderate-to-good agreement (kappa ≥0.5) for fifteen out of twenty-five items
(Weich et al., 2001).
Pikora et al. (2002) described the first reported attempt to develop a comprehensive
instrument to measure the physical environment factors related to walking and cycling in
local neighbourhoods. The development and assessment of reliability for the Systematic
Pedestrian and Cycling Environmental Scan instrument (known as SPACES) was created
after developing a framework of the potential environmental influences on walking and
cycling based on published evidence and policy literature, interviews with experts and a
Delphi study. The framework involved four features, including function, safety, aesthetic and
destination; as well as the hypothesized factors that contribute to each of these features of the
environment (Pikora, Giles-Corti, Bull, Jamrozik & Donovan, 2003). This instrument was
used to collect data over a total of 1,987 kilometres of roads in metropolitan Perth, Western
Australia. Inter and intra-rater reliability was used to assess the reliability of the instrument
using 16 observers to collect the data. Both the intra- and inter-rater reliability of the 37 items
included in the SPACES instrument were generally high, and the observers reported that the
audit instrument was easy to use (Pikora et al., 2002).
Craig, Brownson, Cragg and Dunn (2002) presented an audit instrument known as
‘Neighborhood Active Living Potential’ to measure the physical design of neighbourhoods
which influenced behaviour. This instrument allowed the potential effects of income,
university education, poverty, and degree of urbanization to be separated from the relationship
between walking to work and neighbourhood design characteristics. Their study used
Canadian data from 27 neighbourhood observations with information on walking to work
from the 1996 census (Craig, Brownson, Cragg & Dunn, 2002). The potential reliability and
32
validity of the neighbourhood active living instrument was investigated using the principles of
ecometrics. Eight observers were recruited and were given a map of a predetermined walking
route constructed through joining ten randomly selected street blocks, in order to rate 112
neighbourhoods using an 18-item observation grid. The three dimensions of the
neighbourhood active-living potential instrument have good reliability and convergent
validity (Gauvin et al., 2005).
There is a need for measurement indicators and assessment instruments which evaluate the
suitability of walking and bicycling conditions in community environments in order to create
more active community members. Therefore, Emery, Crump and Bors (2003) determined the
reliability and validity of two instruments to assess the suitability of footpaths for walking and
roads for bicycling. Community members and professionals would use these instruments to
collect data on the walking and bicycling environment. Two data collectors used 31 road
segments to collect data using walking and bicycling suitability assessment instruments. A 7-
point Likert scale was used by three transportation experts to assess walking and bicycling
conditions within the same segments. The reliability of each assessment instrument and the
reliability of the Likert response system were calculated based on intra-class correlations; one
of the very few studies to check for this. Some variables of the walking and bicycling
suitability assessment instruments had lower reliability and validity than was ideal (Emery,
Crump & Bors, 2003).
Two versions of an audit instrument were created by Brownson, Ramirez, Hoehner and Cook
(2003) to measure key indicators of activity friendly communities. Audits were used to
evaluate the walkability and bikeability of environments through understanding the
relationships between street-scale environments and rates of physical activity. Both an
‘analytic’ (with Likert-scale and ordinal-response choices) and a ‘checklist’ (with
dichotomous response choices) were used to audit 147 street segments. The same segments
were re-audited to assess inter-rater reliability (Brownson et al., 2004). These audit tools
focus on the physical activity rate and scores are based on personal judgments.
A brief one-page instrument called the Physical Activity Resource Assessment (PARA)
instrument was developed by Lee, Booth, Reese-Smith, Regan and Howard (2005) to describe
and assess the type, features, amenities, quality and incivilities of a variety of physical activity
resources. The PARA instrument was used to assess all publicly available physical activity
resources in thirteen urban neighbourhoods (Lee et al., 2005).
33
Cunningham, Michael, Farquhar and Lapidus (2005) discussed the importance of developing
reliable measures of the physical features of an environment which influence physical activity
among seniors. They developed and tested the Senior Walking Environmental Audit Tool
(SWEAT) to evaluate the effects of the physical environment on walking among seniors in
Portland, Oregon. SWEAT was developed based on a review of urban planning and health
literature and feedback from urban planning and health research professionals. It was used to
evaluate a total of 355 neighbourhood segments. The reliability of this instrument was tested
using inter-rater reliability for thirty-six neighbourhood segments which were randomly
selected. Raters found agreement for 67% of the items on most items included in SWEAT
(Cunningham, Michael, Farquhar & Lapidus, 2005).
Williams et al. (2005) proposed an objective tool to assess footpath maintenance based on a
collaborative partnership between researchers and community members for the success of this
study. They determined the most important characteristics of footpaths based on the content
of tools located on the internet, and reviewed the literature and input from the research team
and partnering community. Characteristics involved in audit tools included levelness,
blockages, cleanliness, and surface conditions. The audit tool was tested using inter-rater
reliability by three raters for each item on the footpath assessment tool, which found adequate
inter-rater reliability (Williams et al., 2005).
Dannenberg, Cramer and Gibson (2005) evaluated walkability using an audit instrument to
measure elements affecting walkability. Nine elements of walkability were involved in
developing the audit tool, including pedestrian facilities, pedestrian-vehicle conflicts,
crosswalks, route maintenance, footpath width, roadway buffer, universal accessibility,
aesthetics and shade. The walkability was assessed using a five-point scale from 20 to 39
(poor), to 40 to 69 (fair), to 70 to 100 (good) (Dannenberg, Cramer & Gibson, 2005).
Day, Boarnet, Alfonzo and Forsyth (2006) designed an audit tool called the Irvine Minnesota
Inventory (IMI) to systematically and reliably measure the characteristics of the built
environment which are potentially linked to active living. This audit tool was developed based
on a literature review, focus group interviews, a panel of experts, field testing in 27 settings
and pilot testing. The Irvine Minnesota Inventory consists of 162 items that cover four
domains including accessibility (62 items), pleasurability (56 items), perceived safety related
to traffic (31 items), and perceived safety related to crime (15 items) (Day, Boarnet, Alfonzo
& Forsyth, 2006). The reliability of IMI was tested on a wide range of study areas by
34
calculating inter-rater reliability using a percentage agreement between observers (Boarnet,
Day, Alfonzo, Forsyth & Oakes, 2006). There was no score for intermediate conditions in this
point system method, and it also had limited input from a broader range of pedestrians.
Ewing, Clemente, Handy, Winston and Brownson (2006) developed operational definitions
and a field survey instrument for subtler urban design qualities believed to be related to
walkability. Measurement instruments were developed to evaluate five of nine urban design
qualities including imageability, visual enclosure, human scale, transparency, and complexity.
A field survey instrument included urban design qualities, physical features relating to urban
design qualities, procedures for field observations and data collection and scoring procedures
for translating objectively measured physical features into urban design quality scores (see
Figure 7). Although this instrument included a wide range of factors relating to urban design
qualities, it did not include the factors affecting walking facilities (Ewing et al., 2006).
Figure 7 Conceptual framework of study by Ewing et al. (2006)
Troped et al. (2006) developed the Path Environment Audit Tool (PEAT) to measure the trail
characteristics which affect use. The PEAT items consisted of three major categories
including 11 trail design features, 16 amenities, and 7 maintenance/aesthetics, some with sub-
items. The inter-observer reliability of the PEAT was measured by two independent observers
who audited 185 trail segments at 6 Massachusetts facilities. The validity of a sub-set of
PEAT items was assessed using global positioning system (GPS) derived measures as a ‘gold
standard’. All the kappa (k) statistics, observed agreement and intra-class correlation
coefficients (ICCs) were used to assess inter-observer reliability and validity (Troped et al.,
2006).
35
The observation method was used by Suminski, Petosa and Stevens (2006) for observing
physical activity along residential footpaths and streets using the Block Walk Method (BWM)
in 12 U.S. census block groups. The BWM was developed so that it would be a suitable
instrument, providing reliable information regarding the number of individuals walking,
biking, and running on residential streets and footpaths. The BWM had high reliability with
regard to determining types of physical activity, the number of individuals performing the
physical activity, and the geographical location where the physical activity was performed.
The level of agreement between independent observers of BWM was 99.8% for activity type
(Suminski, Petosa & Stevens, 2006). Therefore, Suminski, Petosa and Stevens (2006)
developed the BWM to use in evaluating the relationships between the environmental
characteristics of footpaths and streets and the physical activities (the number of individuals
walking, bicycling, and jogging) on urban residential footpaths and streets (Suminski,
Heinrich, Poston, Hyder & Pyle, 2008).
Hoehner, Ivy, Brennan Ramirez, Handy and Brownson (2007) tested the reliability of the
Active Neighborhood Checklist which was used for the systematic assessment of factors in
the physical and social environment that hinder or facilitate physical activity. Fifteen public
health researchers and seven community stakeholders conducted audits in sixty-four street
segments in St. Louis and southeastern Missouri in the U.S. These segments were selected
among diverse areas that varied with respect to socioeconomic levels, urbanization, and land
use. Inter-rater reliability was measured for the items in each section of the checklist (land use
characteristics, footpaths, shoulders and bike lanes, street characteristics, and quality of the
environment for a pedestrian) using observed agreement and the kappa statistic. The checklist
audit tool demonstrated strong reliability with both the observed agreement for 57 evaluated
items and the kappa statistic (Hoehner, Ivy, Brennan Ramirez, Handy & Brownson, 2007).
Clifton, Smith and Rodriguez (2007) presented the development of the audit methodology and
described consistent, reliable, and efficient methods to collect information about the walking
environment. They developed and tested a comprehensive environmental audit methodology
called the Pedestrian Environmental Data Scan (PEDS) to measure the critical elements of the
walking environment. Raters audited 955 segments to measure the overall rater reliability
(intra and inter) of the audit; the variation in reliability by urban context; and the influence of
various modes of implementation of the audit on reliability of the data collected. They used
kappa statistics, percent agreement and the concordance correlation coefficient to test
36
reliability (Clifton, Smith & Rodriguez, 2007). Some modifications were conducted by the
authors for PEDS as a result of the relatively low reliability of a number of measures in the
PEDS.
Badland, Opit, Witten, Kearns and Mavoa (2010) adopted the same methodological principles
which were used by Pikora et al. (2002) to measure built environment attributes associated
with walking and cycling in New Zealand. They developed the New Zealand Systematic
Pedestrian and Cycling Environment Scan (NZ-SPACES) that was used in 48 street segments
drawn from four neighbourhoods in Auckland, New Zealand. Audits were conducted to assess
the level of agreement between the physical (criterion) and virtual (test) using two-way mixed
model intra-class correlation coefficients (ICC), and the virtual audits took less time than the
physical audits (χ=115.3 min (virtual), χ=148.5 min (physical)) (Badland et al., 2010).
Frackelton et al. (2013) developed the PIN3 Neighborhood Audit Instrument as a primary
source to collect data for assessing urban and rural neighbourhood characteristics. The PIN3
consists of 43 items related to neighbourhood characteristics including 7 key aspects as a
priori constructs. Three of seven aspects were derived from the study by Caughy et al. (2001),
consisting of physical incivilities, territoriality and social space. The four remaining aspects
were derived from a study by Pikora et al. (2002) and included functionality, safety and
aesthetics and destinations that directly related to physical activity in and around the
neighbourhood. Sixteen raters were trained in 2005 and 2006 to audit and collect data from
10,770 road segments. Reliability testing was conducted until raters reached an acceptable
level of agreement. The PIN3 was adapted to collect data using handheld electronic devices
(Frackelton et al., 2013).
Some aspects of the neighbourhood built environment affecting residents’ physical activity
were discussed by Su et al. (2014). They developed the China Urban Built Environment Scan
Tool (CUBEST) consisting of 41 items to assess the physical activity related built
environment in six neighbourhoods in Hangzhou, and potentially in other cities in China, and
conduct necessary reliability and validity tests. A review of existing reliable auditing tools
was adopted to create the main CUBEST items. The key factors which were included to
create CUBEST items involved residential density, street connectivity, accessibility,
walkability (segments with footpaths), bikeability (segments with bike lanes) and aesthetics.
Inter-rater reliability was measured by a pair of raters independently using 309 street
segments, while a total of 208 street segments (104 segments for each rater) were assessed to
37
measure intra-rater reliability. The results of the reliability test showed that the inter-rater
reliability for most items was above 0.8 while the intra-rater reliability for most items was
above 0.4, and was lower than corresponding inter-rater reliability (Su et al., 2014).
2.6.2 Conceptually framing previous audit tool studies

Table 3 summarizes the key characteristics of the prominent audit instruments. It is apparent
from this table that there is significant variation in the detail with which various researchers
measure various features, from a few items to many items measuring the various distinct
characteristics of footpaths, streets, built environments, trails, parks, urban design qualities,
and so on. Many audit tools shown in this table were not empirically tested, as they were
often too complex and required knowledge of urban planning concepts. The examination of
the environment activity relationship is limited because often only a few specific
environmental aspects were assessed. Although several attempts to develop walkability audits
have been undertaken by planners and engineers by including a number of features that affect
the entire roadway corridor environment, there has been no consensus on the degree of
importance each of these features has for pedestrians, using statistically reliable methods.
Also, these features have not yet been statistically tested or widely applied. It is apparent from
the existing instruments summarized in Table 3 that there is a need for a footpath pedestrian
level of service audit tool that is reliable and valid, one which is simple to implement in the
collection and measurement of factors affecting PLOS in commercial areas. But what factors
should be included remains in dispute and is likely culturally contingent.
38
Table 3 Summary of the key characteristics of the most used audit tools
Areas
Instrument Sources of items Reliability Reliability Raters covered by Time
Author Region Purpose Items Ranges covered
name included test results no. reliability required
test
The type and condition
of buildings, the
Building upon condition of grounds
previous research, and undeveloped
Measuring
extensive literature spaces, indications of
neighbourhood
Systematic Social review, pilot test by Inter-rater 1135 hundred block uniformity /
Caughy et characteristics relevant Average reliability 5–10 min
U.S.A. Observation conducting drive- 45 reliability 2 block in 57 territoriality, type of
al. (2001) to the health and well- = 0.87 per block
(paper form) throughs in two neighbourhoods street, presence of
being of families and
socioeconomically graffiti / litter,
children
diverse neighbourhood
neighbourhoods. resources, and
presence and activities
of people.
Height and age of
housing, number of
Three national dwellings and type of
Built Environment
housing surveys, 11 access, provision of
Site Survey
Weich et The built environment texts on urban Inter-rater kappa ≥0.5 for 15 housing areas gardens, use of public
U.K. Checklist 27 2 3 Hours
al. (2001) assessment design and reliability out of 25 items across the two space, amount of
(BESSC)
published studies in wards derelict land, security
(paper form)
this field and accessibility of
local shops and
amenities.
27 segments
4 for
(140 m length) The functional, safety,
Systematic Published evidence Inter-rater Intra- and inter- inter-
Measure the physical for inter-rater aesthetic and
Pedestrian and and policy reliability rater reliability of rater
environment factors reliability destination; the
Pikora et Cycling literature, + the items ≥ 70% & 40 min per 2
AU related to walking and 67 hypothesised factors
al. (2002) Environmental interviews with Intra-rater 16 for km
cycling in local 83 segments that contribute to each
Scan (SPACES) experts and a reliability kappa for 70% of intra-
neighbourhoods (140 m length) of these features of the
(paper form) Delphi study items ≥0.4 rater
for intra-rater environment.
reliability
Canadian data from
27 neighbourhood
Neighborhood Measure the physical 12 hours per
observations with 4 pairs Ten street Activity friendliness,
Craig et al. Active Living design of Inter-rater Average reliability one
Canada information on 18 of raters blocks of 112 safety, density of
(2002) Potential (NALP) neighbourhoods reliability ≥ 79% neighbourhoo
walking to work n=8 neighbourhoods destinations. d
(paper form) influencing behaviour
from the 1996
census
39
Traffic volume, traffic
Walking speed, footpath
Suitability Assessing the 17 road condition and buffer
Emery et Intra-class
U.S.A. Assessment Form suitability of footpaths Not reported 11 γ = 0.79 2 segments width, surface quality, Not reported
al. (2003) correlations
(WABSA) for walking (2 miles length) and supportive
(paper form) amenities.
Bicycling
Suitability Assessing the 31 road General road factors,
Emery et Intra-class
U.S.A. Assessment Form suitability of roads for Not reported 27 γ = 0.90 2 segments pavement factors, and Not reported
al. (2003) correlations
(WABSA) bicycling (2 miles length) location factors.
(paper form)
Transportation
Measuring key
An extensive environment, land-use
indicators of activity-
literature review Reliability environment,
Brownson Analytic & friendly communities
and 3-stage Delphi 27 for Inter-rater measured as % of 6 in 147 segments recreational facilities, 10.6 min per
et al. U.S.A. Checklist Audit to evaluate the
process with experts each reliability items with 75% pairs (147 m length) physical disorder segment
(2003) Instruments walkability and
from a variety of agreement (aesthetics), signage,
bikeability of
backgrounds. and social
environments
environment.
An exhaustive list
of terms identified
Physical Activity by pilot testing, Overall characteristics,
Assessing all publicly Reliability tests of 10 min per
Resource internet and the number, type and
Lee et al. available physical Reliability a 10% overlap Not 17 medium-
U.S.A. Assessment telephone book 43 quality of features and
(2005) activity resources in 17 tests showed good reported neighbourhoods sized
(PARA) searches and amenities it possessed,
urban neighbourhoods reliability (rs > .77) resource
(paper form) windshield surveys and overall incivilities.
by trained assessors.
36 Functionality (types of
Reviewed urban
Neighbourhood buildings, footpaths,
Senior Walking planning and health Acceptable
Evaluating the effects Inter-rater segments (a and street life),
Cunningha Environmental literature and agreement for 67%
of the physical reliability and Not segment is a personal safety (e.g., 17 min per
m et al. U.S.A. Audit Tool provided feedback 188 of the items,
environment on agreement reported section of road adequate lighting) and segment
(2005) (SWEAT) by urban planning kappa (Κ) =0.6-
walking among seniors scores between traffic safety (e.g.,
(paper form) and health research 0.75
two speed)), aesthetics,
professionals.
intersections) destination.
The content of tools
located on the
Assess footpath
Objective internet as well as
maintenance based on 1190 footpath Levelness, blockages, 8–12 min
Williams et Sidewalk reviewing the Inter-rater Overall kappa
U.S.A. a partnership between 5 3 blocks (99.4 cleanliness, and per footpath
al. (2005) Assessment Tool literature and input reliability coefficient=0.61
researchers and miles) surface condition. block
(paper form) from research team
community members
and partnering
community
40
8 settings (15-
In California
20 segments per 10 min per
76.8% of 2 for
A literature review, Inter-rater setting for Accessibility, segment
Measuring the variables> 80% Californ
Irvine Minnesota focus group reliability and California pleasurability, (California)
characteristics of the agreement ia place
Day et al. Inventory (IMI) interviews, a panel a % of & perceived safety &
U.S.A. built environment that 162 & &
(2006) (paper form and a of experts, and field agreement 891 segments related to traffic, and 20 min per
are potentially linked in Minnesota 3 for
tablet PC survey) testing in 27 between for Minnesota perceived safety segment
to active living 99.2% of Minneso
settings raters (a segment= related to crime. (Minnesota)
variables> 80% ta place
two facing
agreement
block fronts)
Measurement Measuring the urban Recruiting an expert (ICCs) ratio of Imageability, visual
Ewing et Instrument design qualities panel; shooting Inter-rater between groups Not enclosure, human 20 min per
U.S.A. 27 Not reported
al. (2006) for Urban Design believed to be related video clips of reliability variance to total reported scale, transparency, segment
Qualities (paper to walkability streetscapes variance = 0.4–0.6 and complexity.
Average
agreement = 86.9%
Bedimo-Rung Assessing park A series of meetings 15 Features, conditions,
Bedimo- Inter-rater &
Assessment Tools characteristics that may with an expert panel teams of Eight target access, esthetics,
Rung et al. U.S.A. 181 reliability & average Not reported
(BRAT-DO) be related to physical using the observer areas safety of park
(2006) validity agreement with
(paper form) activity Delphi method pairs characteristics.
gold standard =
78.7%
15/16 primary
amenity items k
≥0.49,
The Path A brief review of
all had observed
Environment various literature,
agreement k≥81%. Trail design features,
Audit Tool Measuring the trail brief intercept Inter-rater 45 intersecting
Troped et design, amenity; amenities, and
U.S.A. (PEAT) characteristics that survey with adult 34 reliability & 2 road and 185 Not reported
al. (2006) maintenance maintenance/aesthetics
(paper form & affect use users, input from validity trail segments
& , some with sub-items.
Tablet PC) transdisciplinary
agreement between
team
the first observer
and GPS-derived
items≥ 77%
12 U.S. census
Observing physical
The average levels block groups Physical activity type,
Suminski Block Walk activity along The two-stage Inter-observer 10 min per
U.S.A. 3 of agreement = 2 (five 305-m location, and number
el at. (200 Method (BWM) residential footpaths expert panel review reliability segment
97.7% segments/each of individuals.
and streets using
block groups)
Land use
Systematic assessment Kappa(k)=0.68
The observed characteristics,
Active of factors in the &
agreement of footpaths, shoulders
Hoehner et Neighborhood physical and social average observed 64 street 11.7 min per
U.S.A. Not reported 57 items and 15+7 and bike lanes, street
al. (2007) Checklist environment that agreement for 57 segments segment
kappa characteristics, and
(paper form) hinder or facilitate evaluated items =
statistic quality of the
physical activity 0.87
environment for a
41
pedestrian.
Pedestrian Land uses, vehicle-
Measuring the critical Kappa(k)= 0.4 for
Clifton et Environmental Based on previous Intra and inter 955 segments pedestrian interactions, 3–5 min per
U.S.A. elements of the 31 most items 2
al. (2007) Data Scan (PEDS) efforts reliability (400 ft. length) safety and appeal, segment
walking environment
(paper/tablet PC) subjective assessment.
An extensive
Environmental Assessing the extent to Kappa(k)= 0.11 for
literature review on
supports for which environmental walking surface,
environmental Inter-rater 112 Walking surface,
Spivock et people with buoys promoting active Kappa(k)= 0.66 for 4 pairs
Canada buoys in 3 reliability neighbourhoods signage, and Not reported
al. (2007) disabilities living among signage, of raters
architecture, urban (census tracts) surroundings.
(paper form) individuals with Kappa(k)= 0.32 for
planning, disability
disabilities surroundings
studies, and health
New Zealand
Adopting the same Reliability 48 street
Systematic Measuring built Walking function,
methodological between the segments
Badland el New Pedestrian and environment attributes Not cycling function,
principles which 51 two measures ICCs ≥0.70 (defined as 148.5 min
at. (2010) Zealand Cycling associated with reported safety, aesthetics, and
were used by Pikora (physical and intersection to
Environment Scan walking and cycling destinations.
et al. (2002) virtual audits) intersection
(NZ-SPACES)
464 road
segments
Neighborhood (segment as
Physical incivilities,
Audit Instrument Study by Caughy et Spearman length
Frackelton Assessing urban and Reliability by territoriality, social
(PIN3) al. (2001), the study correlation between two
et al. U.S.A. rural neighbourhood 48 level of 16 space, functionality, Not reported
(handheld by Pikora et al coefficients (r = intersections or
(2013) characteristics agreement safety, aesthetics,
electronic (2002) 0.96) between an
destinations.
devices) intersection to a
cul-da-sac or
dead end road)
309 street
segments for 2 Residential density,
scan for inter – street connectivity,
Assessing the physical Kappa(k)≥0.8 for
rater accessibility,
China Urban Built activity-related built Adopting a review Inter and inter –rater
Su et al. & walkability (segments 7.4 min per
China Environment Scan environment in of existing reliable 41 intra-rater & 2
(2014) 208 street with footpaths), segment
Tool (CUBEST) Chinese auditing tools reliability ICCs ≥ 0.4 for
segments (104 bikeability (segments
neighbourhoods intra-rater
segments for with bike lanes) and
each rater) for aesthetic.
intra- rater
42
2.7 Research gap
The literature review reveals that the content of LOS has been significantly studied in extant
academic research. Since the 1990s, most investigations have paid more attention to LOS
research in order to meet the needs of vehicle oriented design without considering the needs
of those who actually use the streets. However, this approach has changed as a result of the
increasing growth of the concept of sustainable development, and because of the growing
awareness of the need to promote sustainability and livability through more attention being
paid to issues relating to pedestrians.
It is critical that the existing walking facilities are evaluated in order to develop a walking
environment for pedestrians by measuring the factors affecting level of service for pedestrians
on footpaths. Calculating PLOS is no easy issue, with behaviour influenced by complex
factors. Some of these factors, especially the qualitative factors, need complex methods of
measurement, analysis and interpretation. Overall, the methodology to evaluate the pedestrian
level of service mainly involves determining the various factors which influence pedestrians
in terms of perceived safety and comfort.
Through a review of the previous methods used to calculate PLOS, it is clear that the majority
of the current evaluation methods have key drawbacks. Some of these studies have dealt with
an assessment of pedestrians, in the same way as an assessment of motor vehicles, based on
automobile LOS. In addition, previous studies’ selections of the factors influencing PLOS
were not conducted based on determined criteria or were not based on pedestrians'
perceptions. However, researchers’ personal estimates played a prominent role in the factors
selected. Further, the methods of data collection and methods of measurement of the factors
affecting PLOS have not been based on a systematic and reliable approach. Also, the
determining of the factors affecting PLOS has not taken into consideration the factors that
may affect the disabled, children or the elderly.
A method is needed to objectively quantify pedestrians’ perception of safety and comfort in
the footpath environments in commercial areas. Therefore, the challenge is to develop an
assessment model that uses reliable methods to identify and measure factors affecting PLOS
for a full range of pedestrians.
Regarding this gap and adopting an empirical and systematic manner, this research attempts
to develop a reliable model which reflects the perceived safety and/or comfort of pedestrians
along footpath segments. This model can then be used to objectively collect and measure
43
factors influencing pedestrian LOS. Hence, a new weighted model based on the rating of a
panel of experts and pedestrian perceptions to evaluate PLOS will be produced.
2.8 Research objective

To aid in the development of reliable measures of footpaths characteristics, this research seeks
to achieve the following objective:
• To identify the various factors affecting PLOS on different pedestrian facilities using a
systematic method.
• To select the appropriate factors for the Australian context that mainly influence the
PLOS on footpaths in commercial areas using systematic and reliable methods.
• To rate the factors impacting walking conditions in terms of the degree of their
importance in influencing PLOS, based on the opinions of a panel of experts and
pedestrians' perceptions.
• To develop appropriate methodology which leads to a new PLOS model appropriate
for the Australian context.
• To develop a comprehensive PLOS model using an audit tool method to measure the
overall PLOS value.
• To define the limits of different PLOS (on a scale of A to F) for footpaths in
commercial areas under affected factors.
• To prepare a guideline manual for using a proposed pedestrian level of service audit
tool.
2.9 Research question
To fulfill the aims of this research, it is necessary that the following primary and procedural
questions be addressed:
Primary question 1
What were the previous approaches and methodologies and factors used for evaluating the
44
Primary question 2
What do experts think should be included in PLOS models appropriate for Australian cities?
environment?
Primary question 3
What do pedestrians think is the importance level of a wide range of factors affecting PLOS?
developed and how can differing age groups and genders be explored?
Primary question 4
How could a more optimal pedestrian level of service (PLOS) audit tool for footpaths in
commercial streets be developed which would be appropriate for Australian cities?
45
46
3 Research approach and methodology
3.1 Research paradigm
Guba defined a paradigm as a set of key beliefs which guides and organizes action (Guba,
1990, p. 17). In order to provide a systematic basis for a research study, it is important that a
worldview or a paradigm of a study be specified (Creswell & Clark, 2007). The development
of a comprehensive reliable audit tool which could efficiently work to evaluate pedestrian
LOS on footpaths in commercial streets is the ultimate task of this investigation. Choice of
paradigm for this study influences the research design.
Social science and philosophy scholars and researchers have suggested diverse sets of
worldviews or paradigms, such as are shown in studies by Guba and Lincoln (1994) and
Teddlie and Tashakkori (2009) (Guba & Lincoln, 1994; Teddlie & Tashakkori, 2009).
Although the research paradigm in transport planning has often concentrated on the
generalisation and prediction of travel behaviour based on positivist paradigms (Banister,
2002), often, paradigms for social research have been developed and have been given
increasing attention. Alternative paradigms include post-positivism, constructivism, critical
theory, participation, pragmatism and critical realism (Creswell, 2013; Lincoln, Lynham &
Guba, 2011; Næss & Jensen, 2002).
Sayer (2000) explained that “the particular choices of research methods should always depend
on the nature of the object of study and what one wants to learn about it” (Sayer, 2000, p. 19).
As Creswell (2013) stated, the important task of a researcher is to understand the research
problem, the ways used to solve it and how the research arrived at the result (Creswell, 2013,
p. 11). It seems difficult to use only one paradigm as was outlined by Guba and Lincoln
(1994, p. 109) in answering the research questions. Therefore, this research adopts a
pragmatic research approach supported by post-positivist and critical realist paradigms (see
Table 4).
47
Table 4 The basic beliefs of received and alternative inquiry paradigms (Guba & Lincoln, 1994).
Item Positivism Postpositivism Critical theory et al. Constructivism
Historical realism – virtual
Critical realism- ‘real’
Naïve realism- ‘real’ reality shaped by social, Relativism – local and
reality but only imperfectly
Ontology reality but able to be
and probabilistically able
political, cultural, economic, specific constructed
apprehended. ethnic, and gender values; realities.
to be apprehended.
crystallized over time.
Modified dualist/
Dualist/objectivist; objectivist; critical Transactional/subjectivist; Transactional/subjectivist
Epistemology findings true. tradition/community; value-mediated findings. ; created findings.
findings probably true.
Modified
Experimental/manipulati experimental/manipulative;
ve; verification of critical multiplism;
Methodology hypotheses; chiefly falsification of hypotheses;
Dialogic/dialectical. Hermeneutical/dialectical.
quantitative methods. may include qualitative
methods.
Scholars’ use of the dualism of quantitative and qualitative methods has previously been
given more attention and more heavily emphasised as a result of increasing methodological
debates in social science research, and in particular, in the fields of transport planning
(Banister, 2002; Denzin & Lincoln, 2005; Guba & Lincoln, 1994). The use of mixed methods
is supported by a pragmatic approach, mixing both quantitative and qualitative research
methods (Creswell, 2013). Pragmatism focuses on using suitable methods, in accordance with
the problem, aims and questions of the research, regardless of the challenges and limitations
inherent in qualitative or quantitative methodologies.
In dealing with the topic of PLOS, it is clear that the investigation addresses a complex issue,
as it is dealing with the various aspects of a space. Stakeholders use a street in many different
ways, so these spaces are considered to be multidimensional and complex. This study deals
with the different kinds of factors (in relation to both the environment and the users) affecting
the PLOS. It aims to answer the question of ‘How could a more optimal pedestrian level of
service (PLOS) measurement for footpaths in commercial streets be appropriate for
Australian cities?’ These factors might be quantitatively measurable (such as physical factors
of environment) or qualitatively measurable (such as users' perceptions of factors’
importance). Therefore, it seems to be difficult to stick solely to a post-positivist paradigm
when aspects of understanding the qualitative factors with users will require digging through
many layers of meaning, demanding that the researcher enter into dialogue with those users.
As Creswell and Clark (2007) argued, “The use of quantitative and qualitative approaches in
combination provides a better understanding of research problems than either approach alone”
(Creswell & Clark, 2007, p. 5).
48
3.2 Case study approach
This research effort is located in one case study location. There are a number of advantages in
using case studies as follows:
1. Case study research is broadly appropriate to cover contextual and other complex
conditions giving way to the particular unit studied (Susam-Sarajeva, 2009; Yin,
2012).
2. Case study methods enable a researcher to closely examine the data within a specific
context (Yin, 1994, p. 21; Zainal, 2007).
3. Case studies allow for both quantitative and qualitative analyses of the data through
using different approaches, including intrinsic, instrumental and collective methods
(Zainal, 2007)
4. Case studies help to produce detailed qualitative accounts which are not only used to
explore or describe the data in real-life environment, but also used to explain the
complexities of real-life situations (Zainal, 2007).
5. By using case study methods, a researcher can go beyond the quantitative statistical
results and understand behavioural conditions through actors’ perspectives (Zainal,
2007).
Despite these advantages, case studies have received criticisms, which may be considered as
disadvantages, as follows:
1. Case studies are often criticized as a result of lack of accuracy (Zainal, 2007).
2. Case studies are accused of allowing the use of equivocal evidence or biased views to
influence the direction of the findings and conclusions (Yin, 1994, p. 21).
3. Case studies are accused of an inability to generalize scientifically as a result of the
use of a small number of subjects, and some such studies are conducted using only one
subject (Yin, 1994, p. 21).
4. Case studies are often described as being too long and difficult to conduct and
producing a massive amount of documentation (Yin, 1994, p. 21).
3.2.1 Description of the study area

Brisbane City’s central business district was selected. Brisbane is the capital city of
Queensland - Australia’s second largest state by area. In addition to being the third most
populous city in Australia, Brisbane City Councils area is, geographically, the largest of
49
Australia’s capital cities. Brisbane inner city has a total area of 29.5 square kilometres. A
considerable increase in population will generate pressure on the local government to provide
a degree of comfort, safety and ease of navigation for users, particularly pedestrians. The
Brisbane City Council and the Queensland Government are trying to encourage active
transportation. For example, they jointly funded the construction of a cycle centre as part of
King George Square station. This is Australia’s first major end-of-trip facility for cyclists and
pedestrians using active transport to the CBD (Brisbane City Council, 2014). Also, the
footpaths in the Brisbane CBD can be a good example of footpaths in commercial streets,
making them suitable for applying our methodology for developing PLOS in Australia.
Consequently, we can consider that the commercial streets in Brisbane CBD are
representative of the commercial streets of other Australian cities.
For this reason, Brisbane’s central business district (CBD) has been chosen as the main case
study as it provides a useful location in which to answer the research questions of this study.
This location has an abundance of pedestrians, and is almost exclusively comprised of
commercial streets.
50
Figure 8 Map of Brisbane Inner City (source: Google map).
3.3 Research design

This study is established to consist of a set of tasks and steps that show the structure of the
research (see Table 6). In the conclusion, these tasks lead to the achievement of the key aim of
this research. These tasks and steps are stated as follows.
Task 1: A study of literature and context

This study began with a study of the literature and the context. The task involved an
examination of peer reviewed literature; a review of path design guidelines, the previous
PLOS model, as well as previous audit tools. A brief review of parks and recreation,
landscape architecture, transportation, and planning literature, as well as the previous LOS
and audit tool studies was conducted to identify the factors that have been documented to
influence use.
It is very important to review previous studies to show key methodologies and practices that
have been used to evaluate street conditions for pedestrians over the past four decades. The
majority of the previous studies and the current prominent attempts were discussed in this task
and were considered as references for this study’s efforts. The advantages and disadvantages
of PLOS models were considered to determine a reasonable PLOS model to measure the
factors affecting pedestrians with respect to safety, comfort, and mobility.
In addition to the narrative review in Chapter 2, a systematic quantitative literature review
was conducted adopting the Preferred Reporting Items for Systematic Review
51
Recommendations (PRISMA) protocol (shown in Chapter 4). This involved systematically
searching and categorising the relevant literature. PLOS models were compiled from the peer
reviewed literature, the internet, and from experts in the fields of transportation and health.
Our review focused on studies related to quality of the walking environment, and explicitly, to
PLOS. We sought to categorise and tabulate what researchers have suggested are the most
important approaches, methodologies and factors in generating PLOS outputs. Factors derived
from existing PLOS models were classified into five groups, including geometric
characteristics, obstructions and interruptions, aesthetics and amenities, pedestrian traffic and
adjacent traffic. This helps by selecting relevant items derived from these studies with factors
which are probably affecting PLOS (with respect to sense of comfort, safety and mobility)
and identifying where new items are needed. Another reason to review the existing audit
instruments used to capture walking conditions, was to inventory and evaluate individual
measures of environmental factors used in these models.
The key aim of this task is to identify which factors previous researcher have identified which
influence PLOS in different pedestrian facilities. Based on this literature review, a summary
table is created, which involves factors that were either empirically studied or received
prominent attention in the previous studies, as shown in Table 6.
Task 2: Data collection and analysis

There were two steps in this task as follows:
1. Delphi Process
Methodologically, potential factors affecting pedestrian level of service were identified in
terms of safety, comfort and mobility to be used for development of the proposed PLOS
models in this step. Two rounds of the Delphi process were applied in this step, first an online
survey, then a related workshop (a ‘walkshop’) with a set of experts who are members and
associates of the Pedestrian and Bicycle Transport Institute of Australasia (PedBikeTrans), a
key professionals’ association that is especially active in Brisbane. The first round was an
online survey. This round involved experts from a variety of backgrounds, including urban
planning, urban design, engineering, transport planning, and pedestrian planning. The final
sample of 36 participants who completed the first round came from local and state
governments, universities and private consultancies, along with one recently retired expert.
The draft of the hierarchical framework obtained from the literature review was organised in
the form of a questionnaire. The experts were asked to rate the importance of each factor
52
using a five-point scale, ranging from 1 ‘not at all important’ to 5 ‘extremely important’, and
to suggest new factors, previously omitted, that they believed were important.
The second round was a walkshop with 20 experts from the round one sample. This round
comprised of two phases with participants undertaking a walking course and then
participating in a further questionnaire. The walking course was designed to subject the
participants to a variety of factors affecting PLOS. Participants were asked to complete a
PLOS rating sheet independently as they walked each footpath segment along eight streets in
the Brisbane central business district (CBD). Whilst walking they were encouraged to use
their imagination and accumulated experience to consider which factors have more effect and
make more contribution to PLOS. In the second phase, the experts were provided with
another sheet with the revised list of the factors affecting PLOS obtained from the results of
the first round (the online survey). The experts were then asked to make any further changes
to the list of factors and to rate them using a five-point Likert scale ranging from ‘extremely
important’ to ‘not at all important’ for each factor based on their perceptions obtained during
the walking course, and the results supplied from round one. Based on the results of this step,
the final list of the factors affecting PLOS on footpaths in commercial streets was developed
to be used to obtain pedestrians’ perceptions regarding the degree of importance of each
factor. This is shown at length in Chapter 5.
2. Pedestrian intercept survey
An on–street intercept questionnaire survey was then employed to gather pedestrians'
perceptions of the factors affecting PLOS on these same commercial street segments within
the Brisbane CBD. An online questionnaire method using a tablet PC was adopted. The
questionnaire was designed to measure pedestrians’ perceptions in five main categories with
33 sub-items or factors, which were the outputs from the Delphi process. The first part of the
questionnaire obtained respondents’ characteristics including age, gender, education, trip
purpose and the frequency of use of the footpaths. The second part assessed the importance
and satisfaction of each existing footpath facility for the pedestrians. Images were used to
help define the factor in easy-to-understand terms, and trained researchers were able to further
describe each factor to respondents who were uncertain. Respondents were asked to rate both
the satisfaction (how satisfied they are) and the importance (how important to them is good
performance) using a five-point Likert scale for each of the factors in relation to these
footpaths in the study area. The scale used for the level of satisfaction was from 1 = very
53
dissatisfied, to 5 = very satisfied and the level of importance was from 1 = not at all important
to 5 = extremely important. Footpaths with different characteristics in the CBD area can be
evaluated with the questionnaire because it involves a wide range of footpath aspects and can
be used in various contexts. The outputs of this survey were used to generate the relative
importance and satisfaction for each aspect of the footpath environment. 312 respondents
aged 16 and over were randomly selected to ensure that the perception of the wide range of
people toward the importance and satisfaction of factors affecting PLOS on footpaths were
captured.
The key advantage of this step is that the perceptions which were obtained were based on
walking experiences in real situations. Furthermore, the scores which were given by the
pedestrians were considered as the basis to produce a new model for PLOS based on
pedestrians' perceptions. Also, the results of this task helped to develop a weighted
assessment sheet for calculating pedestrian level of service using an audit method.
Task 3: Development of PLOS models

This task includes a set of steps as follows:
1. Weighting all the factors affecting PLOS
As mentioned earlier, an on–street intercept questionnaire survey was employed to gather
pedestrians' perceptions regarding the importance and satisfaction rating of the factors
affecting PLOS in the commercial street segments within the Brisbane CBD. The identified
factors had different impacts on the overall PLOS conditions. Pedestrians’ perceptions at this
stage reflect a level of importance and the priority of improvement for each aspect of the
footpath environment that contributes to increase or decrease pedestrian level of service. First
the relative importance or weight for each footpath element was generated using the following
equation:
Where, RW is the realistic weight of importance level for each factor as shown in Table 7, IR
is the importance rating of each factor, IRs is the sum of importance rating for all factors, i is
the factor number (33 factors in this study), and j is the number of participants (312
participants in this study).
54
2. Developing the proposed PLOS model based on a satisfaction approach
The first proposed model for assessing PLOS was adopted using the satisfaction approach
developed by Khisty (1994), which used a point system method. The level of pedestrian
satisfaction for each aspect of the walking environment represented by the factors was
identified as a measurement of performance. The percentages of respondents who rated
performance of the same 33 factors as satisfied or strongly satisfied were calculated. The
PLOS model was built by allocating a point to each factor affecting PLOS using a point
system method based on the percentage of satisfaction rating. Factors were selected to meet
the need for detailed information about the footpath environment with measurement of PLOS.
Each factor was built to assess individual elements of the walking environment with respect to
PLOS as well as overall PLOS value. Factors included various aspects of footpath features at
the microscale level.
3. Developing the proposed PLOS model based on disgruntlement
The second proposed PLOS model was adopted using a disgruntlement approach, which also
used a point system method. The disgruntlement measure can be obtained by calculating both
the performance, or satisfaction and importance ratings, for each factor. First we calculated
the relative importance of a wide range of factors affecting PLOS by calculating percentages
of respondents who rated each factor as important or extremely important. Second, the
percentages of respondents who rated performance of the same 33 factors as satisfied or
strongly satisfied were calculated. The percentage of pedestrians who were disgruntled for
each factor was derived by cross-tabulating the percentage of respondents who were
dissatisfied or strongly dissatisfied and who rated each aspect as important or extremely
important. Then, the modified 'gap analysis' method was applied to identify the aspects of the
walking environment most in need of improvement. This method provides a visual inspection
of how dissatisfaction and importance are distributed amongst the various PLOS measures
and then identifies the weaknesses in level of service for any aspect of the targeted footpaths.
The PLOS model was built by allocating a point for each factor affecting PLOS using the
point system method based on the percentage of pedestrian disgruntlement. This model is
more realistic in identifying PLOS value than the satisfaction model because it takes into
consideration importance levels in identifying the satisfaction percentage of the factors
affecting PLOS.
55
4. Determining pedestrian level of service scale
It is necessary to determine the actual pedestrian LOS grade on a specific scale to be assigned
to targeted footpaths. On the basis of LOS grades adopted in the HCM scale, a PLOS range is
allocated to one of the grades from A to F based on the point system method in this study.
These grades are related to the current situation of footpaths regarding the adequacy of
footpaths’ aspects in providing a high level of comfort, safety and mobility, and are
intrinsically related to all the 33 factors affecting PLOS. The range for pedestrian level of
service values might be between 1 and 6 points for both these two kinds of PLOS models,
based on the current situation by the value of the obtained points.
Task 4: Developing PLOS Audit Tool model

This task included a set of steps as follows:
1. Weighting all the factors affecting PLOS
The same weighting results from the previous task (Task 3, step 1) were used to build the
weighting vales for the factors affecting PLOS.
2. Developing the proposed PLOS audit tool (PLOSAT)

This proposed model for assessing PLOS adopted an audit tool approach using a point system
method. The pedestrian level of service model was developed to provide a systematic, reliable
and empirical method to measure factors affecting PLOS, and efficiently evaluate PLOS for
footpaths along commercial streets. This model consists of two parts. The first part uses a
fieldwork form of the factors affecting PLOS called the Pedestrian Level of Service Audit
Tool (PLOSAT). PLOSAT consists of 33 items in these five categories. The thirty-three items
are distributed into 7 items belonging to the geometric characteristics category, 4 items
belonging to the obstruction and interruption category, 14 items belonging to the aesthetic and
amenity category, 4 items belonging to the pedestrian traffic category, and 3 items belonging
to the adjacent traffic category. All items were selected based on the previous stages of this
study, that is the literature review and the Delphi process. The proposed tool adequately
evaluates all these factors in a CBD area to ensure the provision of a high level of service for
pedestrians. Some factors or items included in this part are evaluated by assessing the current
condition for each item compared with standard requirements in the pedestrian guidelines.
Other items are evaluated based on eligibility criteria developed according to the previous
iterations of PLOS models. In this context, the subjective assessment of raters may play a role
56
in the assessment process, relating to how they measure each item in the field, based on
eligibility criteria. For this reason, the reliability test of this tool is absolutely necessary in
developing a robust version of PLOSAT. A four-point scale was allocated for each item,
starting from point zero representing the worst condition scenario (i.e., very
unsafe/uncomfortable), to point 3 representing the best condition scenario (i.e., very
safe/comfortable). The allocated scale reflects how each item was built in the real
environment and makes the comparison according to the standard requirement, meeting the
needs of pedestrians for safety, comfort and mobility.
The second part is a weighted assessment form which was used to calculate the overall
pedestrian level of service (PLOS). This part contains all 33 factors included within the
fieldwork form.
This new audit tool can be used to measure overall pedestrian walking conditions on footpaths
in commercial areas in a systematic and empirical manner.
3. Determining pedestrian level of service scale
The same method used for developing the PLOS scale in the previous task (Task 3) were
adopted to create the PLOS scale for this model. But the definition of pedestrian level of
service grade in this model depends on the maximum and minimum weighted assessment
score that a footpath could receive. The range of maximum and minimum weighted
assessment score might be between 0 and 300, based on current value of the obtained points.
4. Preparing a guideline manual for using proposed PLOSAT
A guideline manual was prepared for raters to provide specific information relating to the
evaluation of pedestrian level of service (PLOS) on footpaths in commercial streets in
Brisbane using the Pedestrian Level of Service Audit Tool (PLOSAT). This document
provides instructions on how to organise and conduct an audit, and the standards for
pedestrian facilities with supporting information on how to use the forms when conducting an
onsite audit.
5. Conducting reliability tests for the proposed PLOSAT
This step includes a reliability test for a new PLOS model. The reliability or validity of the
new audit tool was assessed using inter- and intra-rater reliability. Raters used a model of the
new audit tool and the guidelines manual which was developed in the last steps. A training
session was held with 6 raters on August 2019 for 2 hours to provide a good understanding of
each of the audit items for all raters. Raters were recruited from the transport PhD cohort at
57
Cities Research Institute/ Griffith University who have experience in the transport planning
field. First, a guideline manual was distributed to participants one week before holding the
training workshop, and participants were asked to read it carefully. The workshop comprised
of three main parts: i) an overview of the audit, ii) a detailed description of each audit
element, and iii) determination the targeted footpath segments.
A comprehensive review of the audit tool item by item was conducted. Then, each item and
selection option was reviewed in depth with photographic examples. Raters were given a
chance to ask questions and discuss any problems or concerns regarding the audit process
during the training workshop. Then, the locations of the targeted audit segments to be
conducted in the field were determined, and maps identifying the required segments were
provided to each participant. The training workshop and protocol materials provided detailed
background information so that raters, once trained, could conduct the audit reliably.
Inter/intra-rater reliability was estimated from data that were observed on footpath segments
by six raters. A measure of agreement between raters leads to measuring the degree of
reliability of the PLOSAT. Furthermore, kappa (κ) statistics were used to overcome any level
of agreement that could occur by chance (Viera & Garrett, 2005).
The key aim of this task was to obtain the degree of agreement among participants. It gives a
score of how much homogeneity, or consensus, there is in the ratings given in order to assess
the degree of reliability for a new model, as well as being a test of usability.
3.4 Research methodology

A summary of the key tasks, the research questions being addressed, the methods and related
result chapters are presented in Table 6.
58
Table 5 Research tasks
Tasks Research questions Activities Methods Chapters Publications

RQ1: What were the previous approaches,
methodologies and factors used for evaluating
the pedestrian level of service for different
pedestrian facilities?
 What were the previous approaches for
calculating the level of service for different
pedestrian facilities?
 What were the previous factors for inclusion in
the level of service calculation for different Chapter 4:  Transport
Identify the previous methodologies and
pedestrian facilities? A systematic What are the most Research Board
approaches. Assess which factors may
A study of  What the methods were used regarding evaluation quantitative important factors for Annual Meeting
influence PLOS in different pedestrian
literature of PLOS? literature review pedestrian level of 2017
facilities and factors that were either
 Where is there research on PLOS, where was the using PRISMA service estimation? A  Transport
empirically studied or received prominent
research undertaken, who has undertaken the technique systematic review of the Research Record
research and where was it published? attention in the previous studies.
literature. 2018
 Which study have applied a reliability test for
data collection methods?
 What evaluation techniques are there for
assessing PLOS?
 What methods have been used to identify factors
affecting PLOS?
 What most important factors were revealed in the
literature for inclusion in assessing PLOS?
Step 1 includes two rounds:
RQ2: What do experts think should be  Online survey with 36 experts to rate the
included in PLOS models appropriate for importance of each factor affecting
Chapter 5:
Australian cities? PLOS derived from the first task using a An earlier version of
Identifying the most
 What are the possible factors for inclusion in the five-point scale, ranging from 1 ‘not at the paper presented in
Data important factors for
pedestrian level of service calculation for all important’ to 5 ‘extremely Chapter 5 was first
collection Delphi technique pedestrian level of
footpaths in commercial streets? important’, and to suggest new factors, presented at the
and analysis of two rounds service measurement:
 How can an extensive list of the factors identified previously omitted, that they believed Transport Research
Step 1 results from a Delphi
be refined to select those factors that should be were important. Board Annual
study in Brisbane,
included in an examination of PLOS appropriate  Based on the results of this step, the Meeting 2018.
for Australian cities?? Australia.
final list of factors affecting PLOS on
footpaths in commercial streets was
developed.
59
Step 2:
 The questionnaire was designed to
measure pedestrians’ perceptions in five
main categories with 33 sub-items or
RQ3: What do pedestrians think is the factors that were the outputs of the
Chapter 6:
importance level of a wide range of factors Delphi process. An earlier version of
An earlier title
affecting PLOS?  312 participants were asked to rate both the paper presented in
Data “Investigating
Collection  How can the importance level of a wide range of the satisfaction (how satisfied) and the
Pedestrian relationship between the
Chapter 6 was first
factors affecting PLOS be examined? importance (how important good presented at the
and analysis intercept survey importance of the factors
 How can this relationship, which may differ for performance is) using a five-point Likert World Conference on
Step 2 different ages and genders of pedestrians in
affecting PLOS and level
scale for each of the factors in relation to Transport Research
different walking environments, be explored? of satisfaction on
these paths in the study area. (WCTRs) 2019.
footpaths”
 The outputs of this survey were used to
generate the relative importance and
satisfaction for each aspect of the
footpath environment.
 A model for assessing PLOS on

Chapter 5:
footpaths in commercial streets was
 How can the PLOS model be appropriately Identifying the most A revised version of
developed by adopting a satisfaction
created? A satisfaction important factors for the paper presented in
approach and point system method.
 How can an overall PLOS and individual PLOS approach and pedestrian level of Chapter 5 was then
 PLOS grade that ranged from A to F
for each factor based on the level of pedestrian point system service measurement: submitted to the
satisfaction for the factors affecting PLOS in a based on point system method was
method results from a Delphi Journal of Case
specific environment be calculated? determined.
Development study in Brisbane, Studies on Transport
of PLOS Australia. Policy.
models
based on  A model for assessing PLOS on
previous footpaths in commercial streets was Chapter 6:
 How can a PLOS model be appropriately In 2019, a revised
approaches developed by adopting a disgruntlement Final title “Developing
created? Gap analysis, a version of the paper
approach and point system method. Pedestrian Level of
 How can a new PLOS model based on a disgruntlement presented in Chapter
 A PLOS grade that ranged from A to F Service (PLOS) model
pedestrian disgruntlement approach be developed approach and 6 was then submitted
and how can these differing age groups and based on point system method was for footpaths in
point system to the Journal of
genders be explored?? determined. commercial streets using
method Travel Behaviour and
pedestrian
Society.
disgruntlement”.
Developing RQ4: How could a more optimal pedestrian  Weighting all the factors affecting Audit tool Chapter 7:
PLOS audit level of service (PLOS) audit tool for PLOS. method, point Developing and testing
60
tool model footpaths in commercial streets appropriate for  Developing the proposed PLOS Audit system method pedestrian level of
Australian cities be developed? tool (PLOSAT). and inter/intra- service (PLOS) audit tool
 How can a PLOS model be appropriately created  Determining pedestrian level of service rater reliability for footpaths in
for the Australian context? scale. commercial streets.
 How can that draft PLOS audit tool be  Prepare a guideline manual for using the
developed?
proposed PLOSAT.
 Is that tool reliable?
 Conducting reliability tests for the
 How would the proposed audit tool be used? proposed PLOSAT
 Is that tool easy to use and practical in the field?
61
3.5 Ethics
Ethical clearance was obtained for this research. The aspects of this research have been
designed to be compatible with the ethical framework of the National Statement for Ethical
Conduct in Research Involving Humans (National Statement on Ethical Conduct in Human
Research 2015), which involves the ethical responsibilities of academic researchers in relation
to research participants. Moreover, Griffith University’s ethical requirements have been
considered as shown in the Griffith University Code for the Responsible Conduct of Research
(2007). Subsequently, users’ survey data will take into consideration the privacy,
confidentiality of the participants, and all participants will be asked to sign informed consent
after the key aims of this research and data usage are explained. This research is designed to
use an online survey with a panel of experts, a walkshop with experts and intercept user
surveys with pedestrians, hence ethical approval for the research was sought and requested
from Griffith University after completing all required documents.
62
4 What are the most important factors for pedestrian level of service? A
systematic review of the literature
An earlier version of the paper presented in Chapter 4 was first presented as:
Level-of-Service Estimation? A Systematic Review of the Literature. Paper presented
at the 97th Annual Meeting of Transportation Research Board (TRB), January 7-11,
2018, Washington, D.C., USA.
 Raad, N., & Burke, M. (2018). What Are the Most Important Factors for Pedestrian
Level-of-Service Estimation? A Systematic Review of the Literature. Transportation
Research Record.
 This paper was also won a 2018 Most Outstanding Paper Award from the
committee of the TRB. It was published in the journal of Transportation Research
Record on 15/03/2018
4.1 Research summary

This chapter summarises the results of Task 1. The key aim of this task was to systematically
identify the various approaches and methodologies previously used for evaluating the
pedestrian level of service in different pedestrian facilities. Another aim was to identify the
various factors that influence PLOS in different pedestrian facilities using systematic
methods. This aim focused on studies related to the quality of the walking environment, and
explicitly to PLOS. A systematic quantitative literature review was adopted to address the
objectives of this study. The systematic review includes a set of procedures based on the
protocol developed by the Preferred Reporting Items for Systematic Review
Recommendations (PRISMA). These procedures include developing the research questions,
search terms used in electronic databases, a search of electronic databases, the search strategy
and initial selection, content assessment and eligibility criteria, data collection process, cross-
check and synthesis of results. The review grouped and categorised what almost all previously
published research efforts have suggested are the most important factors in generating PLOS
outputs. Fifty-eight of the 617 studies surveyed were determined as PLOS models directly
related to different pedestrian facilities, including streets, footpaths, intersections and
midblock crossings over the past four decades. A key output of the review was a list of all
63
factors affecting PLOS derived from the literature, with an indication of how frequently those
factors were used across the many studies. The PLOS factors were themed into five
categories. These categories were: 1) geometric characteristics; 2) obstructions and
interruptions; 3) aesthetics and amenities; 4) pedestrian traffic; and, 5) adjacent traffic. The
result shows that the factors most used were, in order: footpath width; obstructions to
pedestrian flow; motor vehicle speeds and volumes; shoulder widths; and, buffers such as on-
street parking. The results obtained in this current study can help practitioners to gain a clear
image and outstanding understanding by categorising and tabulating what the researchers
have suggested are the most important approaches, methodologies and factors in generating
PLOS outputs.
4.2 Research questions addressed in this paper and contributions
4.2.1 Research question
Primary questions:
RQ1: What were the previous approaches and methodologies and factors used for evaluating
the pedestrian level of service in different pedestrian facilities?
Sub-Questions:
64
4.2.2 Summary of contributions
1. The methodological contributions of this study:
This study is the first to systematically review the extant literature on the previous pedestrian
level of services (PLOS) models to provide a clear image of detailed aspects of PLOS. This
includes identifying the main approaches being used to determine PLOS and, for the first
time, identifying and theming the most common factors used in assessments. There has been
no systematic attempt to synthesise this research and provide an overarching perspective. Our
review surveyed PLOS models from the peer reviewed literature using a protocol developed
by the Preferred Reporting Items for Systematic Review Recommendations (PRISMA).
2. Practical and applied contributions of the study:

 The outcomes of this research can help practitioners and academics to understand the
different options of the main approaches being used to determine PLOS and the most
common factors used in assessments.
 The outcomes can assist practitioners in moving towards more robust and standardised
approaches and to identify problems with PLOS measurement that require further
investigation.
 The outcomes can assist practitioners in identifying gaps in PLOS and suggesting
options for future enquiries, in order to create more reliable and comprehensive PLOS
measures.
 The outcomes highlighted the need to improve methods for selecting, including and
measuring factors in PLOS calculations, using techniques that go well beyond the
individual researcher to include the input of a representative sample of relevant
experts.
 The study highlighted the current lack in PLOS studies. The study also highlights the
need for a wider research base, taking into account more regions of the world, and in
particular in Australia, as the majority of the literature is from North America.
 Presenting the disaggregation results by facility type, particularly for different types of
pedestrian facilities (streets, footpaths, intersections and midblock crossings) provides
an easy and clear understanding of various PLOS approaches in different pedestrian
facilities.
65
4.3 Statement of contribution to the co-authored published paper
This chapter includes a co-authored paper that summarises the results of Task 1. The
bibliographic details of the co-authored paper, including all authors, are:
Level-of-Service Estimation? A Systematic Review of the Literature.
The authors listed below have certified that:
1. They meet the criteria for authorship in that they have participated in the conception,
execution, or interpretation, of at least that part of the publication in their field of
expertise;
2. They take public responsibility for their part of the publication, except for the
responsible author who accepts overall responsibility for the publication;
3. There are no other authors of the publication according to these criteria;
4. They agree to the use of the publication in the student’s thesis and its publication on
the Griffith University database consistent with any limitations set by publisher
requirements.
Contributors Statement of contribution

Nowar Raad  Developed the study conception and design.
 Designed the data collection process including the development of a
systematic review protocol, research questions, search terms used in
electronic databases, electronic databases searched, search strategy and
initial selection and content assessment and eligibility criteria.
 Compiled, scanned and analysed the data.
 Synthesised and interpreted the results.
 Conceived, planned and wrote the first draft of the manuscript.
 Managed the academic editorial process for the manuscript.
 Submitted and revised the manuscript at the suggestion of the conference
reviewers.
 Prepared presentation slides to be presented at the conference.
 Managed journal submission as corresponding author.
 Responsible for revisions made at the suggestion of the journal reviewers
and subsequently resubmitted the paper.
Matthew Burke  Provided assistance in the theoretical framing of the research.
 Provided assistance in the synthesis and interpretation of the results.
 Provided suggestions in response to journal reviewers.
 Provided editing revisions in the initial and the reviewed manuscript.
 Provided assistance in preparing the presentation slides and presented the
paper at the conference.
66
(Signed) _________________________________ (Date) 26/08/2019
Student and Corresponding Author: Nowar Raad
(Signed) __________________________________ (Date) 26/08/2019
Principal Supervisor: A/Prof Matthew Burke
This chapter is an exact copy of the paper referred to above
67
What are the most important factors for pedestrian level of service
estimation? A systematic review of the literature
Abstract
Pedestrian environments are becoming more important in an urbanising world where walking
is increasingly being encouraged. Engineers and planners determine a link’s performance
using measures codified as levels-of-service, which provide guidance on acceptable or
desirable standards. A range of approaches are used in determining pedestrian levels-of-
service (PLOS), incorporating a wide variety of factors and with much debate as to what
should or should not be used. There has been no systematic attempt to synthesise this research
and provide an over-arching perspective. Our review surveyed PLOS models from the peer
reviewed literature using a systematic quantitative literature review method based on the
protocol developed by the Preferred Reporting Items for Systematic Review
Recommendations (PRISMA). PLOS models are increasingly being developed. Twenty-two
of the 58 studies surveyed were produced since 2013. Earlier work adapted approaches used
to determine automobile LOS to pedestrians. Later approaches use a much wider range of
factors but with very little consistency across the studies surveyed. Collectively these factors
can be grouped in themes of comfort, safety and mobility. The most used factors were, in
order: footpath width; obstructions to pedestrian flow; motor vehicle speeds and volumes;
shoulder widths; and, buffers such as on-street parking. However, many of the factors being
included have not been empirically studied and almost none of the tools in use have been
tested for such matters as inter-rater reliability.
Keywords: Pedestrian level of service, street evaluation, pedestrian facilities, systematic

quantitative review.
68
1. Introduction
Standards and performance measures often guide the planning and design of pedestrian
environments. Pedestrian level-of-service (PLOS) is one of the more important measures
used. For almost fifty years, transport engineers, urban designers and planners have developed
and applied PLOS to determine how a pedestrian environment may function in terms of a
pedestrian’s comfort, safety and amenity. As will be shown, there is dispute as to how PLOS
should be calculated, and what factors should be considered. This paper systematically
reviews the literature from 1971 to 2016 to help identify the main approaches being used to
determine PLOS and identify and theme for the first time the most common factors used in
assessments. This should assist practitioners in moving towards more robust and standardised
approaches and to identify problems with PLOS measurement that require further
investigation.
Levels-of-service can be defined as an assessment of the quality of service or an overall
measure of current conditions on any piece of transport infrastructure, covering aspects such
as the facilities themselves, the contextual situation, equipment levels and ancillary
infrastructures. The level-of-service is defined as “a quantitative stratification of a
performance measure or measures that represent quality of service” (L. Kang et al., 2013).
Gallin defined PLOS as “an overall measure of walking condition on route, path and facility”
and suggested PLOS reflects users' perceptions in terms of sense of comfort and safety
(Gallin, 2001).
Street designs should meet the needs of all users, including pedestrians, especially in urban
areas (Zohreh Asadi-Shekari et al., 2013). The US Highway Capacity Manual (HCM)
developed a six-level scale for pedestrian LOS that ranges from A, which represents the ‘best’
level-of-service, through to F, which represents the ‘worst’ (Gallin, 2001). The HCM
provides possible methods for evaluating level of service in specific conditions (Ensley,
2012). However, whilst vehicle LOS was quite advanced as far back as the 1980s,
underpinned by measures of volume and capacity, the means to produce PLOS were poorly
developed by comparison (Landis et al., 2001). Conceptually, to measure PLOS generally
requires quite different and more complex approaches than for vehicle LOS. Ideally, one
might wish to capture users’ perceptions across different groups of pedestrians, including the
young, seniors and the less physically able, given footpaths must be designed to create
accessibility for the majority of users. Footpaths should also have attractive features for those
69
individuals who may choose to walk rather than those with no other mode options (Boodlal,
2004).
Increasingly it is qualitative factors, along with traditional volume and capacity factors, that
are important in the measurement of PLOS. This recognises that the pedestrian experience
cannot be seen only as physical movement from point A to point B (Jaskiewicz, 2000). PLOS
is therefore being tied to agendas to make streets and roadway environments more liveable
(Landis et al., 2001).
PLOS dates back to the 1970s and Fruin's model, based on footpath capacity and pedestrian
volume, which ignored other built environment concerns (Landis et al., 2001). In subsequent
years many researchers adapted Fruin’s approach with some additions or modifications
(Davis & Braaksma, 1987; Lam et al., 1995; Polus et al., 1983; Pushkarev & Zupan, 1975;
Tanaboriboon & Guyano, 1989). Others suggested alternative approaches were based on
alternative measures of pedestrian comfort, safety and convenience (Henson, 2000; Khisty,
1994; Miller et al., 2000; Sarkar, 1993, 2003). Many of these focused on the micro-level
conditions of comfort for pedestrians (i.e., shade, lighting, greenery) but used only a limited
set of factors to measure PLOS. A third set of researchers used a combination of Fruin-style
volume/capacity and other qualitative street characteristics, attempting to gain the best of both
approaches, as well as applying PLOS to a wider set of facility types (Landis et al., 2001;
Mozer, 1994). Researchers such as Gallin (Gallin, 2001) went a little further, seeking to group
and theme measures into categories, such as physical characteristics, location factors and user
factors (Muraleetharan, Adachi, Uchida, Hagiwara & Kagaya, 2004). However, very few
researchers used field studies to calibrate user perceptions, with notable exceptions including
the work of Petritsch et al. (Petritsch et al., 2006). Asadi-Shekari & Zaly-Shah tried to group
factors derived from more than 20 other studies and guidelines in an attempt at being more
robust (Asadi-Shekari & Zaly-Shah, 2011). Other researchers, such as Hidayat, Choocharukul
and Kishi (Hidayat et al., 2011), tried to develop PLOS measures more suited to developing
countries, where street vendor activities proliferate.
As will be shown, across all these different approaches and methods there is a very wide
variety in the choice of factors suggested for PLOS evaluations. To help practitioners
understand the different options and to try to identify, group and theme the more commonly
suggested factors, a systematic review was performed. The review gathered PLOS methods
from the literature and categorised them by input data, model type, analysis method, factors
70
included and use of reliability testing. The next section of this paper outlines the methodology
used to perform the systematic review and the analysis; following are the results; and, finally
a discussion is provided, identifying gaps and suggesting options for future enquiry to create
more reliable and comprehensive PLOS measures.
2. Approach and method

A systematic quantitative literature review was conducted, systematically searching and
categorising the relevant literature. We sought to categorise and tabulate what researchers
have suggested are the most important factors in generating PLOS outputs.
2.1. Systematic review protocol
We used the Preferred Reporting Items for Systematic Review Recommendations (PRISMA)
protocol (Moher, Liberati, Tetzlaff, Altman & The, 2009) following the steps outlined in
Pickering and Byrne (2014) (Pickering & Byrne, 2014), as used successfully in a range of
past studies (Ballantyne & Pickering, 2015; Guitart, Pickering & Byrne, 2012; Roy, Byrne &
Pickering, 2012; Steven, Pickering & Guy Castley, 2011). The approach seeks to be both
robust and replicable, minimising potential biases that can occur in narrative reviews, and by
being transparent in selecting and categorising the papers according to specific eligibility
criteria (Petticrew, 2001). The approach documents (i) where, when, and by whom research
was published; (ii) the geographical spread of the literature; (iii) types of methods used; (iv)
types of subjects examined; (v) types of variables measured; (vi) different disciplines
assessing the topic; and (vii) the types of results obtained (Guitart et al., 2012; Pickering &
Byrne, 2014; Roy et al., 2012). The approach begins with the identification of a topic, in this
case: “factors affecting PLOS and/or the evaluation of PLOS”. The second step involves
database search and priority selection. The third step is content assessment and study
selection. Finally, data collection processes are performed, extracting and handling data from
within each paper for use in the review.
2.2. Research questions, search terms and information sources “database”
We developed five research questions as shown in Figure 1. Electronic databases were
searched to identify original research papers related to ‘pedestrian level of service’. We
sought original research papers published in English language academic journals and
conferences with the search conducted progressively from January to November 2016. Whilst
we began with only the term ‘pedestrian level of service’ we progressively added other search
terms. Each paper was scanned to check if it met the eligibility criteria developed for the
71
review. Only peer reviewed academic papers were included. Book chapters, guidelines and
non-academic reports were used to help identify additional academic papers. We excluded
evaluations of LOS for motorised vehicles, bicycles and public transport.
2.3. Search strategy and initial selection
A total of 617 papers including many duplicates were initially obtained. Because the concept
of PLOS first emerged with the work of Fruin (1971) the period selected for searching the
papers required was set from 1971 to 2016. Zotero, EndNote X7.7, Mendeley, Jabref and
Excel software were used to support the screening stages. Zotero was used to transfer data
obtained from Google Scholar to Endnote.i Figure 1 shows the review protocol and the data
included at each stage. At the end of the initial selection stage 120 articles were selected for
content assessment.
2.4. Data extraction process and synthesis of results
The remaining articles were read in full and citations exported. Necessary data was extracted
including basic data on the paper (paper title, author, journal, year of publication, study
location, type of work); methodological information (type of target area, sampling methods,
data type, analysis method, model type); results (factors / variables affecting PLO,
classification of factors, the relevant importance for each factor), as shown in Figure 1.
72
Fig. 1 Review protocol (left) and papers obtained after the identification, screening and eligibility phases of the review, using the PRISMA technique
(right).
73
3. Results
3.1. Basic data on PLOS research papers
The final set of research papers are shown in Table 1. Although research on PLOS has been
conducted in 18 different countries, eighteen papers (31.1%) were conducted and published in
the USA. China produced 19 papers (17.2%) and India seven (12.1%), as shown in Figure 2.
74
Table 1 Previous studies on PLOS identified via the review
75
76
77
78
79
Research on PLOS has been accelerating, as shown in Figure 2a, with 25 of the 58 papers
published in the few short years between 2012 and 2016. Nineteen of the journal articles were
published in Transportation Research Record and the other 22 were published across a range
of journals. Six conference papers were published in the Annual Meeting of the
Transportation Research Board (TRB) and the other nine were published in different
conferences. There were two key scientific reports (Table 2).
a. Number of papers conducted per year b. Geographic distribution of the papers

Fig. 2 (a & b) Annual and geographic distribution of the 58 published research papers
80
Table 2 Journal and conference distribution of 58 papers on PLOS models assessed in this study
Number of
Publication Percentage
Title papers
Type %
included
Transportation Research Record 19 46.35%
Transportation Research Part A - Policy and Practice 3 7.30%
Journal of Transportation Engineering - ASCE 2 4.8%
Journal of the Eastern Asia Society for Transportation Studies 2 4.8%
Research Journal of Engineering and Technology 1 2.45%
Journal of Intelligent & Fuzzy Systems 1 2.45%
International Journal of Students' Research in Technology &
1 2.45%
Management
Land Use Policy 1 2.45%
Journal of advanced transportation 1 2.45%
Advances in Mechanical Engineering 1 2.45%
Journal
International Journal of Urban Sciences 1 2.45%

Transport Problems 1 2.45%
Journal of Transportation Systems Engineering and
1 2.45%
Information Technology
Infrastructure Planning Review 1 2.45%
Transportation Quarterly 1 2.45%
Road & Transport Research 1 2.45%
International Bicycle Fund 1 2.45%
ITE Journal - Institute of Transportation Engineers 1 2.45%
Mechanical Engineering 1 2.45%
Total 41 100%
Transportation Research Board 6 40.0%

Applied Mechanics and Materials 2 13.3%
Procedia-Social and Behavioral Sciences 1 6.67%
Conference
Malaysia University of Technology 1 6.67%

International Transport Research Conference 1 6.67%
Eastern Asia Society for Transportation Studies 1 6.67%
7th International Congress on Transportation Research 1 6.67%
ICRTSTM-16 1 6.67%
EDP Sciences 1 6.67%
Total 15 100%
3.2. Types of facilities and methods of evaluation

Researchers have been trying to develop PLOS for four key types of facilities, as shown in
Table 3. Of the 58 papers: 21 (36%) advanced PLOS for mixed streets; 19 PLOS for
footpaths, per se; 14 (24%) PLOS for intersections (14 papers, 24%) and 4 (7%) PLOS for
midblock crossings.
Many papers used quantitative data only (39.7%). We label these researchers the
geometricians given their focus on capacity, volume and geometric factors. Only nine papers
(15.5%) used qualitative data only – a group we label the experientialists given their focus on
non-geometric factors that influence the pedestrian experience. A larger proportion used a
81
combination of quantitative and qualitative data (44.8%) with many of these studies bridging
the two camps.
Methods included surveys, observations, video, or audits. Twenty-one studies (36.2%) used
mixed methods (two or more) for data collection; 14 studies (24%) used surveys (whether this
be questionnaires, field-surveys or intercept surveys), nine (15%) used observations and eight
(13.8%) used video. Interestingly, none of the studies used focus groups or Delphi techniques
with experts to help identify and rank PLOS factors, which at least in theory would offer a
viable approach.
Table 3 Methods of research papers assessing PLOS by facility type (1971-2017)
No. of PLOS
No. of PLOS No. of PLOS No. of PLOS
papers for
Category papers for papers for papers for Total
midblock
mixed areas footpaths intersections
crossings
Target Area 21 19 14 4 58
Sampling Methods
1. Survey (Questionnaire,
4 5 5 - 14
Field, Intercept)
2. Observations 8 1 - - 9
3. Video 2 4 2 - 8
4. Audit 3 1 - - 4
5. Results of guidelines 2 - - - 2
6. Mixed 2 8 7 4 21
Data Type
1. Quantitative 6 9 7 1 23
2. Qualitative 5 3 1 - 9
3. Mixed 10 7 6 3 26
Data Analysis
1. Regression 11 13 12 4 40
 Linear 2 9 5 - 16
 Stepwise 7 3 7 1 18
 Ordinal 2 1 - 3 6
2. Simulation - 1 - - 1
3. Fuzzy neural network - 2 - 2
4. Point system 9 2 1 - 12
5. Other 1 1 1 - 3
Approach Type
1. Capacity based 1 2 1 - 4
2. Street characteristics 7 2 4 1 14
3. Combination 13 15 9 3 40
82
Researchers have also tended to use four main approaches for analysing data to produce
PLOS outputs: a points system (12 papers, 20.6%) or some form of empirical regression (40
papers, 69%) simulation (1 paper), or a fuzzy neural network approach (2 papers). Of the
latter approaches, capacity-based modelling is particularly common. That said, only four
papers (6.9%) used a capacity-based model alone; 40 papers (69%) used a combination of
capacity-based modelling with other factors.
3.3. Factors used to measure PLOS
We themed and grouped the factors used in the studies to measure PLOS, under each facility
type, as shown in Table 5. There is considerable variation. Factors included distinct objects in
the built environment (i.e., the presence of bollards, water fountains or pedestrian signals)
collected from audits or geographic information systems (GIS) and more subjective factors
that represent an environment’s effect on pedestrians (i.e., concepts of comfort, connectivity
or security). Geometrical factors dominate with 45 of the 83 factors used in PLOS for mixed
streets, 22 of the 62 factors used in PLOS for footpaths, 15 of 41 of the factors used in PLOS
for intersections and 18 of the 30 in PLOS for midblock crossings. User behavioural factors
were generally given the least attention, and often only by the experientialists. Methods used
to select factors were usually unreported in these studies. Very few papers used primary
research with users to select network, environmental or behavioural factors, seemingly relying
instead on the researchers’ own expert judgement.
To answer our final research question, we sought to identify the priority being given to these
factors across the many studies for each facility type. This is not easy given the range of
factors and their potential overlap. We decided that the simplest measure of priority would be
produced by tabulating the number of times the factor was included across the 58 studies in
the dataset, as shown in Table 4. Given the reasonable sample size this frequency measure
offers the best objective measure we can provide of how the entire set of researchers valued
each factor. Many factors were not repeated across studies: thirty-two of the factors (39%)
were only used once; twenty-three (28%) were used twice, seven (8%) were used three times.
83
Table 4 Factors used to measure PLOS on different facilities
PLOS for Midblock
PLOS for Streets PLOS for Footpaths PLOS for Intersections
Crossings
Factors detail No. Factors detail No. Factors detail No. Factors detail No.
1.Factors related to the pedestrian environment
a.Geometric/Physical characteristics
Footpath width 12 Footpath width 10 Corner radius 2 Crosswalk distance 3
Number of traffic
On-street parking 5 On-street parking 4 Crossing distance 2 2
lanes
Median opening
Shoulder width 5 Shoulder width 4 Crosswalk marking 2 2
width
Number of traffic Crosswalk surface
5 Buffer presence 3 2 Median width 2
lanes condition
Buffer presence 4 Surface quality 3 Pedestrian signal 2 Crosswalk dummy 1
Number of traffic Farside crossing
Buffer width 4 Buffer width 2 2 1
lanes width
Pedestrian-vehicle
Driveway frequency 4 2 Vehicle lane width 2 Far-side cycle length 1
separation
Lighting 4 Adequate footpath 1 Corner space 1 Median presence 1
Mid-block crosswalk
Marking 4 Benches 1 Crosswalk width 1 1
type
Bicycle- vehicle Near-side crossing
Vehicle lane width 4 1 handicapped access 1 1
separation width
Corner island 3 Crossing facilities 1 Median openings 1 Near-side cycle length 1
Refuge island Pedestrian crossing
Curb ramp 3 Disabled guiding 1 1 1
presence time
Pedestrian waiting
Shading 3 Driveway frequency 1 Turn island presence 1 1
time
Two-step crossing Pedestrian-signal
Support facilities 3 Lateral streets 1 1 1
presence dummy
Advance stop bar 2 Marking 1 Warning signs 1 Signal Spacing 1
Width of painted
Benches 2 Ramp 1 - - 1
median
Width of restricted
Bollard 2 Shading 1 - - 1
median
Buffer barrier 2 Support facilities 1 - - Land-use type 1
Transportation
Crossing distance 2 1 - - - -
facilities
Disabled drinking
2 Tree 1 - - - -
fountain
Disabled guiding 2 Vehicle lane width 1 - - - -
Road facility
Disabled warning 2 1 - - - -
conditions
Facility type 2 - - - - - -
Fire hydrant 2 - - - - - -
Garbage facilities 2 - - - - - -
Grade 2 - - - - - -
Median 2 - - - - - -
Mid-block crossing 2 - - - - - -
Pedestrian-vehicle
2 - - - - - -
separation
Signals 2 - - - - - -
Slope 2 - - - - - -
Trees 2 - - - - - -
Awnings 1 - - - - - -
84
Crossing facilities 1 - - - - - -
Crossing
1 - - - - - -
opportunities
Disabled facilities 1 - - - - - -
Enclosure 1 - - - - - -
Elevator 1 - - - - - -
Public health facility 1 - - - - - -
Recreational
1 - - - - - -
facilities
Surface quality 1 - - - - - -
Toilet presence 1 - - - - - -
Transportation
1 - - - - - -
facilities
Trash receptacles 1 - - - - - -
b.Network Characteristics or (Environmental characteristics)
Obstructions 6 Obstructions 7 Businesses’ presence 2 - -
Footpath condition 5 Comfort 3 Aesthetics 2 - -
Footpath
Comfort 4 Footpath condition 3 2 - -
performance
Landscape
4 Safety 3 Pollution 2 - -
development
Safety 4 Security 3 Crosswalk presence 1 - -
Security 4 Weather conditions 3 Safety 1 - -
Attractiveness 3 Footpath continuity 2 Security 1 - -
Footpath presence 3 Movement easiness 2 - - - -
Complexity 2 Noise 2 - - - -
Continuity 2 Businesses’ presence 1 - - - -
Convenience 2 Aesthetics 1 - - - -
Footpath
Accessibility 1 1 - - - -
performance
Amenities 1 Pollution 1 - - - -
Building articulation 1 Social space 1 - - - -
Connectivity 1 - - - - - -
Footpath
1 - - - - - -
performance
Maintenance 1 - - - - - -
Transparency 1 - - - - - -
Transportation
1 - - - - - -
facilities
Social space 1 - - - - - -
System coherence 1 - - - - - -
Weather conditions 1 - - - - - -
2. Factors related to the traffic system and users
a.Traffic Characteristics
Vehicle speed 9 Pedestrian flow 11 Turning vehicle 10 Vehicle volume 2
Pedestrian volume 6 Pedestrian speed 5 Delay 9 Far-side total volume 1
Far-side turning
Vehicle volume 6 Vehicle volume 4 Turn conflicts 5 1
movements
Near-side total
Delay 5 Pedestrian density 3 Bicycle volume 4 1
volume
Pedestrian crossing Near-side turning
Pedestrian flow 4 Pedestrian space 3 4 1
time movements
Vehicle conflict
3 Vehicle speed 3 Vehicle volume 4 Pedestrian speed 1
points
Traffic flow 2 Bicycle flow 2 Pedestrian volume 3 - -
Adjacent roadway 1 Bicycle volume 2 Turning bicycle 3 - -
85
volume
Bicycle volume 1 Pedestrian volume 2 Pedestrian flow 2 - -
Heavy vehicle
1 Traffic conditions 2 Bicycle flow 1 - -
volumes
Adjacent roadway
Motor vehicle mix 1 1 Bicycle speed 1 - -
volume
Opposite bicycle
Pedestrian space 1 Bicycle speed 1 1 - -
flow
Electronic bike
Turning vehicles 1 1 Pedestrian space 1 - -
volume
Vehicle LOS 1 Mix of path users 1 Pedestrian speed 1 - -
Pedestrian crossing
- - 1 Vehicle flow 1 - -
time
Vehicle conflict
- - 1 Vehicles speed 1 - -
points
Vehicle-vehicle
- - Vehicle flow 1 1 - -
Conflict
b.Users' Characteristics & Behaviours
Encroachment 1 Day time trip 1 Noncompliance 1 Pedestrian age 3
Pedestrian-bicycle Frequency of using
Mix of path users 1 Economical 1 1 2
interaction crosswalk
- - Mix of path users 1 - - Pedestrian gender 2
- - Pedestrian age 1 - - Trip purpose 2
Pedestrian conflicts Pedestrian perceived
- - 1 - - 2
points difficulty
Pedestrian perceived
- - Encroachment 1 - - 2
safety
Horizontal
- - 1 - - - -
manoeuvres
- - Vendors attraction 1 - - - -
- - Vertical manoeuvres 1 - - - -
Note: (-) means: no additional factors existent. No. means: the number of times the factor was included across
the 58 studies in the dataset. PLOS means: pedestrian level of service
The geometry factor of footpath width was the most used factor for PLOS measurement for
both street and footpath studies, appearing in 12 of the PLOS for streets studies and 10 of the
PLOS for footpaths studies. Of the environmental factors, obstructions was used in six PLOS
for mixed streets studies, and seven PLOS footpath studies. Of the traffic characteristics
factors, pedestrian-flow was used in ten PLOS for footpath studies; vehicle speed was used in
nine PLOS for streets studies. None of the user behaviour factors were commonly used.
The function ‘Word Clouds’ was used to create a compound word cloud image and highlight
the more frequently used factors for PLOS measurement for each facility type. Compound
factors were developed using a ‘word-word’ formula. The word clouds for each facility type
are provided in Figure 3.
86
a. PLOS for streets
b. PLOS for footpaths
87
c. PLOS for intersections
d. PLOS for misblock crossings

Figure 3 (a, b, c & d) Word clouds demonstrating relative frequency of factors used to determine
PLOS by facility type.
88
4. Discussion
The growth in the development of different approaches to measurement of PLOS since 2013
is indicative of increased policy and planning interest in walking provision. The review
dataset shows that PLOS studies initially adopted capacity-based modelling approaches. But
though this approach continues in the work of those we call the geometricians, it has been
challenged by the alternative approaches of the experientialists. These two competing camps
fundamentally conceive PLOS in different ways: the former as a problem of facility capacity,
flow and geometric design; the latter as about the more subjective and experiential factors of
the built environment that produce pedestrian amenity. Many researchers have sought to
bridge these two conceptions, merging aspects of each in their work. But the litany of
different approaches and methods show that the field remains in significant dispute as to how
authorities should proceed. There are increasing studies but increasing disagreement. Even
within each camp the sets of factors proposed differs widely across authors, with little
convergence in recent years.
The advantages of the geometricians’ approach are relatively straight-forward. Their measures
of PLOS are relatively objective, based on the fundamentals of well-understood
volume/capacity ratios and with variations allowing for factors such as turning movements or
buffers from moving vehicles. Their approaches are both empirical and relatively replicable.
The disadvantages of their approaches are also the reason for the emergence of the competing
camp. The geometricians might rate a footpath that offers very good freedom of movement to
the pedestrian very highly even if it offers no shade, comfort or aesthetic amenity, and is
located immediately adjacent to a noisy and polluted, ten-lane, high-speed arterial.
The advantages of the experientialists’ approach are in its capacity to include almost any
factor via use of a points system or another form of multi-criteria evaluation. They can
incorporate the most subjective factors, such as comfort and attractiveness. But the
experientialists can’t seem to agree on what those factors might be, or how one should
measure them. Too many of their studies rely on ‘best guess’ approaches from learned
scholars, rather than surveys with pedestrians to inform factor selection, albeit the best guess
approach is a beneficial method in qualitative research to open new perspectives for further
inquiry (Baxter & Eyles, 1997). Worryingly, none of the studies reviewed using audit tool
approaches seemed to report any inter-rater reliability testing (the agreement between two or
more observers using the same tool) meaning that we were not certain how replicable these
89
approaches might be even within the same urban context, let alone if transferred to other
places.
Those who try to bridge the two camps attempt to get the best of both worlds. But the
limitations of the research based on subjective factors, in particular, makes such attempts
problematic. A reset may not be needed. But it would be helpful to try to articulate a research
agenda that can overcome these limitations. We suggest:
1. Improved methods for selecting and including factors in PLOS calculations are
needed, using techniques such that go well beyond the individual researcher to
include the input of a representative sample of relevant experts (i.e., via Delphi
studies) or a representative sample of pedestrian facility users (i.e., via surveys or
interviews).
2. There is a lack of studies identifying the importance of specific factors and especially
pedestrian treatments. This issue goes well beyond PLOS, per se, and is needed to
inform a range of transport and land use planning policies and programs. One under-
researched factor is pedestrian connectivity. Whilst analysis methods such as Space
Syntax offer interesting takes on connectivity and are becoming widely used in
academia and practice, more real-world analysis of pedestrian networks based on
sound theoretical conceptions is needed to identify the key effects of network designs.
3. A wider research base from more regions of the world is needed. The majority of the
literature is from North America. Walking rates in the USA are especially low by
world standards. PLOS is a critical issue for cities in Saudi Arabia, Iraq, Iran and
India, where the world’s largest pilgrimages occur, and where stampedes have
resulted in catastrophic loss of life (Aljohani, 2015; Haghighati & Hassan, 2013).
While sunlight may be considered a desirable pedestrian factor in the northern USA
or Canada, shade may be much more desirable in equatorial climates. Developing
factors that can be adapted across places within an overall framework would assist.
4. Meaningful inter-rater reliability testing is needed to prove that PLOS methods
produce replicable results, at least at the level of the urban context where they were
developed and applied, if not beyond.
5. There’s little research examining PLOS factors for disabled or special needs users in a
wide range, except for the study by Asadi-Shekari et al. (Asadi-Shekari, Moeinaddini
& Shah, 2013). Understanding how the use of PLOS for able-bodied pedestrians
90
might impact on those with disabilities would be useful.
6. Further disaggregation by facility type, particularly for different types of streets, may
make sense. Developing separate PLOS measures suitable for commercial streets, for
mixed-use streets or for suburban streets, may have significant merit.
7. Studies of the use of PLOS by agencies and the outcomes they obtain would be
desirable. It is not yet known quite how influential different approaches to measuring
PLOS are for real-world transport and built environment outcomes, or how design
changes in the transport/land use system influence user behaviour. Comparative
studies may be possible where local authorities under similar national/state policies
may take different approaches to measuring PLOS, creating different outcomes in
street, link or intersection design, and eventually changes in behaviour.
Endnotes
i
EndNote was used for organizing references and identifying duplicates. Further filtering was
undertaken using Mendeley and Jabref software and for transfer to Excel. Concurrently, all
papers considered relevant to the research objective were recorded and classified using Excel
spreadsheets at each of these stages.
References
Kang, L., Y. Xiong, and F. Mannering. Statistical Analysis of Pedestrian Perceptions of Sidewalk
Level of Service in the Presence of Bicycles. Transportation Research Part A: Policy and
Practice, Vol. 53, 2013, pp. 10-21.
Gallin, N. Quantifying Pedestrian Friendliness--Guidelines for Assessing Pedestrian Level of Service.
Road & Transport Research, Vol. 10, No. 1, 2001, p. 47.
Asadi-Shekari, Z., M. Moeinaddini, and M. Zaly Shah. Non-Motorised Level of Service: Addressing
Challenges in Pedestrian and Bicycle Level of Service. Transport reviews, Vol. 33, No. 2,
2013, pp. 166-94.
Ensley, J. Application of Highway Capacity Manual 2010 Level-of-Service Methodologies for
Planning Deficiency Analysis. Vol., 2012.
Landis, B., V. Vattikuti, R. Ottenberg, D. McLeod, and M. Guttenplan. Modeling the Roadside
Walking Environment: Pedestrian Level of Service. Transportation Research Record: Journal
of the Transportation Research Board, Vol., No. 1773, 2001, pp. 82-88.
Boodlal, L., "Accessible Sidewalks and Street Crossings-an Informational Guide," 2004).
Jaskiewicz, F. Pedestrian Level of Service Based on Trip Quality. Transportation Research Circular,
TRB, Vol., 2000.
Pushkarev, B., and J. Zupan. Urban Space for Pedestrian. A Report of the Regional Planning
Association, The MIT Press, Cambridge, Mass, Vol., 1975.
Polus, A., J. Schofer, and A. Ushpiz. Pedestrian Flow and Level of Service. Journal of Transportation
Engineering-Asce, Vol. 109, No. 1, 1983, pp. 46-56.
Tanaboriboon, Y., and J. Guyano. Level of Service Standards for Pedestrian Facilities in Bangkok: A
Case Study. ITE journal, Vol. 59, No. 11, 1989, pp. 39-41.
Lam, W., J. Morrall, and H. Ho. Pedestrian Flow Characteristics in Hong Kong. Transportation
Research Record, Vol., No. 1487, 1995, pp. 56-62.
Davis, D., and J. Braaksma. Level-of-Service Standards for Platooning Pedestrians in Transportation
91
Terminals. ITE journal, Vol. 57, No. 4, 1987, pp. 31-35.
Sarkar, S. Determination of Service Levels for Pedestrians, with European Examples. Transportation
Research Record, Vol., No. 1405, 1993.
Khisty, C., "Evaluation of Pedestrian Facilities: Beyond the Level-of-Service Concept," 1994).
Henson, C. Levels of Service for Pedestrians. Institute of Transportation Engineers. ITE Journal, Vol.
70, No. 9, 2000, p. 26.
Miller, J., J. Bigelow, and N. Garber. Calibrating Pedestrian Level-of-Service Metrics with 3-D
Visualization. Transportation Research Record: Journal of the Transportation Research
Board, Vol., No. 1705, 2000, pp. 9-15.
Sarkar, S. Qualitative Evaluation of Comfort Needs in Urban Walkways in Major Activity Centers.
Transportation Quarterly, Vol. 57, No. 4, 2003, pp. 39-59.
Mozer, D. Calculating Multi-Mode Levels-of-Service. International Bicycle Fund, Vol., 1994.
Muraleetharan, T., T. ADACHI, K.-e. UCHIDA, T. HAGIWARA, and S. KAGAYA. A Study on
Evaluation of Pedestrian Level of Service Along Sidewalks and at Crosswalks Using Conjoint
Analysis. Infrastructure Planning Review, Vol. 21, 2004, pp. 727-35.
Petritsch, T., B. Landis, P. McLeod, H. Huang, S. Challa, C. Skaggs, M. Guttenplan, and V. Vattikuti.
Pedestrian Level-of-Service Model for Urban Arterial Facilities with Sidewalks.
Transportation Research Record: Journal of the Transportation Research Board, Vol., No.
1982, 2006, pp. 84-89.
Asadi-Shekari, Z., and M. Zaly-Shah. Practical Evaluation Method for Pedestrian Level of Service in
Urban Streets' International Transport Research Conference. Penang, Malaysia, Vol., 2011.
Hidayat, N., K. Choocharukul, and K. Kishi. Pedestrian Level of Service Model Incorporating
Pedestrian Perception for Sidewalk with Vendor Activities. Journal of the Eastern Asia Society
for Transportation Studies, Vol. 9, 2011, pp. 1012-23.
Moher, D., A. Liberati, J. Tetzlaff, D. Altman, and P.The. Preferred Reporting Items for Systematic
Reviews and Meta-Analyses: The Prisma Statement. PLoS Med, Vol. 6, No. 7, 2009, p.
e1000097.
Pickering, C., and J. Byrne. The Benefits of Publishing Systematic Quantitative Literature Reviews
for Phd Candidates and Other Early-Career Researchers. Higher Education Research &
Development, Vol. 33, No. 3, 2014, pp. 534-48.
Ballantyne, M., and C. Pickering. The Impacts of Trail Infrastructure on Vegetation and Soils: Current
Literature and Future Directions. Journal of Environmental Management, Vol. 164, 2015, pp.
53-64.
Guitart, D., C. Pickering, and J. Byrne. Past Results and Future Directions in Urban Community
Gardens Research. Urban Forestry & Urban Greening, Vol. 11, No. 4, 2012, pp. 364-73.
Roy, S., J. Byrne, and C. Pickering. A Systematic Quantitative Review of Urban Tree Benefits, Costs,
and Assessment Methods across Cities in Different Climatic Zones. Urban Forestry & Urban
Greening, Vol. 11, No. 4, 2012, pp. 351-63.
Steven, R., C. Pickering, and J. Guy Castley. A Review of the Impacts of Nature Based Recreation on
Birds. Journal of Environmental Management, Vol. 92, No. 10, 2011, pp. 2287-94.
Petticrew, M. Systematic Reviews from Astronomy to Zoology: Myths and Misconceptions. BMJ :
British Medical Journal, Vol. 322, No. 7278, 2001, pp. 98-101.
Fruin, J., "Pedestrian Planning and Design," 1971).
Mori, M., and H. Tsukaguchi. A New Method for Evaluation of Level of Service in Pedestrian
Facilities. Transportation Research Part a-Policy and Practice, Vol. 21, No. 3, 1987, pp. 223-
34.
Dixon, L. Bicycle and Pedestrian Level-of-Service Performance Measures and Standards for
Congestion Management Systems. Transportation Research Record: Journal of the
Transportation Research Board, Vol., No. 1538, 1996, pp. 1-9.
Milazzo II, J., N. Rouphail, J. Hummer, and D. Allen. Quality of Service for Interrupted-Flow
Pedestrian Facilities in Highway Capacity Manual 2000. Transportation Research Record:
Journal of the Transportation Research Board, Vol., No. 1678, 1999, pp. 25-31.
Baltes, M., and X. Chu. Pedestrian Level of Service for Midblock Street Crossings. Transportation
92
Research Record, Vol., No. 1818, 2002, pp. 125-33.
Zhang, L., and P. Prevedouros. Signalized Intersection Level of Service Incorporating Safety Risk.
Transportation Research Record: Journal of the Transportation Research Board, Vol., No.
1852, 2004, pp. 77-86.
Steinma, N., and D. Hines. Methodology to Assess Design Features for Pedestrian and Bicyclist
Crossings at Signalized Intersections. Transportation Research Record: Journal of the
Transportation Research Board, Vol., No. 1878, 2004, pp. 42-50.
Petritsch, T., B. Landis, P. McLeod, H. Huang, S. Challa, and M. Guttenplan. Part 2: Pedestrians:
Level-of-Service Model for Pedestrians at Signalized Intersections. Transportation Research
Record, Vol., No. 1939, 2005, pp. 53-62.
Muraleetharan, T., T. Adachi, T. Hagiwara, and S. Kagaya. Method to Determine Overall Level-of-
Service of Pedestrians on Sidewalks Based on Total Utility Value, Presented at Proceedings of
the 83rd Annual Meeting of Transportation Research Board, Washington DC, 2004 B.
Muraleetharan, T., T. Adachi, T. Hagiwara, and S. Kagaya. Method to Determine Pedestrian Level-of-
Service for Crosswalks at Urban Intersections. Journal of the Eastern Asia Society for
Transportation Studies, Vol. 6, 2005, pp. 127-36.
Kim, K., L. Hallonquist, N. Settachai, and E. Yamashita. Walking in Waikiki, Hawaii: Measuring
Pedestrian Level of Service in an Urban Resort District. Transportation Research Record,
Vol., No. 1982, 2006, pp. 104-12.
Tan, D., W. Wang, J. Lu, and Y. Bian. Research on Methods of Assessing Pedestrian Level of Service
for Sidewalk. Journal of Transportation Systems Engineering and Information Technology,
Vol. 7, No. 5, 2007, pp. 74-79.
Muraleetharan, T., and T. Hagiwara. Overall Level of Service of Urban Walking Environment and Its
Influence on Pedestrian Route Choice Behavior: Analysis of Pedestrian Travel in Sapporo,
Japan. Transportation Research Record, Vol., 2007.
Jensen, S. Pedestrian and Bicyclist Level of Service on Roadway Segments. Transportation Research
Record: Journal of the Transportation Research Board, Vol., No. 2031, 2007, pp. 43-51.
Huang, T., and C.-J. Chiun, Modeling Level of Service on Pedestrian Environment, Presented at
Proceedings of the Eastern Asia Society for Transportation Studies, 2007.
Daniel, B., S. Choong, K. Ambak, and M. Abdullah, Pedestrian Level of Service Model for
Crosswalks at Signalized Intersections, Presented at Malaysia University of Technology, 2007,
2007.
NCHRP, " Seismic Analysis and Design of Retaining Walls, Buried Structures, Slopes, and
Embankments " (Transportation Research Board: NATIONAL COOPERATIVE HIGHWAY
RESEARCH PROGRAM, 2008), p. 137.
Petritsch, T., B. Landis, H. Huang, and R. Dowling. Pedestrian Level-of-Service Model for Arterials.
Transportation Research Record, Vol., No. 2073, 2008, pp. 58-68.
FDOT. Quality/Level of Service. Tallahassee, Florida Department of Transportation, Vol., 2009.
Bian, Y., J. Ma, J. Rong, W. Wang, and J. Lu. Pedestrians' Level of Service at Signalized Intersections
in China. Transportation Research Record: Journal of the Transportation Research Board,
Vol., No. 2114, 2009, pp. 83-89.
Vedagiri, P., and D. Anithottam, Pedestrian Flow Characteristics and Level of Service for Sidewalks in
Mixed Traffic Conditions, Presented at Transportation Research Board 91st Annual Meeting.
Christopoulou, P., and M. Pitsiava-Latinopoulou. Development of a Model for the Estimation of
Pedestrian Level of Service in Greek Urban Areas. Procedia-Social and Behavioral Sciences,
Vol. 48, 2012, pp. 1691-701.
Bunevska Talevska, J., and M. Malenkovska Todorova. Contribution to Sidewalk Pedestrian Level of
Service Analysis. Transport Problems, Vol. 7, No. 3, 2012, pp. 5-13.
Nagraj, R., and R. Vedagiri. Modeling Pedestrian Delay and Level of Service at Signalized
Intersection Crosswalks under Mixed Traffic Conditions. Transportation Research Record,
Vol., No. 2394, 2013, pp. 70-76.
Ling, Z., Y. Ni, C. Cherry, and K. Li. Pedestrian Level of Service at Signalized Intersections in China
Using Contingent Field Survey and Pedestrian Crossing Video Simulation. Transportation
93
Research Record, Vol., 2013.
Jensen, S., Pedestrian and Bicycle Level of Service at Intersections, Roundabouts and Other
Crossings, 2013, 2013.
Kim, I., and H. Kang, A Study of Delay-Based Level of Service on Pedestrian Facility, 2013, 2013.
Bian, Y., J. Lu, and L. Zhao, "Method to Determine Pedestrians Level of Service for Unsignalized
Intersections," in Applied Mechanics and Materials, 2013), pp. 1936-43.
Asadi-Shekari, Z., M. Moeinaddini, and M. Shah. Disabled Pedestrian Level of Service Method for
Evaluating and Promoting Inclusive Walking Facilities on Urban Streets. Journal of
Transportation Engineering-Asce, Vol. 139, No. 2, 2013, pp. 181-92.
Zhao, L., Y. Bian, J. Lu, and J. Rong. Method to Determine Pedestrian Level of Service for the Overall
Unsignalized Midblock Crossings of Road Segments. Advances in Mechanical Engineering,
Vol. 6, 2014, p. 652986.
Meng, J., Z. Zhu, and J. Zeng, The Research on the System of Level of Service of the City
Commercial Pedestrian Street, 2014, 2014.
Kim, S., J. Choi, S. Kim, and R. Tay. Personal Space, Evasive Movement and Pedestrian Level of
Service. Journal of Advanced Transportation, Vol. 48, No. 6, 2014, pp. 673-84.
Kadali, B. R., and P. Vedagiri, Modeling Pedestrian Crosswalk Level of Service (Los) under Mixed
Traffic Condition at Mid-Block Locations, Presented at Transportation Research Board 93rd
Annual Meeting 2014.
Asadi-Shekari, Z., M. Moeinaddini, and M. Shah. A Pedestrian Level of Service Method for
Evaluating and Promoting Walking Facilities on Campus Streets. Land Use Policy, Vol. 38,
2014, pp. 175-93.
Zhao, L., Y. Bian, J. Rong, X. Liu, and S. Shu, Modeling Pedestrian Level of Service on Sidewalks
with Multi-Factors Based on Different Pedestrian Flow Rates, Presented at Transportation
Research Board 94th Annual Meeting 2015.
Ye, X., J. Chen, G. Jiang, and X. Yan. Modeling Pedestrian Level of Service at Signalized Intersection
Crosswalks under Mixed Traffic Conditions. Transportation Research Record, Vol., No. 2512,
2015, pp. 46-55.
Kadali, B., and P. Vedagiri. Evaluation of Pedestrian Crosswalk Level of Service (Los) in Perspective
of Type of Land-Use. Transportation Research Part a-Policy and Practice, Vol. 73, 2015, pp.
113-24.
Hasan, T., A. Siddique, M. Hadiuzzaman, and S. Musabbir. Determining the Most Suitable Pedestrian
Level of Service Method for Dhaka City, Bangladesh, through a Synthesis of Measurements.
Transportation Research Record, Vol., No. 2519, 2015, pp. 104-15.
Archana, G. Analysis of Pedestrian Level of Service for Crosswalk at Intersections for Urban
Condition. International Journal of Students' Research in Technology & Management, Vol. 1,
No. 6, 2015, pp. 604-09.
Lazou, O., A. Sakellariou, S. Basbas, E. Paschalidis, and I. Politis. Assessment of Los at Pedestrian
Streets and Qualitative Factors. A Pedestrians’ Perception Approach. Vol., 2015.
Zhao, L., Y. Bian, J. Rong, X. M. Liu, and S. Shu. Evaluation Method for Pedestrian Level of Service
on Sidewalks Based on Fuzzy Neural Network Model. Journal of Intelligent & Fuzzy Systems,
Vol. 30, No. 5, 2016, pp. 2905-13.
Raghuwanshi, A., and V. Tare. Assessment of Pedestrian Level of Service for Mixed Lane. Research
Journal of Engineering and Technology, Vol. 7, No. 1, 2016, p. 11.
Daniel, B., S. Nor, M. Rohani, J. Prasetijo, M. Aman, and K. Ambak, Pedestrian Footpath Level of
Service (Foot-Los) Model for Johor Bahru, 2016, 2016.
Chandana, K., M. Ibrahim, and R. Kumar. Analysis of Pedestrian Level of Service on Footpath in
Hyderabad. Vol., 2016.
Baxter, J., and J. Eyles. Evaluating Qualitative Research in Social Geography: Establishing ‘Rigour’in
Interview Analysis. Transactions of the Institute of British geographers, Vol. 22, No. 4, 1997,
pp. 505-25.
Aljohani, A., Pilgrim Crowd Dynamics, University of Birmingham, 2015.
94
Haghighati, R., and A. Hassan. Modeling the Flow of Crowd During Tawaf at Masjid Al-Haram.
Journal of Mek, Vol. 36, 2013, pp. 2-18.
95
5 Identifying and measuring the most important factors for pedestrian
level of service measurement: results from a Delphi study and a
pedestrian perception survey in Brisbane, Australia
An earlier version presented in Chapter 5 was first introduced as:

 Raad, N., & Burke. M. (2018). Identifying the most important factors for pedestrian
level of service measurement: results from a Delphi study in Brisbane, Australia.
Paper presented at the 98th Annual Meeting of Transportation Research Board (TRB),
January 13−17, 2019, Washington, D.C., USA.
A revised and much improved version was then submitted to the journal of “Case Studies on
Transport Policy” as:
pedestrian level of service measurement: results from a Delphi study and pedestrian
perception survey in Brisbane, Australia.
 First Submission on 26/08/2019

The previous chapter reviewed prior PLOS models including the methodologies and
approaches used for establish these models, factors affecting the pedestrian level of service
examined, the limitations of these models and suggestions for new approaches. An intensive
list of previously examined factors affecting pedestrian level of service was identified. The
various pedestrian facilities identified included whole streets, footpaths, intersections and
midblock crossings. However, in much research, the possible factors that should be included
in the pedestrian level of service calculations for footpaths in commercial streets, or which are
appropriate for Australian cities have not been explicitly referred to. This chapter summarises
the results of Task 2, step 1 (in detail), and Task 3 including steps 1, 2 and 4. The main aim of
this chapter is to select the appropriate factors for the Australian context that mainly influence
the PLOS on footpaths in commercial areas using systematic and reliable methods. The main
method used to achieve this aim was a Delphi process with a panel of experts. Volunteer
experts for the Delphi process were recruited with the cooperation of the Pedestrian and
Bicycle Transport Institute of Australasia (PedBikeTrans). Based on the results of this study,
the final list of the factors affecting PLOS on footpaths in commercial streets was developed,
96
and was then used to obtain pedestrians’ perceptions regarding the degree of importance of
each factor. The results of a pedestrian intercept survey are almost complete in the following
chapter (Chapter 6) but summary materials are used to contribute to this chapter. A pedestrian
intercept survey was employed to gather pedestrians' perceptions regarding the satisfaction
and the importance of the factors affecting PLOS on footpaths in commercial streets within
the Brisbane CBD. The relative importance or weight for each factor affecting PLOS was
generated based on pedestrians’ perceptions. Then the PLOS calculation model was
developed based on a satisfaction approach and using a point system method.

Primary questions:
RQ2: What do experts think should be included in PLOS models appropriate for Australian
cities?
Sub-questions:
environment?

This study is believed to be the first to use the Delphi technique with a large set of experts
from a variety of backgrounds, including urban planning, urban design, engineering, transport
planning, and pedestrian planning, in order to systematically and reliably identify the factors
affecting PLOS on footpaths in commercial streets. This study could be also considered a
unique attempt to establish a PLOS model using a combination of the Delphi process,
pedestrian perceptions and the satisfaction or performance approach. Most of the key
measures in this study have attained statistical significance.
97
 The outcome of this study demonstrates a novel, robust method for determining and
measuring the factors that affect PLOS using a literature review, an online survey, a
‘walkshop’ as part of an inter-linked Delphi process, a pedestrian intercept survey and
a point system method.
 The factors identified in this study, as the most important affecting PLOS which is
appropriate for the Australian context, can be used by practitioners to create new
PLOS models.
 The Delphi technique used in this study helps to refine an extensive list of factors
affecting PLOS for footpaths in commercial streets. This list of factors can be used by
practitioners in other PLOS investigations.
 The approach and methods applied in this study could be used for evaluation of PLOS
on footpaths in commercial streets in Australia as well across a range of different
facility types and in differing urban contexts.
 The approach and methods applied in this study could be adopted to identify the
factors affecting LOS for different pedestrian facilities such as intersections, and other
transportation facilities, such as bicycle level of service.

This chapter includes a co-authored paper that summarises the results of Task 2 including step
1 (in detail) and step 2, and Task 3 including steps 1, 2 and 4. The bibliographic details of the
co-authored paper, including all authors, are:
pedestrian level of service measurement: results from a Delphi study and pedestrian
perception survey in Brisbane, Australia.
execution or interpretation, of at least that part of the publication in their field of
expertise;
98
requirements.

Nowar Raad  Designed the data collection process for the Delphi technique, including the
site selection, participant recruitment, and questionnaires.
 Attended the study area to conduct the walking course and workshop with
experts.
 Managed and guided the panel of experts in both the walking course and the
workshop.
 Designed the data collection process for the pedestrian intercept survey,
including the site selection, participant recruitment, and questionnaires.
 Complied, cleaned and analysed the data.
reviewers.
 Prepared slides to be presented at the conference.
 Responsible for revisions made at the suggestion of the journal reviewers
and resubmitting the paper.
 Provided assistance in the synthesis and interpretation the results.
 Provided assistance in preparing the slides to be presented at the conference.
 Attended the study area to conduct the walking course and workshop with
experts.
(Signed) _________________________________ (Date) 26/08/2019

(Signed) __________________________________ (Date) 26/08/2019

This chapter is an exact copy of the paper referred to above
99
Identifying and measuring the most important factors for pedestrian level of service
measurement: results from a Delphi study and pedestrian perception survey in
Brisbane, Australia
Abstract
In today’s rapidly changing environment, there is a need for quality people places that serve a
diverse range of users through innovative planning and design. Pedestrian level of service
(PLOS) is a common method used for assessing the quality of pedestrian facilities.
Identification of the most important factors relating to footpaths and their surrounding
environments is one of the major challenges for researchers and practitioners. Previous
approaches and methods for factor selection have had numerous limitations, with many
studies using expert judgement, often from a single researcher. This paper reports on a
research effort that aims to overcome these problems by proposing a novel method for PLOS
factor identification and applying the selected factor to measure PLOS value. The approach
uses the outputs of a previous systematic quantitative literature review on PLOS factors as
inputs to a Delphi study involving experts in Brisbane, Australia. In 2017, some 36 experts
recruited from the professional association, the Pedestrian and Bicycle Transport Institute of
Australasia (PedBikeTrans), participated in a questionnaire survey. Twenty of the same
experts then participated in a follow-up ‘walkshop’ involving walking a dedicated route and a
workshop to re-examine the possible set of PLOS factors. Thirty-three factors affecting PLOS
were eventually identified for inclusion in the proposed PLOS model. The proposed model for
assessing PLOS adopts a satisfaction approach using a point system method. An on-street
intercept questionnaire survey was employed to gather pedestrians' perceptions regarding the
importance and satisfaction rating of the 33 factors affecting PLOS with 312 pedestrians in
the Brisbane CBD. Adelaide and George Streets in Brisbane’s CBD were chosen as the setting
for applying the methodology of PLOS measurement. The overall PLOS score shows that the
condition of measuring segments was better than moderate, which reflects pedestrians’
perceptions. The results are limited only to commercial streets in city/activity centres in the
Australian setting, but the approach and methods may open the pathway for much more
robust PLOS measures to be developed for a range of facility types and across different urban
contexts.
100
1. Introduction
Footpaths (known in the USA as sidewalks) are vital components of any transportation
network and are key in achieving sustainable travel behaviours (Saito, Oguma, Inoue, Tanaka
& Kobori, 2013; Van Cauwenberg et al., 2012). Walking is directly influenced by the footpath
and its surrounding environment (Sung, Go, Choi, Cheon & Park, 2015). The promotion of
current walking levels means creating improved conditions for walking. This can be done by
first examining the potential methods used to assess the conditions for walking in cities
(Mateo-Babiano, 2016). Since the 1970s increasing attention has been given to assessing
pedestrian environments and footpaths for how they facilitate walking (Raad & Burke, 2018).
A key focus has been the development of pedestrian level of service (PLOS) measures. A
level of service may be defined “as an assessment of the quality of service or an overall
infrastructures” (Raad & Burke, 2018).
We recently reported the findings of a major systematic quantitative literature review on
PLOS estimation which identified a large set of factors that have been used previously by
researchers in the field. In this paper we report on a Delphi study using experts in pedestrian
planning to obtain a more robust set of factors for use in PLOS estimation. The main
contributions of the paper are methodological, with an additional set of applied findings that
are of particular use in the Australian case and which may have use in similar contexts, such
as many North American cities.
Previous approaches to identifying factors for PLOS estimation include reviews of previous
studies, observations in the field or by video, questionnaires, focus groups, interviews or
expert judgement (Hess et al., 2017; Smith, Troped, McDonough & DeFreese, 2015; Spittaels
et al., 2009; Weliange, Fernando & Gunatilake, 2014). As shown in our review, the set of
studies we examined had a set of key limitations, with some poor levels of rigour in approach
and methods (Raad & Burke, 2018). Further, as cities change, and as pedestrians themselves
change, the set of factors that influence walking will also be dynamic (Hess et al., 2017).
Finding methods to estimate PLOS that are relevant to the facility type and local context, and
that can be used and regularly updated without too much difficulty, is a major challenge.
Our approach is to combine a review of past approaches with a Delphi study of experts,
followed by intercept questionnaires with pedestrians. In this paper we report in detail the
101
Delphi study phase of our research agenda and the results gained from intercept survey. The
paper is structured as follows: i) the background research and issues are explained; ii) the
approach and methods are outlined; iii) the data analysis methods and key results are
provided; and, iv) a discussion spells out the contributions and limitations of the work
presented, along with some of the remaining issues for researchers in the field of PLOS.
2. Background
PLOS methods are commonly used by urban and transportation researchers to assess
pedestrian facilities. The concept of pedestrian level of service (PLOS) is often used to
objectively quantify how well footpaths in streets accommodate pedestrians. There is
significant dispute as to what should, or should not, be included when calculating PLOS. A
wide range of measures and tools has been developed for different pedestrian environments.
These aid designers and engineers in developing or retrofitting streets, crossings and paths.
Level of service (LOS) method can indicate existing conditions for street users or forecast
possible conditions for hypothetical designs to be built in future. The first attempts at PLOS
assessment considered only geometric factors, especially volume vs. capacity (Fruin, 1970).
Researchers began to identify a set of more qualitative factors, realizing that the pedestrian
experience cannot be seen only as a trip from point A to point B (Jaskiewicz, 2000). As shown
in our previous review paper (Raad & Burke, 2018) these qualitative factors, such as safety,
security, comfort and convenience, began to be included in PLOS estimation, with significant
debate as to which factors were important, and how they might be calculated. Factors such as
pedestrian path width, surface quality, presence of a shoulder, on-street parking and
obstructions began to be used. Facilities such as benches and drinking fountains, and
amenities such as path lighting were used in other studies (Raad & Burke, 2018).
A great number of experts have driven this research effort forward over time. Since Fruin’s
pioneering work (Fruin, 1970) research teams such as Landis et al. have explained how the
factors affecting pedestrians' perceptions of safety and comfort in street environments are
based on various complex factors, including personal safety (the threat of accidents), personal
security (the threat of assault), architectural interest, shade, lighting and amenities, the
presence of other pedestrians and conditions at intersections (Landis et al., 2001). Recently,
researchers have been working to develop reliable and valid tools to assess walkability by
measuring the characteristics of the neighbourhood built environment (Troped et al., 2006)
and the quality of the walking environment (Ewing & Handy, 2009). Other assessment
102
instruments were developed to examine walkability by capturing the contextual elements on
neighbourhood streets which promote or inhibit walking and cycling across different domains
(Pikora et al., 2003; Pikora et al., 2002). Pikora et al. tried to develop more comprehensive
methods using factors grouped into four themes; functional, safety, aesthetic and destination
issues, with judgments as to how individual factors may influence walking behaviour (Pikora
et al., 2003; Pikora et al., 2002). Williams et al. proposed methods more targeted at footpath
maintenance including factors such as path levelness, blockages, cleanliness, and surface
condition (Williams et al., 2005). Dannenberg et al. developed an audit tool that covered such
factors as pedestrian facilities, pedestrian-vehicle conflicts, crosswalks, route maintenance,
walkway width, roadway buffer, universal accessibility, aesthetics, and shade (Dannenberg et
al., 2005). Although several attempts to develop walkability audits have been undertaken by
planners and engineers by including a number of features that affect the entire roadway
corridor environment, there has been no consensus on the degree of importance to pedestrians
for each feature, using statistically reliable methods. Also, such features have not yet been
statistically tested or widely applied.
Eventually there were separate PLOS methods that could be applied to footpaths (per se),
intersections and mid-block crossings, as experts understood bespoke solutions are needed in
each of these situations. Gathering a variety of information about users’ perceptions and
consideration of the various requirements of different groups of pedestrians are significant
issues, and can affect design decisions, therefore since the 1970s, there have also been
changes in planning philosophy. It is now recognized that a street design for one group may
actually reduce other individuals’ capabilities. Footpaths must be designed to create
accessibility for the majority of users, meet disability standards, and also have attractive
features for those who choose to walk (Boodlal, 2004). To assess and calculate walking
conditions, the most common method uses the PLOS. Various general conditions were
determined by previous studies for selecting effective indicators to include in PLOS. Two
general approaches were used to model PLOS. Of the latter approaches, capacity based
modelling is particularly common, and another is a roadway characteristic based model. Some
previous studies have attempted to use combination models to evaluate PLOS, while other
studies have relied on a single approach (Raad & Burke, 2018). Methods including surveys,
observations, video, or/and audits have been used for data collection (two or more).
Interestingly, none of the studies used focus groups or Delphi techniques with experts to help
103
identify and rank PLOS factors, which at least in theory would offer a viable approach. Most
of the studies also adopted two main approaches for analysing data to produce PLOS outputs,
including either a points system or regression analysis (Raad & Burke, 2018).
Often, investigators’ personal judgements have been central to decisions to include, exclude
or weight certain factors. Although some previous investigators have identified several factors
significantly correlating with PLOS in their models, the relative importance of each factor and
how it apparently influences different walking behaviour in a given context has been a
weakness in previous research efforts. Whilst this is acceptable for early research efforts in a
particular field, it becomes less acceptable over time. Very few of the methods and approaches
in the literature use either statistical methods to ensure robustness in gaining consensus from
experts, gather perceptions of pedestrians themselves, or ensure there is inter-rater reliability
for assessors (ensuring consistency in the scores that are given to a facility by assessors using
a given PLOS tool) (Lee, Jang, Wang & Namgung, 2009; Raad & Burke, 2018). This sets a
challenge for those seeking to improve PLOS measurement.
3. Methodology
If a new approach and method is needed, the challenge is to develop an assessment model that
uses reliable methods to identify and measure factors affecting the PLOS for the full range of
pedestrians. In this paper we look at developing a satisfaction approach to evaluate PLOS on
footpaths in commercial streets, as shown in Figure 1. In our systematic quantitative literature
review, we reported that nearly all the factors experts have previously used to measure PLOS
had been identified. This information was used as a key input in the second phase of our
broader research effort: a Delphi process with a panel of experts from PedBikeTrans in
Australia. The aim of this work was to refine the list of factors identified in the initial
systematic quantitative literature review, toidentify those factors that should be included for
examining PLOS in commercial streets appropriate for Australian cities, and to measure these
factors to calculate perceived PLOS. The Australian experts’ viewpoints were adopted to
avoid a bias that might be perceived by the authors in selecting the factors affecting PLOS.
Many researchers have adopted experts’ viewpoints and the Delphi method to identify the
factors related to the characteristics of the built environment, the pedestrian and cycling
environment in neighbourhoods (Troped et al., 2006), the quality of the walking environment
(Ewing & Handy, 2009) and aspects of footpaths and neighbourhoods (Pikora et al., 2003)
and so on. Pikora et al. (2002) described the first reported attempt to develop a comprehensive
104
instrument to measure the physical environmental factors related to walking and cycling in
local neighbourhoods. A framework of the potential environmental influences on walking and
cycling was developed based on published evidence and policy literature, interviews with
experts and a Delphi study (Pikora et al., 2003). Also, the physical features of an environment
that influence walking among seniors were determined by Cunningham et al. (2005) based on
a review of urban planning and health literature and feedback from urban planning and health
research professionals (Cunningham et al., 2005). Day, Boarnet et al. (2006) identified the
characteristics of the built environment with 162 items that are potentially linked to active
living based also on a literature review, focus group interviews, a panel of experts (Day et al.,
2006). Park characteristics affecting physical activity were also determined by Bedimo-Rung
et al. (2006) through a series of meetings with an expert panel using the Delphi method
(Bedimo-Rung et al., 2006). Troped et al. (2006) identified the trail characteristics that affect
use through adopting a trans disciplinary team viewpoint (Troped et al., 2006).
3.1. Data collection

3.1.1. Inputs from the literature
Our previous review systematically searched and categorised the relevant literature (Raad &
Burke, 2018). The list of potentially important factors affecting PLOS for footpaths is shown
in Table 1.
Figure 1 The proposed methodology of the PLOS model for footpaths in commercial areas.
105
3.1.2. Delphi process
The Delphi technique is a process of deriving group consensus for any topic when the
required information is not available by using a systematic approach (Pikora et al., 2003). In
the Delphi technique, the group survey is an efficient and dynamic process. The main features
of a survey in the Delphi technique are an anonymous, written, multi-stage survey process and
controlled feedback (von der Gracht, 2012). A Delphi consensus approach was selected in part
to decrease the required number of Delphi rounds. The systematic review was considered as
the first round of the Delphi process for this study. The second round was an online survey of
key experts from relevant disciplines, and the third round was a ‘walking workshop’ or
‘walkshop’ (Figure 2).
The full procedure, with the review paper included as the ‘first round’ of the Delphi process,
is shown in Figure 2.
Online survey (second round of Delphi)
The aim of this round was to gain primary experts’ perceptions and to examine the content
validity of the extensive list of potential factors affecting PLOS developed in the literature
review. The draft of the hierarchical framework obtained from the first round (the literature
review) was organised in the form of a questionnaire. A letter introducing the study and the
Delphi process was sent to members of PedBikeTrans, inviting them to participate. Then, a
link to the online survey was sent. The experts were asked to rate the importance of each
factor using a five-point scale, ranging from 1 ‘not at all important’ to 5 ‘extremely
important’, and to suggest new factors, previously omitted, that they believed were important.
A total of 36 experts completed and submitted the online survey. Twenty-four participants
were male, twelve were female, reflecting the gender bias in these professions in Australia.
Walkshop (third round)
This round comprised two components, an initial walking course and workshop. The experts
who completed the online survey were invited to attend the walkshop, with a prize draw as an
incentive for participation. Participants registered online and names were matched. A total of
twenty experts participated in both the walking course and the workshop components. Twelve
participants were male, eight were female. Participants were provided with a sheet that
included information about the study and instructions. They were briefed on the purpose,
rules, and instructions for the two components. They were made to walk the walking course
prior to (re-)evaluation and weighting of the list of PLOS factors obtained in the earlier round.
106
Figure 2 Flow chart of the Delphi process.
107
Experts were asked to consider the factors affecting PLOS only on footpaths, and were asked
to neglect the factors affecting other pedestrian facilities such as intersections and midblock
crossings. The walking course was a 3km loop which included a set of eleven commercial
streets segments in Brisbane’s central business district. The route was designed to subject the
participants to the full variety of factors affecting PLOS identified in the previous round. For
risk management reasons, walkers were allocated to two trip leaders, who helped guide them
through the itinerary. Participants were asked to complete a PLOS rating sheet independently
as they walked each segment. While walking, they were encouraged to use their imagination
and accumulate experiences in order to consider which factors have more effect and make
more contributions to PLOS than others, as well as to the PLOS rating. The course included
footpath segments with traffic and footpath conditions typical of Australian city centres.
During the walking event stage, adjacent posted vehicular traffic speeds ranged from 30 to 50
km/hr. At key locations, the experts were asked to make a personal judgement as to how they
rated the PLOS on that street for each segment. In addition to the PLOS rating sheet, the
experts were provided with a map that displayed each required segment and were asked to
record their evaluation of the PLOS for each segment on a 6-point (LOS ‘A’ to LOS ‘F’)
scale. They were advised to evaluate how safe/comfortable they felt it is for pedestrians as
they travel each segment. Level ‘A’ was considered the most safe / comfortable (or least
hazardous), while level ‘F’ was considered the least safe / comfortable (or most hazardous).
This procedure was adopted only to provide the experts with a general idea about the selected
segment conditions in a real situation. The experts were asked to use their imagination and
experience to make balanced decisions while rating PLOS at each segment to prevent any
confusion that may arise, because the ‘comfortable’ walking footpath may not be the ‘most
safe’ or vice versa. They were also asked to highlight which factor negatively or positively
contributed to increase or decrease PLOS value in terms of safety or/and comfort. The results
of this procedure then helped the experts to rate the importance of each factor based on their
perceptions of PLOS obtained during the walking course.
The walking course included streets with two to four lanes, one way and two way streets, and
with undivided, divided, and continuous left turn median lane configurations. There were also
a variety of footpath conditions and characteristics. For example, some segments had striped
shoulders and some included designated bicycle lanes. Other segments had on-street parking,
landscaped buffers, street trees, and buildings adjoining the footpaths whose structures and
108
awnings covered the footpaths. There were different forms of land use and street frontages
including retail commercial development (ranging from large retail shopping centres to small
shops), office buildings, professional services, churches, banks, restaurants, cafes, other
businesses, parking, and some recreational facilities.
Following the walking course, experts returned to the workshop room. They were provided
with a sheet listing the revised set of factors affecting PLOS that emerged from the second
round (online survey) of the Delphi study. The sheet included both the group median scores
and Content Validity Ratios (CVRs) derived from the survey for each factor (Table 1). Then,
the experts were asked to make changes to their previous decisions recorded on the online
survey using a five-point Likert scale ranging from ‘extremely important’ to ‘not at all
important’ for each factor, based on:
 Their perceptions obtained during the walking course; and
 The group median score and CVR for each factor
The data of this round were manually transferred to the SPSS software package and double
checked for accuracy. Likert scale responses were converted to a numerical value for
statistical analysis and computation.
3.1.3. On-street questionnaire survey
An on-street intercept questionnaire survey was then employed to gather pedestrians'
perceptions of the factors affecting PLOS on these same commercial street segments within
the Brisbane CBD. An online questionnaire method using a tablet PC was adopted. The
questionnaire was designed to measure pedestrians’ perceptions in five main categories, with
33 sub-items or factors that were the outputs from the Delphi process (Table 2). These
categories include geometric footpath characteristics, obstructions and interruptions,
aesthetics and amenities, pedestrian traffic issues and adjacent traffic. The first part of the
questionnaire obtained the respondent’s characteristics including age, gender, education, trip
purpose and the frequency of use of the footpaths. The second part was allocated for assessing
the importance and satisfaction of each existing footpath facility for the pedestrians. Images
were used to help define the factor in easy to understand terms, and trained researchers were
able to further describe each factor to respondents who were uncertain. Respondents were
asked to rate both the satisfaction (how satisfied they were) and the importance (how
important good performance is to them) using a five-point Likert scale for each of the factors
in relation to these paths in the study area. The scale used for the level of satisfaction was
109
from 1 = very dissatisfied, to 5 = very satisfied, and the level of importance was from 1 = not
at all important to 5 = extremely important. Footpaths with different characteristics in the
CBD area were evaluated by the questionnaire as a result of involving a wide range of
footpath aspects that can be used in various contexts. The outputs of this survey were used to
generate the relative importance and satisfaction percentage for each aspect of the footpath
environment. 312 respondents aged 16 and over were randomly selected to ensure that the
perceptions of a wide range of people toward the importance and satisfaction of factors
affecting PLOS on footpath were captured. Data collection was performed across the period
16th May to 3th June 2018 on various days of the week, with the surveys undertaken from 10
a.m. to 5 p.m.
3.2. Data analysis
3.2.1. The Delphi technique
One of the main criticisms of using the Delphi method is that standard analysis procedures
have not been uniformly developed. Consensus measurement and stability can be considered
the main goals of a Delphi study and a valuable component of data analysis and interpretation
(Flanders, 1988; Seagle & Iverson, 2001). Responses are considered as being in consensus
when the answers among participants tend to converge with decreasing variability in scores.
The consistency of responses between successive rounds of a study indicates that responses
are considered stable. The most common statistical methods of consensus measurement
include means, standard deviations, medians, interquartile ranges, and a composite score.
Delphi studies more commonly use the Wilcoxon matched-pairs signed-ranks test and
Pearson Product-moment correlation coefficients to measure stability in responses between
successive rounds (von der Gracht, 2012).
For this study, the data analysis comprised primarily descriptive statistics using a practical,
non-parametric approach. The following computations were performed for data analysis: 1) a
composite score for each factor was calculated to rank order the factors; and 2) medians,
interquartile ranges were also calculated for each factor in rounds 2 and 3 as measures of
opinion convergence, and consensus measurement. We could not use Wilcoxon matched-pairs
or Pearson Product correlations due to differences in sample sizes between rounds 2 and 3.
We instead use intra-class correlation coefficients to determine convergence of opinion and to
measure the stability between rounds (Koo & Li, 2016). Volunteer experts for the Delphi
process were recruited with the cooperation of the Pedestrian and Bicycle Transport Institute
110
of Australasia (PedBikeTrans). The Delphi study involved experts from a variety of
backgrounds, including urban planning, urban design, engineering, transport planning, and
pedestrian planning. They came from local and state governments, universities and private
consultancies, along with one recently retired expert.
Online survey (second round of Delphi)
The results of this round were carefully transcribed and summarised (see Table 1). Based on
experts' answers derived in this phase, the group median scores and Content Validity Ratios
(CVRs) were calculated for each factor.
The CVR can simply be defined as an item statistic, which is beneficial for rejecting or
retaining the specific items and can be calculated by following the equation (n - N/2)/ (N/2),
where n is the number of experts, indicating an ‘important’ rating of a particular factor, and N
is the total number of experts (Lawshe, 1975). According to the Lawshe measure for
quantifying consensus and the validity of judgments, and because of the sample size of 36, the
factors with a content validity ratio (CVR) less than 0.33, a mean less than 3.5, and a median
less than 3.5 were eliminated. Eighteen of the 52 total factors were identified and nominated
for elimination. This included 3/10 from the set of factors in the geometric characteristics
group; 2/7 from obstructions and interruptions; 11/23 from aesthetics and amenities; nil from
pedestrian traffic; and 2/7 from adjacent vehicular traffic (see Table 1). These results were
then prepared for the third round of the Delphi process.
Table 1 Mean, Median and CVR results of second Delphi round.

Factors Mean Median Content Validity Ratio
Buffer width 3.50 4.00 0.667
Crossing opportunities 4.25 4.00 0.944
Quality of footpath surface 4.03 4.00 0.944
Footpath slope & grade 3.61 4.00 0.722
Adequate footpath width 4.53 5.00 0.944
On-street parking 2.17 2.00 -0.389
Pedestrian-traffic separation 3.75 4.00 0.889
Ramp 3.64 4.00 0.722
Shoulder width 2.80 3.00 0.278
Height clearance 3.19 3.00 0.444
Footpath connectivity 4.64 5.00 1
Footpath continuity 4.64 5.00 1
Frequency of lateral ways 3.30 3.00 0.778
Ease of movement 4.25 4.00 1
Permanent obstructions 4.19 4.00 0.944
Temporary obstructions 3.44 3.00 0.667
Obstruction width 3.69 4.00 0.778
Awnings 3.55 4.00 0.778
Benches 3.25 3.00 0.556
111
Bollards 3.00 3.00 0.389
Building articulation 3.03 3.00 0.556
Business presence 3.25 3.00 0.722
Crossing facility type 4.05 4.00 0.944
Activation of side facades 3.69 4.00 0.833
Facilities for the disabled 3.97 4.00 0.833
Drinking fountain 3.16 3.00 0.611
Enclosure 2.83 3.00 0.444
Fire hydrant 2.00 2.00 -0.444
Footpath cleanliness 3.64 4.00 0.667
Footpath maintenance 3.97 4.00 0.944
Footpath surveillance 3.50 4.00 0.5
Garbage facilities 3.03 3.00 0.5
Guidance signs 3.75 4.00 0.833
Lighting 4.08 4.00 0.889
Recreational facilities 2.78 3.00 0.222
Shade 4.33 5.00 0.889
Toilet nearby 3.00 3.00 0.389
Public transportation facilities 3.92 4.00 0.778
Trees and plants 4.08 4.00 0.889
Wayfinding guidance 4.19 4.00 0.944
Pedestrian free manoeuvre 3.91 4.00 0.778
Pedestrian conflict points 4.19 4.00 0.889
Pedestrian speed 3.50 4.00 0.722
Pedestrian signs 3.66 4.00 0.722
Pedestrian volume 3.94 4.00 0.833
Bicycle speed 3.44 4.00 0.667
Bicycle volume 3.53 4.00 0.667
Number of traffic lanes 3.22 3.00 0.5
Vehicle conflict points 4.03 4.00 0.944
Vehicle lane width 3.30 3.00 0.611
Vehicle speed 4.16 4.50 0.889
Vehicle volume 4.05 4.00 0.889
3.2.2. Consensus and stability of Delphi rounds

Composite scores
Composite scores were only calculated on the final round data for each factor by adding the
value of the individual responses. Twenty experts’ ratings for each of the 52 factors were
added to obtain a factor composite score. For instance, if all twenty experts rated a factor as
‘strongly important’, the factor would receive a composite score of 100 (20 experts x 5 =
100). As a result of convergence in responses among participants in the final rounds of the
Delphi process, only the results of the final round (after the walkshop) were considered for the
discussion of composite scores because they were the most accurate and valued.
At the composite score stage, consensus in responses for each factor was considered as
established if the two following important statistical criteria were achieved: 1) at least 60%
(12 of 20) of the respondents were in agreement (whether or not they agreed with the factor);
and 2) an ‘agree’ was a composite score greater than 70 (70% of 100) and a ‘disagree’ score
112
was less than 50 (50% of 100). The composite score (group answers) were also considered to
be in either the agreement or disagreement range.
The highest and lowest composite scores generally represent the high level of consensus as
well as agreement or disagreement with the factor. Forty-six of 52 factors (88%) met the two
required conditions for achieving a consensus, as outlined above. Thirty-two factors (61%)
had high composite scores, with a high score in the agreement range (most experts agreed to
include the factor). Fourteen factors (27%) had low composite scores, with a high score in the
disagreement range (most experts agreed to exclude the factor). Only six factors (12%) did
not meet the two required criteria for consensus: height clearance, guidance signs, recreational
facilities, pedestrian traffic signals, bicycle speed, and bicycle volume (Table 2). Several other
statistical procedures, such as interquartile range and intra-class correlation coefficient (ICC),
were used for better analysis and to check a consensus for these six factors, and also to further
determine convergence of opinion for all factors in the next stages.
Using the composite scores, the factors were ranked in descending order of group agreement
with the factors. The factors agreed for inclusion were ranked from 1st to 32nd, and the
factors agreed for exclusion ranked from 39th to 52nd. The six factors that did not meet the
required criteria of consensus were ranked from 33rd to 38th based on the composite score
(Table 2). In the cases of factors having the same composite score, the ties were broken using
the interquartile ranges by giving the factors with the smaller interquartile range the higher
rank. If both composite scores and the interquartile ranges were equal, the factors were ranked
alphabetically.
Table 2 Composite scores of the third Delphi round

Experts Composite Highest number in
Factors Mean Rank
(N) score agreement
Footpath connectivity 4.85 20 97 1 20 (for inclusion)
Adequate footpath width 4.8 20 96 2 20 (for inclusion)
Footpath continuity 4.8 20 96 3 19 (for inclusion)
Shade 4.7 20 94 4 19 (for inclusion)
Pedestrian conflict points 4.65 20 93 5 19 (for inclusion)
Vehicle conflict points 4.45 20 89 6 17 (for inclusion)
Ease of movement 4.35 20 87 7 20 (for inclusion)
Lighting 4.25 20 85 8 19 (for inclusion)
Pedestrian volume 4.25 20 85 9 19 (for inclusion)
Footpath slope & grade 4.05 20 81 10 19 (for inclusion)
Pedestrian free manoeuvre 4.05 20 81 11 18 (for inclusion)
Pedestrian speed 4.05 20 81 12 17 (for inclusion)
Vehicle volume 4.05 20 81 13 18 (for inclusion)
Vehicle speed 4.05 20 81 14 16 (for inclusion)
Crossing opportunities 4 20 80 15 18 (for inclusion)
113
Activation of side facades 3.95 20 79 16 18 (for inclusion)
Crossing facility type 3.95 20 79 17 18 (for inclusion)
Quality of footpath surface 3.95 20 79 18 18 (for inclusion)
Trees and plants 3.95 20 79 19 16 (for inclusion)
Buffer width 3.9 20 78 20 18 (for inclusion)
Facilities for the disabled 3.9 20 78 21 18 (for inclusion)
Footpath maintenance 3.85 20 77 22 17 (for inclusion)
Permanent obstructions 3.85 20 77 23 16 (for inclusion)
Pedestrian-traffic separation 3.85 20 77 24 15 (for inclusion)
Footpath surveillance 3.75 20 75 25 16 (for inclusion)
Wayfinding guidance 3.75 20 75 26 16 (for inclusion)
Awnings 3.75 20 75 27 15 (for inclusion)
Ramp 3.75 20 75 28 15 (for inclusion)
Garbage facilities 3.75 20 75 29 14 (for inclusion)
Public transportation facilities 3.7 20 74 30 15 (for inclusion)
Obstruction width 3.65 20 73 31 13 (for inclusion)
Footpath cleanliness 3.6 20 72 32 13 (for inclusion)
Guidance signs 3.35 20 67* 33 11 (for inclusion)*
Height clearance 3.3 20 66* 34 5 (for inclusion)*
Pedestrian signs 3.3 20 66* 35 5 (for inclusion)*
Bicycle volume 3.1 20 62* 36 5 (for inclusion)*
Bicycle speed 2.95 20 59* 37 3 Both *
Recreational facilities 2.95 20 59* 38 1 (for exclusion)*
Bollards 1.5 20 30 39 17 (for exclusion)
Building articulation 1.35 20 27 40 17 (for exclusion)
Business presence 1.35 20 27 41 18 (for exclusion)
Number of traffic lanes 1.35 20 27 42 18 (for exclusion)
Benches 1.3 20 26 43 18 (for exclusion)
On-street parking 1.3 20 26 44 18 (for exclusion)
Fire hydrant 1.25 20 25 45 18 (for exclusion)
Frequency of lateral ways 1.2 20 24 46 18 (for exclusion)
Shoulder width 1.2 20 24 47 18 (for exclusion)
Toilet nearby 1.2 20 24 48 19 (for exclusion)
Drinking fountain 1.15 20 23 49 19 (for exclusion)
Enclosure 1.15 20 23 50 19 (for exclusion)
Vehicle lane width 1.15 20 23 51 19 (for exclusion)
Temporary obstructions 1.1 20 22 52 19 (for exclusion)
Relative importance index of the factors affecting PLOS

The numerical scores from the questionnaire responses provided an indication of the varying
degrees of influence that each factor has on PLOS, based on the viewpoint of these experts.
To further investigate the data, a relative importance index (RII) was used to rank the factors
according to their influences (Shash, 1993), calculated using the following formula:
Relative important index (%) =
Where (%) is the Relative Importance Index of each factor i, for each group of experts, k;
W is the weighting given to each factor by the experts and ranges from 1 (not at all important)
to 5 (extremely important) and = n1 + 2n2 + 3n3 + 4n4 + 5n5 (n is the number of experts who
114
scored each factor in each range; A = highest weight (five in our case); and N= total number
of sample (experts).
Figure 2 shows the relative important index (RII) for each factor affecting PLOS on footpaths
in commercial streets. Figure 4 shows the same information, but with each factor grouped and
themed according to the schema developed in our paper which was reviewed earlier (Raad &
Burke, 2018).
Figure 3 Factors ranked by relative importance index
The medians and interquartile ranges (IQR)

Medians and interquartile ranges are more appropriate for measuring the convergence of
opinion as long as the distribution of expert responses is expected to be highly skewed
(Flanders, 1988). A larger interquartile range indicated more variability in the scores for that
factor. When the median was two or less, or four or greater, the majority of respondents
appeared in agreement. The use of a measure of variance, such as the interquartile range with
the median, facilitated a more accurate conclusion as to agreement of the group. The
interquartile range (IQR) is a common method used in Delphi research and can be defined as
115
the measure of dispersion for the middle 50% of the observations (von der Gracht, 2012). It is
considered an objective and rigorous method for identifying the level of agreement or
consensus among sets of respondents. The consensus is usually considered to be reached
when there is IQR of 1 or less for 4 or 5 unit scales (Raskin, 1994; Rayens & Hahn, 2000),
which means that more than 50% of all opinions fall within 1 point on the scale (De Vet,
Brug, De Nooijer, Dijkstra & De Vries, 2004).
Table 3 reports the medians and interquartile ranges for rounds two and three responses by
factors. In 16 of the 52 factors (31%), the interquartile range in round two was greater than 1,
which indicates that consensus among experts (whether or not they agreed with the factor)
was not reached. These are highlighted in blue in Table 3. In 46 of the 52 factors (88%), the
interquartile range in round three was smaller than in round 2. In six cases (12%), the
interquartile range did not change from round two to round three, and for no factor did the
interquartile range increase. Smaller interquartile ranges in round three indicated decreasing
variance. The results indicated that the group answers were moving toward consensus. As
measured by interquartile ranges, 67.3% of the factors moved toward the zero, 17.3% moved
toward 0.75, and the rest moved towards 1. The responses were found to provide
comprehensive consensus from round two to round three in all factors (100%), as indicated by
the interquartile ranges value.
The level of agreement or consensus between participants is considered meaningless if group
stability is not achieved along with the consensus (von der Gracht, 2012). For this purpose,
Kendall's Coefficient of Concordance (W) (Brancheau & Wetherbe, 1987; Doke & Swanson,
1995; Schmidt, Lyytinen, Keil & Cule, 2001; Schmidt, 1997) was designed to: 1) better
interpret the reliability; 2) assess the consistency or conformity of responses between two or
more quantitative measurements; 3) identify overall agreement of importance (the 5-point
Likert scale) among experts; and 4) measure the stability of the responses between any
successive rounds (von der Gracht, 2012). The Kendall coefficient of concordance (W) is a
non-parametric statistic appropriate for ordinal variables, and it can be used for assessing the
agreement among multiple raters (Chaturvedi & Shweta, 2015). The ‘W’ score provides an
indication of whether consensus has been reached, or consensus is increasing between rounds,
and also the relative strength of agreement. It ranges from 0 (indicates no agreement) to 1
(indicates complete agreement) (von der Gracht, 2012). The W ≥ 0.7 is interpreted having a
strong agreement with high confidence in ranks (Schmidt, 1997).
116
3.2.3. Measuring the changes in consensus and stability
Overall agreement for importance ratings (W) of the 52 factors common for both rounds 2
and 3 were 0.31 and 0.75, respectively. The Delphi rounds can be stopped if either the
coefficient of concordance indicates a strong consensus (W>0.70), or the level of consensus
for the panel levels off in two successive rounds (Schmidt et al., 2001).
For the second round, a low level of agreement was found among the judges and consensus
was not reached because the W score was 0.31 and, therefore, it was necessary to conduct
another round. We can interpret the low consensus which occurred in this round as a result of
the massive number of factors affecting PLOS (52 factors) included in this study, with a large
number of experts who assessed these factors. In round 3, the raters were moving to higher
levels of agreement. Significant W means that the participants are applying essentially the
same standard in judging the importance of the issues. The level of reliability can be regarded
as excellent because the W value was greater than 0.70 for round three and found to be
statistically significant (at p < 0.001). General responses among experts were found to be very
stable as measured by W. The PedBikeTrans panel reached weak agreement (W=0.31) with
low confidence in their ranks in the second round, and had strong agreement with the high
confidence in ranks for the third round (W>0.70) according to Schmidt’s interpretation of
Kendall’s W. A W of 0.7, allowing for conclusions to be reached at the general level, which
indicates that the general consistency among the participants and stability in responses for
round 2 and round 3 were reached.
It is not common that further rounds would lead to perfect consensus because absolute
consensus would probably never be achieved. Based on the results obtained, it appeared that
there was little value in another round of the Delphi process as the results had achieved
sufficient reliability.
Table 3 Medians and interquartile ranges (IQR)

Round 1 Round 2 Change in IQR
Factors
Median IQR Median IQR R2-R1
Buffer width 4 1 4 0 -1
Crossing opportunities 4 1 4 0 -1
Quality of footpath surface 4 0.75 4 0 -0.75
Footpath slope & grade 4 1 4 0 -1
Adequate footpath width 5 1 5 0 -1
On-street parking Removed 2 Removed 0 -2
Pedestrian-traffic separation 4 2 4 0.75 -1.25
Ramp 4 1 4 0.75 -0.25
Shoulder width Removed 1 Removed 0 -1
117
Height clearance Removed 2 3 0.75 -1.25
Footpath connectivity 5 1 5 0 -1
Footpath continuity 5 1 5 0 -1
Frequency of lateral ways Removed 1 Removed 0 -1
Ease of movement 4 1 4 1 0
Permanent obstructions 4 1 4 0 -1
Temporary obstructions Removed 1 Removed 0 -1
Obstruction width 4 1 4 1 0
Awnings 4 1 4 0.75 -0.25
Benches Removed 1 Removed 0 -1
Bollards Removed 2 Removed 0 -2
Building articulation Removed 1 Removed 0 -1
Business presence Removed 1 Removed 0 -1
Crossing facility type 4 1 4 0 -1
Activation of side facades 4 2 4 0 -2
Facilities for the disabled 4 1.75 4 0 -1.75
Drinking fountain Removed 1 Removed 0 -1
Enclosure Removed 1 Removed 0 -1
Fire hydrant Removed 2 Removed 0 -2
Footpath cleanliness 4 1 4 1 0
Footpath maintenance 4 0 4 0 0
Footpath surveillance 4 1.75 4 0 -1.75
4
Garbage facilities Removed 1.75 1 -.075
(Reinstate)
Guidance signs 4 2 4 1 -1
Lighting 4 1 4 1 0
Recreational facilities Removed 1 Removed 0 -1
Shade 5 1 5 0.75 -0.25
Toilet nearby Removed 2 Removed 0 -2
Public transportation 4 2 4 0.75
-1.25
facilities
Trees and plants 4 1 4 0 -1
Wayfinding guidance 4 1 4 0 -1
Pedestrian free manoeuvre 4 1 4 0 -1
Pedestrian conflict points 4 1 5 1 0
Pedestrian speed 4 1 4 0 -1
Pedestrian signs 4 2 3 0.75 -1.25
Pedestrian volume 4 1.75 4 1 -0.75
Bicycle speed 4 1 3 0 -1
Bicycle volume 4 1 3 0.75 -0.25
Number of traffic lanes Removed 1.75 Removed 0 -1.75
Vehicle conflict points 4 1.75 5 0 -1.75
Vehicle lane width Removed 1 Removed 0 -1
Vehicle speed 4 1 4 0.75 -0.25
Vehicle volume 4 1 4 0 -1
3.2.4. On-street questionnaire survey

The following analysis is based on 312 responses, of which 131 (42%) were female and 181
(58%) were male (Table 4). The sample also skewed towards the young and educated, with
around a quarter of the sample being international visitors due to the large numbers of tourists
and international students who use some of the street segments surveyed. Sixty-one per cent
of respondents were aged 40 or less. Only one respondent had not completed high school;
118
38.5% had undergraduate degrees. The main purpose for travel to the Brisbane CBD was
39.1% for work and 14.7% for education. For the frequency of trip, 30.8% were almost every
day, 19.9% travel at least once per month and 21.5% were less than once per month. Table 8
shows the characteristics and percentage of the sample profiles of participants.
Table 4 Demographic and travel characteristics of the survey participants (sample = 312).
Characteristics % Characteristics %
Gender Trip purpose
Male 58 Work 39.1
Female 42 Education 14.7
Age Shopping 9.3
16-30 27.2 Eat a meal 4.8
31-40 34 Personal business 20.9
41-50 26 Recreation 10.6
Over 50 12.8 Other 0.6
Education level Trip frequency
Did not finish high school 0.3 Almost every day 30.8
Year 12 or equivalent 12.5 At least once per week 27.8
Trade/technical qualification 36.2 At least once per month 19.9
Undergraduate degree 38.5 Less than once per month 21.5
Postgraduate degree (Masters, PhD) 12.5
3.2.5. Factors’ weighting and satisfaction rating

As mentioned earlier, an on–street intercept questionnaire survey was then employed to gather
pedestrians' perceptions regarding the importance and satisfaction rating of the factors
affecting PLOS on the commercial street segments within the Brisbane CBD. Each identified
factor had different impacts on the overall PLOS condition. Also, a perceived importance was
that each factor makes a different contribution to both society and footpath environment,
depending on the pedestrian’s physical condition and characteristics, such as ethnicity, age,
and gender. Pedestrians’ perceptions at this stage reflect the level of importance and the
priority of improvement for each aspect of the footpath environment which contributes to an
increase or a decrease in pedestrian level of service.
First the relative importance or weight for each footpath element is generated using the
following equation:
is the important rating of each factor, IRs is the sum of importance rating for all factors, i is a
factor number (33 factors in this study), and j is the number of participants (312 participants
119
in this study). Table 5 shows the realistic weight (RWi) of each factor affecting PLOS on
footpaths in commercial streets, grouped and themed according to the schema developed in
our paper which was reviewed earlier (Raad & Burke, 2018).
Table 5 The realistic weight of importance level for the factors affecting PLOS
Normalized
Std Cumulative
Factors affecting PLOS Mean weight
dev. raw weight
(RWi)
Presence of adequate footpath width 4.96 0.230 1547 3.2
Presence of adequate crossing opportunities 4.95 0.261 1544 3.2
Characteristics
Provision of appropriate quality of footpath surface 4.95 0.302 1545 3.2

Geometric
Provision of appropriate footpath slope & grade 4.92 0.359 1536 3.1
Presence of adequate buffer area and width 4.94 0.293 1540 3.2
Presence of appropriate pedestrian-traffic separation 4.89 0.425 1526 3.1
Presence of adequate ramp 4.87 0.445 1519 3.1
Provision of an appropriate vertical clearance over footpath 3.04 1.035 950 1.9
Obstructions and
Presence of reasonable footpath connectivity 4.88 0.324 1523 3.1

Interruptions
Presence of adequate footpath continuity 4.96 0.200 1547 3.2

Provision of adequate ease of movement 4.96 0.230 1547 3.2
Absence of permanent obstructions 4.93 0.308 1539 3.2
Presence of adequate awnings 4.48 0.852 1398 2.9

Presence of adequate shade 4.95 0.290 1543 3.2
Provision of appropriate type of crossing facility 4.76 0.603 1485 3.0
Presence of appropriate activation of side facades 3.85 1.125 1201 2.5
Aesthetics and Amenities
Presence of adequate facilities for the disabled 4.96 0.274 1546 3.2
Provision of regular footpath maintenance 4.94 0.293 1540 3.2
Presence of adequate level of footpath cleanliness 4.97 0.238 1550 3.2
Provision of appropriate surveillance for footpath 3.60 1.225 1124 2.3
Presence of adequate garbage facilities 4.98 0.159 1554 3.2
Presence of adequate pedestrian signs 4.97 0.168 1551 3.2
Presence of adequate level of lighting 4.97 0.186 1551 3.2
Presence of appropriate public transportation facilities 4.97 0.168 1551 3.2
Presence of trees and plants 4.94 0.384 1541 3.2
Presence of adequate wayfinding guidance 4.92 0.290 1536 3.1
Provision of adequate level of ease for evasive movements 4.17 1.013 1300 2.7
Pedestrian
Traffic
Presence of minimum of potential pedestrian conflict points 4.94 0.293 1540 3.2
Provision of adequate freedom for selecting pedestrian speed 4.21 0.998 1313 2.7
Presence of reasonable pedestrian volume along travel line
4.91 0.394 1531 3.1
width
T
d
n
a
a
c
e
c
j
f
f
i
120
Absence of conflict points with vehicles 4.97 0.218 1551 3.2
Limitation of vehicle speed adjacent to footpath 4.91 0.356 1532 3.1
Presence of reasonable vehicle volume adjacent to footpath 4.92 0.392 1534 3.1
48835 100.0
Second, the level of pedestrian satisfaction for each aspect of the walking environment
represented by the factors was identified as a measurement of performance. The percentages
of respondents who rated performance of the same 33 factors as satisfied or strongly satisfied
were calculated and is presented in Table 7, ranked from worst (lowest percent agreeing the
aspect is being performed well) to best (highest percent). As may be seen from Table 6, ten
factors received a satisfaction rating below 50%.
4. Measuring PLOS vales using point system method

The proposed model for assessing PLOS was adopted. This model adopts a satisfaction
approach developed by (Khisty, 1994) using point system method. Factors were selected to
meet the need for detailed information about the footpath environment with a measurement of
PLOS. Each factor was built to assess individual elements of the walking environment with
respect to PLOS as well to overall PLOS value. Factors include sections on various aspects of
footpath features at microscale level for different types of walking environments.
4.1. Determining pedestrian level of service scale
to targeted footpaths. On the basis of LOS grades adopted in the HCM scale, a PLOS range is
allocated to one of the grades from A to F, in this study, based on a point system method.
These grades are related to the current situation for footpaths regarding the adequacy of
footpaths’ aspects in providing a high level of comfort, safety and mobility, and are
intrinsically related to all 33 factors affecting PLOS. The range for pedestrian level of service
values might be between 0 and 6 points, based on the current situation by the value of the
obtained points. Table 6 presents the footpath conditions, the proposed criteria, and the range
of scores, acceptability level and the required remedial actions assigned to each PLOS grade
from A to F. Based on a point system method, the PLOS grades can be defined based on a
different range of scores as follows:
 PLOS A: indicates that the footpath is in excellent condition, provides the highest level
of comfort, safety and mobility for pedestrians, and all the factors positively affect
PLOS. PLOS here can be assigned as being 90% or higher of the pedestrian
121
satisfaction rating for each factor.
 PLOS B: indicates that the footpath is in good condition, and few aspects of the
footpath environment (factors affecting PLOS) negatively affect pedestrian safety,
comfort and mobility. Limited procedures are needed for improvement. PLOS can be
assigned as being 80% - 90% of the pedestrian satisfaction rating and is considered to
be acceptable.
 PLOS C: indicates that the footpath is in regular condition, and some aspects of the
comfort and mobility. Some procedures and efforts are needed here for improvement.
PLOS here can be assigned as being 60% - 80% of the pedestrian satisfaction rating
and is considered to be moderately acceptable.
 PLOS D: indicates that the footpath is in basic condition, and significant aspects of the
comfort and mobility. Many procedures and efforts are needed here for improvement.
PLOS here can be assigned as being 40% - 60% of the pedestrian satisfaction rating
and is considered to be unacceptable.
 PLOS E: indicates that the footpath is poor condition, and too many aspects of the
comfort and mobility. Too much effort is needed here for improvement. PLOS can be
assigned as being 20% - 40% of the pedestrian satisfaction rating and is considered to
be highly unacceptable.
 PLOS F: indicates that the footpath is totally inappropriate for walking, and almost all
the aspects of the footpath environment (factors affecting PLOS) negatively affect
pedestrian safety, comfort and mobility. Total improvement and rehabilitation or
rebuilding are needed in this case. PLOS here can be assigned as 20% or less of the
pedestrian satisfaction rating and is considered to be totally unacceptable or invalid.
Table 6 Pedestrian level of service (PLOS) grade scale
PLOS
Conditions Proposed criteria Remedial action
Grade
A Excellent 90% or more satisfaction Very limited
B Good 80% - 90% satisfaction Limited
C Regular 60% - 80% satisfaction Medium
D Basic 40% - 60% satisfaction High
E Poor 20% - 40% satisfaction Unlimited
F Appalling 20% or less of satisfaction Rebuilding
122
4.2. Calculating PLOS value
Two main streets, including Adelaide and George Streets in Brisbane’s CBD, were chosen as
the setting for applying the methodology described in the previous section. A total of 8
footpath segments of about 2 km length in Adelaide Street and 4 footpath segments of about 1
km length in George Street were included to measure PLOS values in this study. To calculate
the PLOS grade, the overall PLOS score should first be calculated. The PLOS score for each
factor can be achieved by multiplying the points allocated to each factor by the relative
importance of that factor (Figure 3 & Table 6). Also, appropriate points can be allocated to
each factor based on the grade assigned for that factor. The grade was determined for each
factor based on the pedestrian satisfaction percentage obtained from the intercept survey and
according to the scale developed in Table 6. Figure 3 shows a comparison in PLOS grade
between all factors affecting PLOS. The overall PLOS score was calculated by adding all
PLOS scores for all the factors. Then, the overall PLOS score was allocated to an appropriate
grade from A to F based on the PLOS score result. The overall PLOS score of 4 as shown in
Table 7 reflects PLOS equal to C.
Figure 3 Pedestrian level of service (PLOS) grade for each aspect of the walking environment
123
Table 6 Pedestrian level of service (PLOS) grade
Satisfaction PLOS Allocated PLOS
Category Factors affecting PLOS RWi %
ratings grade points score
Footpath width 20.5% E 2 0.032 0.064
Crossing opportunities 61.5% C 4 0.032 0.128
Characteristics
Quality of footpath surface 92.0% A 6 0.032 0.192

Geometric
Footpath slope & grade 87.0% B 5 0.031 0.155

Buffer area and width 6.7% F 1 0.032 0.032
Pedestrian-traffic separation 52.5% D 3 0.031 0.093
Ramp 85.0% B 5 0.031 0.155
Clearance over footpath 96.5% A 6 0.019 0.114
Obstructions and
Footpath connectivity C 4 0.031 0.124

Interruptions
67.3%
Footpath continuity 89.7% B 5 0.032 0.16
Ease of movement 56.4% D 5 0.032 0.16
Permanent obstructions 37.5% E 3 0.032 0.096
Awnings 96.0% A 6 0.029 0.174

Shade 49.3% D 3 0.032 0.096
Type of crossing facility 76.6% C 4 0.03 0.12
Activation of side facades 94.5% A 6 0.025 0.15
Facilities for the disabled 53.5% D 3 0.032 0.096

Footpath maintenance 97.0% A 6 0.032 0.192
Footpath cleanliness 96.0% A 6 0.032 0.192
Footpath surveillance 91.6% A 6 0.023 0.138
Garbage facilities 93.2% A 6 0.032 0.192
Guidance signs 82.0% B 5 0.032 0.16
Lighting 37.5% E 3 0.032 0.096
Public transportation 94.5% A 6 0.032 0.192
Trees and plants 11.8% F 1 0.032 0.032
Wayfinding guidance 50.0% D 3 0.031 0.093
Pedestrian
Evasive movements 43.9% D 3 0.027 0.081

Traffic
Pedestrian conflict points 66.6% C 4 0.032 0.128

Pedestrian speed 48.4% D 3 0.027 0.081
Pedestrian volume 23.0% E 2 0.031 0.062
Adjacent
Traffic
Vehicle conflict points 77.2% C 4 0.032 0.128

Vehicle speed 64.4% C 4 0.031 0.124
Vehicle volume 11.8% F 1 0.031 0.031
PLOS Value C 4.00
124
5. Results
Inspection of the results shows that the factors affecting PLOS can generally be themed and
grouped into five categories (Figure 4). It is also possible to suggest which factors are most
important based on experts’ viewpoints using the composite scores. In general, the factors
with very high influence are considered those with composite scores above 0.800. This
includes the factors from rank 1 (footpath connectivity) to rank 15 (crossing opportunities).
Second, the factors with average to high influence are considered with RIIs lying between
0.600 and 0.800. This category includes the factors from rank 16 (activation of side facades)
to rank 36 (bicycle volume). Factors with low influence have RIIs less than 0.600. The results
of the survey provided the level of service value for a total of 12 footpath segments on
Adelaide and George Streets in Brisbane’s CBD. The overall grand total score of 4 as shown
in Table 7 truly shows that the level of service is equal to C level. The main value in using
PLOS measurement is to ensure that different pedestrian facilities are balanced with vehicular
facilities and other land uses. Furthermore, the PLOS scale is considered to be a specific
method for checking the conditions of pedestrian facilities and infrastructure as well as for
evaluating the quality of service. Pedestrian LOS also reflects users' perceptions in terms of a
sense of comfort and safety. PLOS scores can be considered as a basis for selecting or
developing pedestrian facilities and for identifying how well a facility or service operates
from a user’s perspective. It also aids in systematically assessing pedestrians’ needs and in
prioritising improvements, as well strengthening objectivity in the decision making process. It
is necessary to objectively quantify how well footpaths in commercial streets accommodate
pedestrians’ trips, and to quantify pedestrians’ perception of safety, comfort and mobility.
125
Figure 4 Factors and relative importance index grouped by type
126
6. Discussion
This paper has demonstrated a novel, robust method for determining and measuring the
factors that affect PLOS using a literature review, an online survey and a ‘walkshop’ as part of
an inter-linked Delphi process, a pedestrian intercept survey and a point system method.
Applied here on footpaths in commercial streets in Brisbane, the approach and methods could
be used across a range of different facility types and in differing urban contexts.
Measuring PLOS in evaluating footpaths using this approach can lead to many advantages.
First, PLOS can be used as an indicator of the current situation of footpaths and other
pedestrian facilities. Also, it could work well as a tool for creating a clear image to guide
decision makers in evaluating, monitoring and comparing the performance and quality of
pedestrian facilities. The methodology described can be most useful in allocating the
budgeting funds for facility changes and improvements. Further, adopting this approach for
measuring PLOS using qualitative and quantitative factors affecting the quality of the walking
environment can supplement the results obtained through the quantitative PLOS of flow,
speed, and density, as elaborated in the US Highway Capacity Manual (HCM) approach.
Although it is important to evaluate the quality of the walking environment using both the
qualitative and quantitative factors that contribute to the walking experience and perceived
PLOS, there are no guidelines which have been adopted, or recommended methodology given
by HCM, on how to measure or use these factors in designing and assessing pedestrian
facilities. Applying the described methodology can also be most useful in assessing the
operating characteristics and providing the qualitative level of service of the facility being
assessed, as perceived by pedestrians in real situations.
Though the results obtained may look impressive, they do not yet take into account the
perceptions of a broader range of pedestrians. The experts involved in the Delphi process
were predominantly white and middle class, and, at least in the initial survey, were
predominantly male. All were physically able to walk long distances. Though experts in this
field, they are not reflective of the general population. To overcome this limitation, more
verification of the described approach in this paper is needed by applying it over various and
long periods of time to a broad range of users, including residents of all ages and
requirements and visitors in the same city streets in Brisbane, to gain further insights into the
validity and ranking of these PLOS factors.
Unresolved research challenges in the field of PLOS remain. These include: further research
into specific factors, such as pedestrian connectivity, shade and lighting to determine better
127
metrics that are more readily used in PLOS frameworks; a wider research base, and in
particular, the extension of PLOS measurement to developing nations, given the
overwhelming proportion of the research literature that is from North America; ensuring
PLOS tools have meaningful inter-rater reliability so that we can be sure that when someone
is trained to undertake PLOS measurements with a given tool, we will obtain robust outputs;
undertaking similar research as shown in this paper explicitly with and for disabled and
special needs pedestrians and identifying if and how the use of PLOS measures for the
general population may affect the disabled; and studies of the in-situ use of PLOS tools by
transport agencies and local authorities to reveal their impact.
References
Aghaabbasi, M., Moeinaddini, M., Shah, M. Z., Asadi-Shekari, Z., & Kermani, M. A. (2018).
Evaluating the capability of walkability audit tools for assessing sidewalks. Sustainable Cities
and Society, 37, 475-484.
Asadi-Shekari, Z., Moeinaddini, M., Aghaabbasi, M., Cools, M., & Shah, M. Z. (2019). Exploring
effective micro-level items for evaluating inclusive walking facilities on urban streets (applied
in Johor Bahru, Malaysia). Sustainable Cities and Society, 49, 101563.
Bedimo-Rung, A. L., Gustat, J., Tompkins, B. J., Rice, J., & Thomson, J. (2006). Development of a
direct observation instrument to measure environmental characteristics of parks for physical
activity. Journal of Physical Activity & Health, 3, S176.
Boodlal, L. (2004). Accessible sidewalk and street crossing-AN informational guide. Retrieved from
Brancheau, J. C., & Wetherbe, J. C. (1987). Key issues in information systems management. MIS
quarterly, 23-45.
Chaturvedi, H., & Shweta, R. (2015). Evaluation of inter-rater agreement and inter-rater reliability for
observational data: An overview of concepts and methods. JIAAP, 41, 20-27.
Cunningham, G. O., Michael, Y. L., Farquhar, S. A., & Lapidus, J. (2005). Developing a reliable
senior walking environmental assessment tool. American Journal of Preventive Medicine,
29(3), 215-217.
Dannenberg, A. L., Cramer, T. W., & Gibson, C. J. (2005). Assessing the Walkability of the
Workplace: A New Audit Tool. American Journal of Health Promotion, 20(1), 39-44.
doi:doi:10.4278/0890-1171-20.1.39
Day, K., Boarnet, M., Alfonzo, M., & Forsyth, A. (2006). The Irvine–Minnesota inventory to measure
built environments: development. American Journal of Preventive Medicine, 30(2), 144-152.
De Vet, E., Brug, J., De Nooijer, J., Dijkstra, A., & De Vries, N. K. (2004). Determinants of forward
stage transitions: a Delphi study. Health Education Research, 20(2), 195-205.
Doke, E. R., & Swanson, N. (1995). Decision variables for selecting prototyping in information
systems development: A Delphi study of MIS managers. Information & Management, 29(4),
173-182.
Ewing, R., & Handy, S. (2009). Measuring the Unmeasurable: Urban Design Qualities Related to
Walkability. Journal of Urban Design, 14(1), 65-84. doi:10.1080/13574800802451155
Flanders, F. B. (1988). Determining curriculum content for nursery/landscape course work in
vocational agriculture for the twenty-first century: A futures study utilizing the Delphi
technique. (8823794 Ed.D.), University of Georgia, Ann Arbor. ProQuest Dissertations &
Theses Global database.
Fruin, J. J. (1970). Designing for Pedestrians, a Level of Service Concept. Mechanical Engineering,
92(12), 63-&.
128
Hess, F., Salze, P., Weber, C., Feuillet, T., Charreire, H., Menai, M., . . . Oppert, J.-M. (2017). Active
Mobility and Environment: A Pilot Qualitative Study for the Design of a New Questionnaire.
PLoS One, 12(1), e0168986.
Jaskiewicz, F. (2000). Pedestrian level of service based on trip quality. Transportation Research
Circular, TRB.
Khisty, C. J. (1994). Evaluation of pedestrian facilities: beyond the level-of-service concept.
Transportation Research Record(1438), 45-50.
Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and reporting intraclass correlation
coefficients for reliability research. Journal of chiropractic medicine, 15(2), 155-163.
Landis, B., Vattikuti, V., Ottenberg, R., McLeod, D., & Guttenplan, M. (2001). Modeling the roadside
walking environment: pedestrian level of service. Transportation Research Record: Journal of
the Transportation Research Board(1773), 82-88.
Lawshe, C. H. (1975). A quantitative approach to content validity. Personnel psychology, 28(4), 563-
575.
Lee, B. J., Jang, T. Y., Wang, W., & Namgung, M. (2009). Design criteria for an urban sidewalk
landscape considering emotional perception. Journal of Urban Planning and Development,
135(4), 133-140.
Mateo-Babiano, I. (2016). Pedestrian's needs matter: Examining Manila's walking environment.
Transport Policy, 45, 107-115.
Pikora, T., Giles-Corti, B., Bull, F., Jamrozik, K., & Donovan, R. (2003). Developing a framework for
assessment of the environmental determinants of walking and cycling. Social Science &
Medicine, 56(8), 1693-1703.
Pikora, T. J., Bull, F. C. L., Jamrozik, K., Knuiman, M., Giles-Corti, B., & Donovan, R. J. (2002).
Developing a reliable audit instrument to measure the physical environment for physical
activity. American Journal of Preventive Medicine, 23(3), 187-194.
doi:http://dx.doi.org/10.1016/S0749-3797(02)00498-1
Removed because it is a self-reference.
Raskin, M. S. (1994). The Delphi study in field instruction revisited: Expert consensus on issues and
research priorities. Journal of Social Work Education, 30(1), 75-89.
Rayens, M. K., & Hahn, E. J. (2000). Building consensus using the policy Delphi method. Policy,
politics, & nursing practice, 1(4), 308-315.
Saito, Y., Oguma, Y., Inoue, S., Tanaka, A., & Kobori, Y. (2013). Environmental and individual
correlates of various types of physical activity among community-dwelling middle-aged and
elderly Japanese. International journal of environmental research and public health, 10(5),
2028-2042.
Schmidt, R., Lyytinen, K., Keil, M., & Cule, P. (2001). Identifying software project risks: An
international Delphi study. Journal of management information systems, 17(4), 5-36.
Schmidt, R. C. (1997). Managing Delphi surveys using nonparametric statistical techniques. decision
Sciences, 28(3), 763-774.
Seagle, E. D., & Iverson, M. (2001). Characteristics of the turfgrass industry in 2020: a Delphi study
with implications for agricultural education programs. Citeseer.
Shash, A. A. (1993). Factors considered in tendering decisions by top UK contractors. Construction
Management and Economics, 11(2), 111-118. doi:10.1080/01446199300000004
Smith, A. L., Troped, P. J., McDonough, M. H., & DeFreese, J. (2015). Youth perceptions of how
neighborhood physical environment and peers affect physical activity: a focus group study.
International Journal of Behavioral Nutrition and Physical Activity, 12(1), 80.
Spittaels, H., Foster, C., Oppert, J.-M., Rutter, H., Oja, P., Sjöström, M., & De Bourdeaudhuij, I.
(2009). Assessment of environmental correlates of physical activity: development of a
European questionnaire. International Journal of Behavioral Nutrition and Physical Activity,
6(1), 39. doi:10.1186/1479-5868-6-39
Sung, H., Go, D., Choi, C.-g., Cheon, S., & Park, S. (2015). Effects of street-level physical
environment and zoning on walking activity in Seoul, Korea. Land Use Policy, 49, 152-160.
Troped, P. J., Cromley, E. K., Fragala, M. S., Melly, S. J., Hasbrouck, H. H., Gortmaker, S. L., &
Brownson, R. C. (2006). Development and reliability and validity testing of an audit tool for
129
trail/path characteristics: the Path Environment Audit Tool (PEAT). Journal of Physical
Activity & Health, 3, S158.
Van Cauwenberg, J., Van Holle, V., Simons, D., Deridder, R., Clarys, P., Goubert, L., . . . Deforche,
B. (2012). Environmental factors influencing older adults’ walking for transportation: a study
using walk-along interviews. International Journal of Behavioral Nutrition and Physical
Activity, 9(1), 85.
von der Gracht, H. A. (2012). Consensus measurement in Delphi studies: Review and implications for
future quality assurance. Technological Forecasting and Social Change, 79(8), 1525-1536.
doi:https://doi.org/10.1016/j.techfore.2012.04.013
Weliange, S. H. D. S., Fernando, D., & Gunatilake, J. (2014). Development and validation of a tool to
assess the physical and social environment associated with physical activity among adults in
Sri Lanka. BMC Public Health, 14(1), 423.
Williams, J. E., Evans, M., Kirtland, K. A., Cavnar, M. M., Sharpe, P. A., Neet, M. J., & Cook, A.
(2005). Development and use of a tool for assessing sidewalk maintenance as an
environmental support of physical activity. Health Promotion Practice, 6(1), 81-88.
130
6 Developing pedestrian level of service (PLOS) model for footpaths in
commercial streets using pedestrian disgruntlement approach
An earlier version presented in Chapter 6 was first introduced as:
 Raad, N., & Burke. M. (2018). Investigating relationship between the importance of
the factors affecting PLOS and level of satisfaction on footpaths. Paper presented at
the 15th World Conference on Transport Research (WCTRs), May 26-31 2019,
Mumbai, India.
I elected that this publication not be published in WCTR’s conference proceeding. A revised
and much improved version was then submitted to the journal of “Travel Behaviour and
Society” as:
 Raad, N., & Burke. M., Developing Pedestrian Level of Service (PLOS) model for
footpaths in commercial streets using pedestrian disgruntlement.
 First Submission on 26/08/2019.
6.1 Research summery

The previous chapter focused on the research efforts in order to systematically select the
appropriate factors for the Australian context which were the main influencers of the PLOS
on footpaths at commercial areas adopting the Delphi process. In line with the previous
chapter, this paper summarizes the results of Task 2 including step 2 (in detail), and Task 3
including steps 1, 3 and 4. The main focus of this chapter is to examine the importance level
of a wide range of factors affecting PLOS, and their relationship to each other, and to explore
how this relationship may differ for different ages and genders of pedestrians in different
walking environments. Another focus of this study is to identify both the level of pedestrian
disgruntlement and PLOS for each factor based on pedestrians’ perceptions. The methods
used to achieve the aims of this study are a pedestrian intercept survey, a disgruntlement
approach, a point system method and a gap analysis approach.
The outputs of the previous chapter (the results of the Delphi process) were used as inputs to a
survey of pedestrians. An on-street intercept questionnaire survey was employed to gather
pedestrians' perceptions of the factors affecting PLOS on these same commercial street
segments within Brisbane’s CBD. Three hundred and twelve (312) respondents aged 16 and
over were randomly selected to ensure that the perceptions of a wide range of people
regarding the importance and satisfaction of factors affecting PLOS on footpaths were
131
captured. Respondents were asked to rate both the satisfaction (how satisfied) and the
importance (how important good performance is) using a five-point Likert scale for each of
the factors in relation to these footpaths in the study area. The outputs of this survey were
used to generate the relative importance and satisfaction level for each aspect of the footpath
environment. The key advantage of this step is that perceptions which were obtained are
based on walking experiences in real situations. The results of this survey indicate that the
perceived importance of each factor makes a different contribution to both society and the
footpath environment, depending on the pedestrian’s physical condition and characteristics,
such as ethnicity, age, and gender. Pedestrians’ perceptions at this stage reflect the level of
importance and the priority for improvement of each aspect of the footpath environment that
contributes to an increase or a decrease in the pedestrian level of service. Furthermore, the
scores which were given by pedestrians are considered as the basis for the production of a
new model for PLOS based on pedestrians' perceptions.
Based on the outputs of the pedestrian intercept survey, a new PLOS model was produced by
combining measures of pedestrian satisfaction with each factor and perceived importance of
that factor, creating a novel measure we label ‘pedestrian disgruntlement’ to measure PLOS.
A disgruntlement approach, together with a point system method, offers a much more robust
means of obtaining PLOS measures. Pedestrian disgruntlement was first determined by cross-
tabling the level of importance of the features of the walking environment represented by the
factors affecting PLOS and the level of satisfaction for each factor. Then the PLOS model
was built by allocating a point to each factor, thereby affecting the PLOS using a point system
method based on percentage of pedestrian disgruntlement.
The results of applying this model show that although the majority of the aspects of the
walking environment were satisfied by pedestrians with PLOS scores of A or B, pedestrians,
including disabled people, should walk within a narrow width in comparison to the extreme
pedestrian volume, especially during peak hours. Footpath width and buffer area were
assigned as a low level of service value by pedestrians of different ages and genders within
the category of geometric characteristics. However, elderly people complained more about the
lack of obvious pedestrian-traffic separation. Absolute consensus occurred among different
ages and genders of pedestrians regarding footpath connectivity, which occupied the lowest
value, with a PLOS score of C. Also, older users recorded their disgruntlement regarding the
permanent obstructions factor, with the lowest value of a PLOS score of C in comparison with
others. PLOS for facilities for the disabled obtained the worst value by elderly users in
132
comparison to other users, reflecting the need for improvements that cater for older people.
Elderly people and females assigned the lighting and trees and plants factors as occupying the
PLOS scores of C and D, respectively. The overall PLOS score shows that the condition of
measuring segments was better than moderate, reflecting pedestrians’ perceptions.
Finally, the modified gap analysis approach, which has previously been used to measure
consumer perceptions of quality of service (QOS) in the service industries was applied in
order to identify the aspects of the walking environment most in need of improvement. This
study provides a more realistic model in identifying PLOS value than the satisfaction model
because it takes into consideration the importance level in identify the satisfaction percentage
of the factors affecting PLOS.

Primary questions:
RQ3: What do pedestrians think about the importance level of a wide range of factors
affecting PLOS?
Sub-questions:
developed and how can these differing age groups and genders be explored?

The study is the first to introduce a disgruntlement approach in combination with a pedestrian
intercept survey, a point system method and a gap analysis approach in producing a PLOS
model. This approach was applied on footpaths in commercial streets in Brisbane, however
the approach and methods could be used across a range of different facility types and in
differing urban contexts. The main methodological contribution of this study is that it uses the
six-step approach in harnessing the gap analysis technique used in other fields and applies it
for use in PLOS measurement for the first time. As noted in our previous systematic review
133
paper (Raad & Burke, 2018), previous PLOS measurement approaches relied heavily on
experts’ judgements alone, often of only one or two experts, and had little in the way of
testing with ordinary pedestrians. The methodological contributions of this Chapter can be
summarised as follows:
 The described method can be used for different transport services and modes, and
different trip purposes.
 The method can help analyse dissatisfaction and importance for different population
sub-groups.
 The described method provides a visual inspection of how dissatisfaction and
importance are distributed amongst the various PLOS measures. This method then
identifies the weaknesses in level of service of any aspect of targeted footpaths.
 This study opens pathways for others to further refine and improve on PLOS
measures.
 The technique of gap analysis allows the determination of the perceived ‘quality gaps’
of each aspect of a walking environment. It is not difficult to conceive how the
approach could be used with cyclists and others.

 The outcome of this study can reflect the perceived safety or comfort of pedestrians
along footpath segments, and can be objectively used to collect and measure factors
influencing pedestrian LOS.
 This study provides a reliable method of evaluating and prioritising the needs for
footpath retrofit construction on existing streets and also for designing new streets.
 The proposed methodology described may be used as a guide across a range of
different facility types and in differing urban contexts for future studies by planners,
designers and transportation engineers.
 There are also many applied contributions in this study that are of use to city
authorities in Brisbane in terms of planning the city’s streets and footpaths, and the
urban environment of commercial streets.
 The study identifies a range of factors that pedestrians are disgruntled with in the
Brisbane CBD, including buffer width, vehicle volume, presence of trees and plants,
adequate footpath width and pedestrian volume.
134
This chapter includes a co-authored paper that summarises the results of Task 2 including step
1 and step 2 (in detail in step 2), and Task 3 including steps 1, 3 and 4. The bibliographic
details of the co-authored paper, including all authors, are:
 Raad, N., & Burke. M., “Developing Pedestrian Level of Service (PLOS) model for
footpaths in commercial streets using pedestrian disgruntlement”.
execution or interpretation of at least that part of the publication in their field of
expertise;
requirements.

Nowar Raad  Designed the data collection process for the Delphi technique, including the site
selection, participant recruitment, and questionnaires.
 Visited the study area to conduct the walking course and workshop with experts.
 Managed and guided the panel of experts in both the walking course and the
workshop.
 Designed the data collection process for the pedestrian intercept survey, including
the site selection, participants’ recruitment, and questionnaires.
 Complied, cleaned and analysed the data.
reviewers.
 Prepared presentation slides to be presented at the conference.
 Responsible for revisions made at the suggestion of the journal reviewers and
resubmitting the paper.
 Provided assistance in the synthesis and interpretation of the results.
 Provided assistance in preparing the presentation slides and presented the paper at
the conference.
 Visited in the study area to conduct the walking course and workshop with experts.
135
(Signed) _________________________________ (Date) 26/08/2019
(Signed) __________________________________ (Date) 26/08/2019
136
Developing pedestrian level of service (PLOS) model for footpaths in commercial streets
using pedestrian disgruntlement approach
Abstract
Pedestrians are important stakeholders in urban streets and their safety and satisfaction is a
priority for decision makers. Therefore, the development of pedestrian level of service
(PLOS) measures has received considerable academic and practitioner attention, as transport
agencies seek to encourage walking. There remains dispute as to what factors should be
included in PLOS measurement. This study had a set of inter-related aims: to examine the
importance level of a wide range of factors affecting PLOS and their relationship to each
other; to explore how this relationship may differ for different ages and genders of pedestrians
in different walking environments; and, to identify both the level of pedestrian disgruntlement
and PLOS for each factor. We used the gap analysis approach, which has previously been
used in measuring consumer perceptions of Quality of Service (QOS) in the service
industries. The methods included a semi-Delphi survey and ‘walkshops’ with Brisbane
professionals involved in active transport facility design and planning, and intercept surveys
with 312 pedestrians. We have developed new methods to combine measures of pedestrian
satisfaction with each factor and perceived importance of that factor, creating a novel measure
we label ‘pedestrian disgruntlement’ to measure PLOS. The new measure was determined by
cross-tabling the level of importance of the features of the walking environment represented
by the factors affecting PLOS, and the level of satisfaction for each factor. The new PLOS
model was built based on our new approach called ‘pedestrian disgruntlement’, and a point
system method. We applied this approach to a case study site covering a set of footpaths on
commercial streets in the central business district of Brisbane, Australia. However, this
approach method could be used across a range of different facility types and in differing urban
contexts. The results demonstrate how the technique may be used to improve PLOS
measurement and to allow comparison of service to satisfy particular population sub-groups.
Keywords: pedestrian level of service; pedestrian satisfaction; systematic quantitative review;
Delphi technique; disgruntlement percentage
137
1. Introduction
Walking is now considered an important part of urban transport in more and more cities
(Hidayat, Choocharukul & Kishi, 2012). The increased attention being paid to pedestrian
facilities is indicative of increased policy and planning interest in walking provision (Raad &
Burke, 2018). To help transport agencies, various models, tools and techniques for the
planning, designing and assessing of pedestrian facilities have been developed (Hidayat,
Choocharukul, Nakatsuji & Kishi, 2009). Pedestrian trip-making is influenced by
environmental design and urban form (Guo et al., 2014). The impact of the built environment
on pedestrians has been the subject of much research during the past decade to identify factors
that can help get more people walking more often, providing benefits to the transport system,
and to health and sustainability. Different types of walking have been linked with various
factors affecting walkability (Forsyth et al., 2008). The discovery of these factors has led to
the creation of specific methods and tools for measuring walkability (Saelens & Handy, 2008).
However, it has been argued that there are many factors that influence pedestrian behaviour in
the built environment that are yet to be discovered (Praveen K. Maghelal & Cara Jean Capp,
2011).
The literature on environmental factors enabling walking is ever increasing (Ewing et al.,
2014). Urban designers have been especially interested in the environmental qualities of
places that make them better for walking, not only as settings for physical activity, but also as
sensorial and social settings (Mehta, 2008). Research has pointed out how different situations
and various densities can influence people’s behaviour in public spaces (Kerridge et al.,
2001). The completeness of footpath systems has a positive association with pedestrian
volumes and the inclusion of off-street dedicated paths can also encourage walking (Frank et
al., 2003). Other features of policies that improve pedestrian amenity include street lighting,
footpath continuity, and improved aesthetics of the street area (Heath et al., 2006).
Combining all these and associated factors into a set of PLOS measures is challenging. One
must begin with the identification of the set of potentially important factors affecting level of
service as perceived by pedestrians (Raad & Burke, 2018). A level of service is defined as “a
quantitative stratification of a performance measure or measures that represent quality of
service’’ (L. Kang et al., 2013, p. 11). Gallin (2001) defined pedestrian level of service as “an
overall measure of walking condition on route, path and facility. Pedestrian LOS reflects
users' perception in terms sense of comfort and safety” (Gallin, 2001, p. 48). It is important to
gather a variety of information about users’ perceptions and consider the various requirements
138
of different groups of pedestrians which can affect design decisions. Changing the street
design for one group may reduce or enhance other individuals’ capabilities. Therefore,
footpaths must be designed to create accessibility and comfort for the majority of users
(Boodlal, 2004). To do this one must include a number of qualitative factors, in addition to the
traditional volume and capacity factors that were first adopted as PLOS measures (Jaskiewicz,
2000). Walking behaviour is affected by pedestrians’ physical, physiology and psychology
characteristics (Hidayat, Choocharukul & Kishi, 2010). As such, pedestrians’ psychological
characteristics, such as perceptions regarding footpath facilities, the street environment and
traffic, should also be taken into consideration (Hidayat et al., 2010).
This study had a set of inter-related aims: to examine the importance of a wide range of
factors affecting PLOS; to examine the level of pedestrian satisfaction for each factor in a
specific environment; to develop a new PLOS model based on a pedestrian disgruntlement
approach and to explore how these differed across age groups. A Delphi technique with a
panel of experts was adopted, followed by a survey of ordinary, everyday pedestrians. The
Delphi technique is a process of deriving group consensus for any topic when the required
information is not available using a systematic approach (Pikora et al., 2003). The main
features of the survey in the Delphi technique are an anonymous, written, multi-stage survey
process and controlled feedback (von der Gracht, 2012). Good practice application of the
technique requires sound selection of the expert panel, finalizing the number of participants,
carefully formulating the questionnaires, selecting strong consensus formation processes and
criteria; and statistically processing the answers (Sousa, Sequeira & Ferré-Grau). The Delphi
technique was used to help refine a list of factors identified in a recently published systematic
quantitative literature review (Raad & Burke, 2018). The outputs were then used as inputs in a
survey of pedestrians. This approach offers a much more robust means of obtaining PLOS
measures.
The paper is structured as follows. In Section 2 the theories and methodologies of both
pedestrian level of service (PLOS) and user satisfaction found in the literature are outlined,
providing the key conceptual and theoretical framing for the paper. The limitations of the
current assessment methods on the factors affecting pedestrian satisfaction are also identified
in this section. In Section 3 the methodology and computation used in this study are explained
in detail. In Section 4 the results of this study are discussed.
139
2. Background
2.1. Pedestrian level of service (PLOS)
To assess and calculate walking conditions, the most common method uses the PLOS.
Various approaches were determined by examining previous studies in order to select the
most effective indicators to evaluate PLOS. Two general approaches were used to evaluate
PLOS: the first was a capacity based model and the second was a roadway characteristic
based model. Some previous studies have attempted to use combination models to evaluate
PLOS while other studies have relied on the use of a single approach (Z. Asadi-Shekari et al.,
2013).
Capacity based model studies rely on the criteria of highway capacity, which was later
modified and improved to be suitable to evaluate pedestrian facilities. This approach was used
in the planning of pedestrian facilities, but information regarding acceptability by pedestrians
was insignificant (Singh & Jain, 2011). Fruin (1971) proposed the first attempt of a PLOS
model based on footpath capacity and pedestrian volume (Fruin, 1971). In subsequent years
many researchers adapted Fruin’s approach with some additions or modifications. Pushkarev
and Zupan (1975) defined the level of service for walking as essentially the same as Fruin's
LOS standards (Pushkarev & Zupan, 1975). Polus et al. (1983) also suggested a range of LOS
definitions for pedestrian traffic. Their calibrated and evaluated one- and three-regime linear
speed-density regression models were used to analyse the properties and characteristics of
pedestrian flow on footpaths (Polus et al., 1983). Fruin’s LOS principal as a design standard
was used by Tanaboriboon and Guyano (1989) to suggest an LOS method for the design of
footpaths in Bangkok (Tanaboriboon & Guyano, 1989). In addition, Jaskiewicz (2000) went
further than just the study of volumes and capacities, and evaluated PLOS based on trip
quality (Jaskiewicz, 2000).
Roadway characteristics based methods are based on the characteristics and features of the
environment of footpaths or pedestrian facilities. Pedestrians’ perceptions are essentially used
in these methods and the comfort level of pedestrians while encountering certain roadway
characteristics is determined. A milestone in PLOS research was made by Lautso and Murole
(1974) through describing the influence of environmental factors on pedestrian facilities
(Muraleetharan et al., 2005). Sarkar (1993) and Khisty (1994) developed a PLOS model
which qualitatively defined pedestrian environmental features rather than quantifying them to
evaluate PLOS in streets (Sarkar, 1993) (Khisty, 1994). In Sarkar's method, it is difficult to
measure each factor affecting PLOS in a practical field; especially when each factor is related
140
to the other. While Khisty’s method is inexpensive to use and can supplement the qualitative
LOS of the facility on a point scale, it is also considered to be difficult to use because the
results of these scales are not always easy to interpret, and the scaling systems may not
represent pedestrians’ perceptions. Dixon proposed a PLOS evaluation criterion that involves
the provision of basic facilities, conflicts, amenities, motor vehicle LOS, maintenance and
travel demand management, and multimodal provisions (Dixon, 1996). Although the LOS
ratings describe the degree to which facility provisions encourage pedestrian use, the
drawback of this model was that the weighting of factors affecting users was selected
arbitrarily. Sarkar (2003) suggested another PLOS model based only on the key attributes of
comfort that are used to qualitatively grade the physical, psychological and physiological
comfort levels of footpaths (Sarkar, 2003). Although this model has dealt with the micro-level
conditions of comfort for pedestrians, it drew from a small number of standards and factor
details for different street indicators, hence, this model did not provide a comprehensive
evaluation of inclusive facilities.
Another pedestrian LOS model was introduced by Landis et al. (2001), which focused on
determining the factors that may significantly influence a pedestrian's feeling of safety or
comfort on roadway segments between intersections (Landis et al., 2001). The gathering of
these factors into a mathematical expression, tested for statistical reliability, provides a
measure of the roadway segment’s PLOS. This approach is similar to the methods used to
assess the automobile operators’ LOS produced in the HCM. Gallin (2001) developed a
unique model for evaluating PLOS based on several qualitative and quantitative factors that
were classified into physical characteristics, location factors and user’s factors (Gallin, 2001).
The drawback of Gallin’s model is that the weighting of factors affecting PLOS is based on
personal decisions - the factors are not representative of all aspects of the walking
environment. Asadi-Shekari and Zaly-Shah (2011) discussed PLOS using several indicators
from various guidelines which meet different pedestrian needs and determine the failures of
existing streets. This model is useful for improving existing streets and produces a complete
guideline for designing new streets. However, all of the factors were determined only from
previous studies and guidelines based on personal decisions, and the users' perceptions were
not taken into account (Asadi-Shekari & Zaly-Shah, 2011). Hidayat et al. (2011) proposed an
alternative model for evaluating PLOS for footpath segments in commercial areas in Bangkok
and Jakarta with street vendor activities, by incorporating qualitative and quantitative
variables. Regression analysis was used to produce an LOS model as a function of
141
pedestrians’ perceptions of comfort and the problems they considered that they faced (Hidayat
et al., 2011).
Although some previous investigators have identified several factors significantly correlating
with PLOS in their models, the relative importance of each factor and how it apparently
influences different walking behaviour in a given context has been a weakness in previous
research efforts. Often, investigators’ personal judgments have been central to decisions to
include, exclude or weight certain factors. Across all these different approaches and methods,
there is wide variety in the choice of factors suggested for PLOS evaluations. In this paper,
we sought to introduce a new method to establish a more reliable and comprehensive PLOS
measure through identifying and measuring the factors affecting PLOS, and by calculating
overall PLOS value adopting a pedestrian disgruntlement approach (pedestrian satisfaction
rating across pedestrian importance rating).
2.2. Satisfaction approach
In the field of transport research, awareness of the importance of user satisfaction is
increasing. Studies have been undertaken into passenger satisfaction with urban rail transit
from a quantitative and systematic perspective (Shen, Xiao & Wang, 2016), bus service
quality and the preferences of bus passengers (Hensher, 1990), and other public transport
users' satisfaction with service (Friman, Edvardsson & Gärling, 2001). Many of these studies
adopt a stated preference approach to identify user satisfaction (Friman & Gärling, 2001).
Friman et al. (2013) examined the psychometric properties of satisfaction with a travel scale
(STS) using confirmatory factor analyses. Their study found that users travelling by active
transport modes, such as bicycle or walking, are more satisfied than people using other travel
modes (Friman, Fujii, Ettema, Gärling & Olsson, 2013). Celik et al. (2014) developed a novel
framework model to evaluate customer satisfaction level for the rail transit network in
Istanbul by integrating statistical analysis, SERVQUAL, interval type-2 fuzzy sets and
VIKOR (Celik, Aydin & Gumus, 2014). Stradling et al. (2007) proposed an alternative
measure of customer satisfaction/dissatisfaction using something termed the ‘user
disgruntlement’ indicator. This method combines measures of performance and importance to
provide a graphic representation, in which aspects of service are most in urgent need of
improvement (Stradling, Anable & Carreno, 2007).
Surprisingly little published research has been conducted on pedestrian (or cyclists) user
satisfaction (Carreno, Willis & Stradling, 2002; Wang, Li, Wang & Namgung, 2011). Asking
what makes pedestrians satisfied or dissatisfied with footpath environment service is a
142
significant matter in attempts to encourage a walking mode. Carreno et al. (2002) evaluated
the Quality of Service (QOS) provided for pedestrians using an adapted form of 'gap analysis'
used in the measurement of consumer perceptions of QOS in service industries. Their
methods identified which features of the walking environment were causing the greatest
dissatisfaction by examining the relative importance of a wide range of factors, and also
provided a clear vision of the improvements needed for each feature (Carreno et al., 2002). A
path model was used by Wang et al. (2011) to examine the relationship among satisfaction,
identified perception factors, and physical components of footpath environments
simultaneously. Their study found that overall satisfaction is more directly affected by
perception factors than by physical components (Wang et al., 2011).
Some limitations of the previous studies include that most of the previous attempts to
investigate the relationship between satisfaction and the environment have only studied one or
two characteristics of the physical environment. In general, they have not considered the
multiple and complex components of footpath environments that may influence user's
satisfaction. Though single factor research is important, many such studies have examined in
isolation the effects of width (Kim, Choi & Kim, 2011), pavement surface conditions
(Øvstedal & Ryeng, 2004), footpath connectivity (Randall & Baetz, 2001), lighting conditions
(Fujiyama, Childs, Boampomg & Tyler, 2005), urban signscapes (Nasar & Hong, 1999), trees
(Williams, 2002), and flowers (Todorova, Asakawa & Aikoh, 2004) in isolation. Also, many
studies focused on the direct effect of physical components on satisfaction, and neglected the
emotional perceptions of users that directly affect satisfaction (Wang et al., 2011). We sought
to develop methods that would overcome many of these limitations.
3. Methodology
The paper draws upon the method used in three studies of travel provision that adopt a six
stage methodology based upon the gap analysis approach (Stradling et al., 2007). Drawing on
the six stage methodology we established the following stages in the study:
1. Identifying all factors affecting PLOS that shape the aspects of the walking
environment.
a. Selecting a comprehensive list of factors affecting PLOS on footpaths by
conducting a systematic quantitative literature review.
b. Refining the set of factors by using a semi-Delphi approach with professionals.
143
2. Surveying pedestrians’ perceptions in the study area to rate importance and
satisfaction for each selected factors.
3. Measuring pedestrian disgruntlement of each factor.
a. Conducting cross-tabulations between the importance and satisfaction ratings to
compute the percentage of disgruntled pedestrians for each factor.
b. Comparing disgruntlement percentages against importance for all factors.
4. Weighting all the factors affecting PLOS.
5. Measuring PLOS vales using a point system method.
a. Determining pedestrian level of service scale.
b. Calculating PLOS value.
6. Prioritising the potential aspects of footpath environment for remediation.
a. Defining zone conditions and prioritising implementation.
b. Determining the aspects of the walking environment that should receive priority
attention.
3.1. Selecting comprehensive list of the factors affecting PLOS on footpaths
A systematic quantitative literature review conducted by Raad and Burke (2018) was used for
the present study (Raad & Burke, 2018). This focused on studies related to the quality of the
walking environment and explicitly to PLOS. The review grouped and categorised what
almost all previously published research efforts have suggested are the most important factors
in generating PLOS outputs. A key output of the review was a list of all factors affecting
PLOS derived from the literature, with an indication of how frequently those factors were
used across the many studies. The PLOS factors were themed into five categories. These
categories were: 1) geometric characteristics; 2) obstructions and interruptions; 3) aesthetics
and amenities; 4) pedestrian traffic; and, 5) adjacent traffic. The final list of potentially
important factors affecting PLOS for footpaths is shown in Table 1.
3.2. Determining the factors affecting PLOS on footpaths in Brisbane CBD area
The Delphi technique was applied to surveys and a related workshop (a ‘walkshop’) with a set
of experts who are members and associates of the Pedestrian and Bicycle Transport Institute
of Australasia (PedBikeTrans) a key professional association that is especially active in
Brisbane. A slimmed down version of the Delphi technique was adopted as a result of
obtaining a comprehensive consensus among experts and the stability in responses between
only two Delphi rounds. There was little value in holding a third round as the results had
achieved sufficient reliability.
144
The first round was an online survey. Volunteer experts for the Delphi process were recruited
with the cooperation of the PedBikeTrans. This round involved experts from a variety of
backgrounds, including urban planning, urban design, engineering, transport planning, and
pedestrian planning. The final sample of 36 participants who completed the first round came
from local and state governments, universities and private consultancies, along with one
recently retired expert. The draft of the hierarchical framework obtained from the literature
review was organised in the form of a questionnaire. A letter introducing the study and the
Delphi process was sent to members of PedBikeTrans, inviting them to participate. This
membership is mostly Brisbane based, and has detailed knowledge of the study area and its
footpath user groups. Then a link to the online survey was sent. Illustrative images for each
factor were used to provide the best description of each factor during the Delphi rounds. The
experts were asked to rate the importance of each factor using a five-point scale, ranging from
1 ‘not at all important’ to 5 ‘extremely important’, and to suggest new factors, previously
omitted, that they believed were important.
The second round was a walkshop with 20 experts from the round one sample. This round
comprised of two phases with participants undertaking a walking course and then
participating in a further questionnaire. The walking course was designed to subject the
participants to a variety of factors affecting PLOS. Participants were asked to complete a
PLOS rating sheet independently as they walked each footpath segment along eight streets in
Brisbane’s central business district (CBD). The data collection was only undertaken for a
footpath on one side of the street and was done during late-afternoon / early-evening peak
hour when pedestrian loads were high. It is important to manage such issues as conditions can
differ on opposite footpaths on the same street, and at different times of day. Whilst walking,
they were encouraged to use their imagination and accumulated experience to consider which
factors have more effect and make more contribution to PLOS. They were also asked to
provide an overall PLOS rating for each footpath segment in the study area. In the second
phase, the experts were provided with another sheet with the revised list of the factors
affecting PLOS obtained from the results of the first round (the online survey). The experts
were then asked to make changes to the list of factors and to rate them using a five-point
Likert scale ranging from ‘extremely important’ to ‘not at all important’ for each factor, based
on their perceptions obtained during the walking course, and the results supplied from round
one.
145
The data was analyzed mostly using descriptive statistics, using a practical, non-parametric
approach. The following computations were performed for data analysis: i) medians,
interquartile ranges were calculated for each factor in both rounds as measures of convergence
of opinion and consensus measurement; ii) inter-rater reliability for both rounds using
Kendall's coefficient of concordance (W) were calculated to further determine convergence of
opinion and for measuring the stability between rounds (von der Gracht, 2012). After
calculating the inter-quartile range (IQR) for these adjusted scores, we defined those factors
with an IQR of ≤ 1 as having a high level of consensus, and vice versa. Further, inter-rater
reliability for each round was another way of proving the convergence of opinion as well as
for measuring the stability in responses between successive rounds (von der Gracht, 2012).
For this purpose, Kendall's Coefficient of Concordance (W) (Schmidt et al., 2001) was used.
This coefficient is designed to: 1) better interpret the reliability; 2) assess the consistency or
conformity of responses between two or more quantitative measurements; 3) identify overall
agreement of importance (the 5-point Likert scale) among experts; and 4) measure the
stability of the responses between any successive rounds (von der Gracht, 2012). W estimates
were calculated using the SPSS statistical package version 2. A W-value of 0.7 or more
indicates that general consistency among the participants and stability in any two successive
rounds has been achieved (Schmidt et al., 2001).
Table 1 reports all the important factors affecting PLOS that were found to have a
comprehensive consensus by experts from round two to round three, by having inter-quartile
ranges values of ≤1. Overall agreement for importance ratings (W) of all factors common for
both rounds 2 and 3 were 0.31 and 0.75, respectively. The Delphi rounds can be stopped if
either the coefficient of concordance indicates a strong consensus (W>0.70), or the level of
consensus for the panel levels off in two successive rounds (Schmidt et al., 2001). In round 3,
the level of reliability can be regarded as excellent because the W values were greater than
0.70 and were found to be statistically significant (at p < 0.0001). The general responses
among experts were found to be very stable as measured by W. The Pedestrian and Bicycle
Transport Institute of Australasia (PedBikeTrans) panel reached weak agreement (W=0.31)
with low confidence in the ranks in the second round, and had strong agreement with high
confidence in the ranks in the third round (W>0.70), according to Schmidt’s interpretation of
Kendall’s W. A W of 0.7 and more allows conclusions to be drawn at a general level, which
indicates that the general consistency among the participants and stability in responses in
round 2 and round 3 were reached.
146
It is not common that further rounds would lead to perfect consensus because absolute
consensus would probably never be achieved. Based on the results obtained, it appeared that
there was little value in another round of the Delphi process as the results had achieved
sufficient reliability.
Table 1. The important factors affecting PLOS obtained through the Delphi process.
Category Factors Factor definition
Buffer width Buffer area or street furniture zone is used to accommodate all street
furniture elements, and plants.
Crossing opportunities Refers to the number of facilities provided to assist in the safe crossing of
roads and paths by pedestrians.
Quality of footpath Refers to the quality of the surface of the path in terms of general design
surface (firm, stable and slip resistant even when wet), decorative surfacing and
materials.
Footpath slope & grade Slope refers to the slope of the footpath at right angles to the direction of
travel, while gradient refers to the longitudinal gradients (or running
slope).
Geometric Adequate footpath width The minimum width that is considered adequate for most low and high
characteristic pedestrian volumes, and must have enough space to accommodate
s pedestrian trips.
Pedestrian-traffic For example using fencing within a commercial streetscape can enhance
separation a safety feature or can be a functional support rail (leaning rail) for
people to lean against. Railings and fences can help define the street
space.
Ramp A kerb that provides a continuous path to the adjacent street facilities of
travel for pedestrians with mobility impairments or wheeled devices,
such as wheelchairs and prams, to ensure accessibility for all users.
Shoulder width A border area built along streets for the safety of motorists and
pedestrians as well as for aesthetic reasons.
Height clearance A vertical (overhead) clearance over the entire width (including the street
furniture and frontage zones.
Footpath connectivity The directness of links and the density of connections in a footpath
network. It refers to the degree to which the path provides a useful, direct
and logical link between key departure points and destinations.
Footpath continuity Refers to the completeness of the footpath within a road network.
Obstructions
and Ease of movement Refers to the ability of a pedestrian to walk on a path without
interruptions interruptions or delays, such as those that pedestrians experience at
signalled and unsignalled intersections, or when pedestrian do not travel
in clear ‘lanes’, although they may do sometimes under heavy flow.
Permanent obstructions Measures the number of obstructions per width per kilometre on the path
being assessed. Auditing this factor is essential in determining whether
access is available to people with disabilities.
Awnings Retail shops should be encouraged to provide protective awnings to

create shade, protection from rain and snow, and to otherwise add colour
and attractiveness to the street. Awnings are especially important in hot
climates on the sunny side of the street.
Bollards Any small vertical element such as a decorative steel or iron pole, or a
short concrete column intended to allow pedestrian traffic but to restrict
Aesthetics
vehicular traffic. Bollards are normally used in groups to indicate an edge
and
between pedestrian and vehicular areas.
amenities
Crossing facility type Refers to the type of crossing facilities provided to assist in the safe
crossing of roads and paths by pedestrians.
Activation of side facades Activation of side facades by adopting public art that captures and
reinforces the unique character of a place. It can interpret the community
by revealing its culture, history or fantasy.
Facilities for the disabled They are physical and tactile facilities provided along footpaths for the
disabled people such as blind or have low vision in the absence of
147
alternative enough existing cues to help them move independently and
safely through the pedestrian environment.
Footpath cleanliness Refers how footpaths should be clear of overgrown vegetation, fallen
leaves, broken glass, trash, and other unsightly or hazardous debris.
Footpath maintenance Refers how all footpath facilities should be regularly maintained, for
example, removing graffiti, mending broken windows, noticeable signs
of deferred maintenance, well-maintained gardens, and so on.
Footpath surveillance A way of enhancing personal safety or the perception of personal safety,
and provides an opportunity for a person to be seen by other people or
devices (e.g., a surveillance camera).
Trash receptacles should be easily accessible for pedestrians and trash
collection. Their design should relate to other site furnishings as well as
Garbage facilities
building architecture. They must be carefully placed to be unobtrusive
yet effective.
Guidance signs Provide information to assist pedestrians in finding their route to the key
destinations in an area.
Lighting For both safety and security reasons, most footpaths require street
lighting. Lighting is needed for both lateral movement of pedestrians and
for detection by motorists when the pedestrian crosses the roadway.
Shade Refers to the sufficiency of shade and shelter provided on a path that
helps to make the walkable environment more comfortable and more
accessible by protecting pedestrians from heat, rain, and other elements.
Public transportation
Standard bus shelters are placed at stops where there is a clear need.
facilities
Trees and plants Provides protection from weather and make the footpaths more attractive
to pedestrians.
Wayfinding guidance Refers to the type and sufficiency of navigational systems provided on a
footpath that help pedestrians to determine where they are and where
they need to go to reach a destination.
Refers to the ability of a pedestrian to cross another pedestrian traffic

Pedestrian free stream, to walk in the reverse direction of a major pedestrian flow, to
manoeuvre manoeuvre generally without conflicts and changes in walking speed. It
may be unidirectional, bidirectional, or multi directional.
Pedestrian Pedestrian conflict points The points that could arise as a result of users’ behaviour and/or failure
traffic to adequately plan, design, build and maintain a facility to fully account
for the diversity of users.
Pedestrian speed A walking speed at which pedestrian prefer to walk.
Pedestrian volume Refers to the estimate of the number of pedestrians using the footpath per
a specific period.
Refers to the number of potential vehicle conflict points along the route
Vehicle conflict points
including intersections and driveways.
Adjacent Vehicle speed Refers to the preferred average running speed of motor vehicle traffic
traffic (km/hr).
Vehicle volume Refers to the estimate of the average amount of motor vehicle traffic
during a specific period.
3.3. Surveying pedestrians’ perceptions in the study area

Following the Delphi process, an on-street intercept questionnaire survey was employed to
gather pedestrians' perceptions of the factors affecting PLOS in these same commercial street
segments within the Brisbane CBD. An online questionnaire method using a tablet PC was
adopted. The questionnaire was designed to measure pedestrians’ perceptions in five main
categories with 33 sub-items or factors that were the outputs from the Delphi process (Table
1). These categories include geometric footpath characteristics; obstructions and interruptions;
aesthetics and amenities; pedestrian traffic issues and adjacent traffic. The first part of the
148
questionnaire obtained the respondent’s characteristics including age, gender, education, trip
purpose and the frequency of use of the footpaths. The second part was allocated for assessing
the importance and satisfaction of each existing footpath facility for the pedestrians. Images
were used to help define the factor in easy to understand terms, and trained researchers were
able to further describe each factor to respondents who were uncertain. Respondents were
asked to rate both the satisfaction (how satisfied are were) and the importance (how important
good performance was to them) using a five-point Likert scale for each of the factors in
relation to these paths in the study area. The scale used for the level of satisfaction was from 1
= very dissatisfied, to 5 = very satisfied, and the level of importance was from 1 = not at all
important to 5 = extremely important. Footpaths with different characteristics in the CBD area
can be evaluated with the questionnaire because it involves a wide range of footpath aspects
and can be used in various contexts. The outputs of this survey were used to generate the
relative importance of and satisfaction for each aspect of the footpath environment. 312
respondents aged 16 and over were randomly selected to ensure that the perceptions of a wide
range of people toward the importance and satisfaction of factors affecting PLOS on footpath
were captured. Data collection was performed across the period 16th May to 16th June 2018
on various days of the week, with the surveys undertaken from 10 a.m. to 5 p.m. The
following analysis is based on these 312 responses, of which 131 (42%) were female and 181
(58%) male (Table 2). The sample also skewed towards the young and educated, with around
a quarter of the sample being international visitors, due to the large numbers of tourists and
international students who use some of the street segments surveyed. Sixty-one per cent of
respondents were aged 40 or less. Only one respondent had not completed high school; 38.5%
had undergraduate degrees. The main purpose for travel to the Brisbane CBD was 39.1% for
work and 14.7% for education. For the frequency of trip, 30.8% were almost every day,
19.9% travel at least once per month and 21.5% were less than once per month. Table 2
shows the characteristics and percentage of the sample profiles of participants.
Table 2 Demographic and travel characteristics of the survey participants (sample size = 312).
Characteristics % Characteristics %
Gender Trip purpose
Male 58 Work 39.1
Female 42 Education 14.7
Age Shopping 9.3
16-30 27.2 Eat a meal 4.8
31-40 34 Personal business 20.9
41-50 26 Recreation 10.6
Over 50 12.8 Other 0.6
149
Education level Trip frequency
Did not finish high school 0.3 Almost every day 30.8
Year 12 or equivalent 12.5 At least once per week 27.8
Trade/technical qualification 36.2 At least once per month 19.9
Undergraduate degree 38.5 Less than once per month 21.5
Postgraduate degree (Masters, PhD) 12.5
3.4. Measurement of pedestrian disgruntlement for each factor

A disgruntlement measure can be obtained by calculating both the performance or satisfaction
and importance ratings for each factor (Stradling et al., 2007). The modified 'gap analysis'
method was applied to identify the aspects of the walking environment most in need of
improvement. First we calculated the relative importance of a wide range of factors affecting
PLOS, and their relationship to each other, and how this relationship may differ for different
ages of pedestrians in different walking environments. Table 3 shows the percentages of
respondents who rated each factor as important or extremely important, ordered from highest
to lowest rated importance. Table 3 shows that all 33 factors affecting PLOS were rated as
important by 90% or more of the targeted sample, except for the last five factors (issues such
as natural surveillance). Almost none of the respondents suggested additional factors for
consideration. This suggests the earlier phases of the research had identified the salient factors
affecting PLOS.
Table 3. Importance ratings for 33 factors affecting PLOS in footpath environments.

Factors affecting PLOS Importance(%): % extremely important + %
important
Entire Age Age Only Only
sample (16-40) over 41 male female
Presence of reasonable footpath connectivity 100% 100% 100% 100% 100%
Presence of adequate footpath continuity 100% 100% 100% 100% 100%
Presence of adequate guidance signs 100% 100% 100% 100% 100%
Presence of appropriate public transportation 100% 100% 100% 100% 100%
facilities
Presence of adequate footpath width 99% 100% 98% 99% 99%
Presence of adequate crossing opportunities 99% 100% 98% 99% 99%
Provision of appropriate footpath slope & grade 99% 100% 98% 98% 99%
Presence of adequate buffer area and width 99% 100% 98% 98% 99%
Provision of adequate ease of movement 99% 99% 99% 99% 100%
Absence of permanent obstructions 99% 100% 98% 99% 98%
Presence of adequate shade 99% 100% 98% 98% 100%
Presence of adequate facilities for the disabled 99% 100% 98% 98% 100%
Provision of regular footpath maintenance 99% 100% 98% 99% 99%
Presence of adequate level of footpath cleanliness 99% 100% 98% 99% 100%
Presence of adequate garbage facilities 99% 100% 99% 99% 100%
Presence of adequate level of lighting 99% 100% 99% 100% 99%
Presence of adequate wayfinding guidance 99% 100% 99% 99% 99%
Presence of minimum number of potential pedestrian 99% 100% 98% 99% 99%
conflict points
Absence of conflict points with vehicles 99% 100% 98% 99% 99%
150
Provision of appropriate quality of footpath surface 98% 100% 97% 98% 98%
Presence of appropriate pedestrian traffic separation 98% 100% 97% 98% 97%
Presence of adequate ramp 98% 100% 96% 97% 99%
Presence of trees and plants 98% 100% 97% 97% 99%
Presence of reasonable pedestrian volume along 98% 100% 97% 98% 98%
travel line width
Limitation of vehicle speed adjacent to footpath 98% 100% 97% 97% 99%
Presence of reasonable vehicle volume adjacent to 97% 100% 96% 97% 98%
footpath
Provision of appropriate type of crossing facility 93% 98% 90% 93% 93%
Presence of adequate awnings 86% 90% 83% 83% 90%
Provision of adequate freedom level for selecting 73% 42% 92% 74% 72%
pedestrian speed
Provision of adequate level of ease for pedestrian 71% 33% 95% 72% 68%
evasive movements
Presence of appropriate activation of side facades 62% 92% 43% 56% 70%
Provision of appropriate surveillance for footpath 53% 79% 36% 48% 59%
Provision of an appropriate vertical (overhead) 34% 54% 22% 33% 36%
clearance over entire footpath width
Second, the level of pedestrian satisfaction for each aspect of the walking environment
represented by the factors was identified as a measurement of performance. The percentages
of respondents who rated the performance of the same 33 factors as satisfied or strongly
satisfied were calculated and is presented in Table 4 from worst (lowest percent agreeing the
aspect is being performed well) to best (highest percent). As may be seen from Table 4, eight
factors received a satisfaction rating below 50%.
Table 4. Performance ratings for 33 factors affecting PLOS on footpath environments.

Factors affecting PLOS Performance (%): % very satisfied + % satisfied
For
For age For age For only For only
entire
(16-40) over 41 male female
sample
Presence of trees and plants 12% 9% 14% 12% 11%
Presence of reasonable vehicle volume adjacent to
12% 5% 16% 15% 7%
footpath
Presence of reasonable pedestrian volume along
23% 21% 25% 24% 22%
travel line width
Presence of adequate level of lighting 38% 35% 39% 41% 32%
Provision of adequate level of ease for pedestrian
44% 70% 27% 47% 39%
evasive movements
Provision of adequate freedom level for selecting
48% 75% 31% 48% 49%
pedestrian speed
151
Presence of adequate wayfinding guidance 50% 43% 53% 50% 50%
Presence of appropriate pedestrian-traffic separation 53% 29% 68% 51% 54%
Presence of adequate facilities for the disabled 54% 18% 76% 53% 53%
Limitation of vehicle speed adjacent to footpath 64% 25% 81% 65% 63%
Presence of reasonable footpath connectivity 67% 65% 69% 68% 68%
Presence of minimum number of potential pedestrian
67% 62% 70% 68% 65%
conflict points
Provision of appropriate type of crossing facility 77% 52% 92% 78% 75%
Absence of conflict points with vehicles 77% 63% 86% 78% 76%
Presence of adequate guidance signs 82% 80% 83% 83% 81%
Presence of adequate ramp 85% 64% 98% 84% 85%
Presence of adequate footpath continuity 90% 95% 86% 88% 93%
Provision of appropriate quality of footpath surface 92% 92% 92% 92% 91%
Provision of appropriate surveillance of footpath 92% 94% 92% 92% 93%
Presence of adequate garbage facilities 93% 93% 95% 93% 94%
Presence of appropriate activation of side facades 94% 98% 92% 95% 93%
Presence of appropriate public transportation facilities 95% 93% 96% 93% 97%
Provision of an appropriate vertical (overhead)
96% 97% 96% 97% 95%
clearance over entire footpath width
Presence of adequate awnings 96% 97% 95% 96% 96%
Presence of adequate level of footpath cleanliness 96% 96% 96% 95% 97%
Provision of regular footpath maintenance 97% 96% 98% 97% 96%
The percentage of pedestrians who were disgruntled for each factor was derived by cross-
tabulating the percentage of respondents who were dissatisfied or strongly dissatisfied and
who rated each aspect as important or extremely important. This method provides a visual
inspection of how dissatisfaction and importance are distributed amongst the various PLOS
measures, and for then identifying the weaknesses in level of service for any aspect on
targeted footpaths. Table 5 shows one example to calculate the disgruntlement percentage for
footpath width through this cross-tabulating process, based upon the gap analysis approach
developed by Stradling et al. (Stradling et al., 2007). It appears from this example that 33% of
respondents (1.5% + 30% + 1.5% + 0%) had indicated that the aspect of adequate footpath
width is important to them but that they are dissatisfied that it was being performed well.
Table 5. Cross-tabulation of performance and importance ratings for presence of trees and plants.
Satisfaction rating % Importance rating % Total
Not at all Least Neutral Important Extremely
important important important
Very dissatisfied - - - - 27 (8.65) 27
(8.65%)
Dissatisfied 1 (0.32%) 1 (0.32%) - 2 (0.64%) 100 (32%) 104
(0.33)
Neither - - 1 (0.32) 1 (0.32%) 142 (45.5%) 144
dissatisfied/satisfied (0.46%)
Satisfied - 1 (0.32%) 1 (0.32) - 30 (9.6%) 32
152
(10.25%)
Very satisfied - - - 2 (0.64%) 3 (0.96%) 5 (1.6%)
Total % 1 (0.32%) 2 (0.64%) 2 (0.64) 5 (1.6%) 303 (97%) 312
(100%)
The cross-tabulations were repeated for all 33 factors, calculating the percentage of
‘disgruntled users’, as shown in Table 6.
Table 6. Disgruntlement measures for 33 factors affecting PLOS on footpath environments.
Factors affecting PLOS % Disgruntled pedestrians
For For For
For age For age
entire only only
(16-40) over 41
sample male female
Presence of adequate buffer area and width 50 46.6 55.4 44.7 57.2
Presence of reasonable vehicle volume adjacent for 45.2 43.5 47.9 40.3 51.9
footpath
Presence of trees and plants 41.3 39.2 44.6 36.4 48.1
Presence of adequate footpath width 40.7 40.8 40.5 37 45.8
Presence of reasonable pedestrian volume along travel 26.6 29.8 21.4 25.4 28.2
line width
Provision of adequate level of ease for pedestrian evasive 23 34 5.7 20.9 25.9
movements
Presence of reasonable footpath connectivity 22 21.4 23.1 20.4 24.4
Presence of adequate level of lighting 20.8 19.3 23.1 14.9 29
Provision of adequate freedom level for selecting 19.6 28.8 4.9 22.1 16
pedestrian speed
Presence of adequate facilities for the disabled 17.9 9.9 30.5 17.1 19
Presence of appropriate pedestrian-traffic separation 17.6 12 25.6 13.2 22.9
Absence of permanent obstructions 15.4 11 21.4 11.6 14.3
Presence of adequate wayfinding guidance 13.8 14.1 13.2 12.1 16
Presence of adequate crossing opportunities 10.9 5.7 19 8.8 13.7
Presence of adequate shade 10.2 7.8 14 7.1 14.5
Limitation of vehicle speed adjacent to footpath 8.7 3.6 16.5 6.6 10.6
Provision of adequate ease of movement 7.7 5.7 10.7 4.9 8.3
Presence of minimum number of potential pedestrian 6 3.6 9.9 3.8 9.1
conflict points
Absence of conflict points with vehicles 6 3.1 10.7 4.4 8.3
Presence of adequate ramp 4.2 0.5 9.9 4.4 3.8
Provision of appropriate type of crossing facility 4.2 2 7.4 4.4 3.8
Presence of adequate guidance signs 2.2 2 2.5 2.2 2.3
Provision of appropriate footpath slope & grade 1.3 0.5 2 1.1 1.5
Provision of appropriate quality of footpath surface 1 1.5 0 1.1 0.75
Presence of adequate footpath continuity 0.6 1 0 0.6 0.75
Presence of appropriate activation of side facades 0.6 1 0 0 1.5
Provision of an appropriate vertical (overhead) clearance 0.3 0.5 0 0 0.75
Presence of adequate awnings 0.3 0.5 0 0.6 0
Provision of appropriate surveillance for footpath 0.3 0 0 0 0.75
Presence of adequate garbage facilities 0.3 0.5 0 0.6 0
Presence of appropriate public transportation facilities 0.3 0.5 0 0.6 0
Provision of regular footpath maintenance 0 0 0 0 0
Presence of adequate level of footpath cleanliness 0 0 0 0 0
153
3.5. Weighting all the factors affecting PLOS
As mentioned previously, an on–street intercept questionnaire survey was then employed to
gather pedestrians' perceptions regarding the importance and satisfaction rating of the factors
affecting PLOS on the commercial street segments within Brisbane’s CBD. Each identified
factor has different impacts on the overall PLOS condition.
Also the perceived importance of each factor makes a different contribution to both society
and footpath environment, depending on the pedestrian’s physical condition and
characteristics, such as ethnicity, age, and gender. Pedestrians’ perceptions at this stage reflect
the level of importance and the priority of improvement for each aspect of the footpath
environment that contributes to an increase or a decrease in pedestrian level of service. First
the relative importance or weight for each footpath element is generated using the following
equation:
is the importance rating of each factor, IRs is the sum of the importance rating for all factors, i
is the number of factors (33 factors in this study), and j is the number of participants (312
participants in this study). Table 7 shows the realistic weight (RWi) of each factor affecting
PLOS on footpaths in commercial streets, grouped and themed according to the schema
developed in our earlier review paper (Raad & Burke, 2018).
Table 7 the realistic weight of importance level of the factors affecting PLOS.
Normalized
Std Cumulative
Factors affecting PLOS Mean weight
Dev. raw weight
(RWi)
Presence of adequate footpath width 4.96 0.230 1547 3.2
Geometric Characteristics
Presence of adequate crossing opportunities 4.95 0.261 1544 3.2

Provision of appropriate quality of footpath surface 4.95 0.302 1545 3.2
Provision of appropriate footpath slope & grade 4.92 0.359 1536 3.1
Presence of adequate buffer area and width 4.94 0.293 1540 3.2
Presence of appropriate pedestrian-traffic separation 4.89 0.425 1526 3.1
Presence of adequate ramp 4.87 0.445 1519 3.1
Provision of an appropriate vertical clearance over
3.04 1.035 950 1.9
footpath
Interruptions
Obstructions
Presence of reasonable footpath connectivity 4.88 0.324 1523 3.1

and
Presence of adequate footpath continuity 4.96 0.200 1547 3.2

Provision of adequate ease of movement 4.96 0.230 1547 3.2
Absence of permanent obstructions 4.93 0.308 1539 3.2
154
Presence of adequate awnings 4.48 0.852 1398 2.9
Presence of adequate shade 4.95 0.290 1543 3.2
Provision of appropriate type of crossing facility 4.76 0.603 1485 3.0
Presence of appropriate activation of side facades 3.85 1.125 1201 2.5
Presence of adequate facilities for the disabled 4.96 0.274 1546 3.2
Provision of regular footpath maintenance 4.94 0.293 1540 3.2
Presence of adequate level of footpath cleanliness 4.97 0.238 1550 3.2
Provision of appropriate surveillance for footpath 3.60 1.225 1124 2.3
Presence of adequate garbage facilities 4.98 0.159 1554 3.2
Presence of adequate guidance signs 4.97 0.168 1551 3.2
Presence of adequate level of lighting 4.97 0.186 1551 3.2
Presence of appropriate public transportation
4.97 0.168 1551 3.2
facilities
Presence of trees and plants 4.94 0.384 1541 3.2
Presence of adequate wayfinding guidance 4.92 0.290 1536 3.1
Pedestrian Traffic
Provision of adequate level of ease for evasive

4.17 1.013 1300 2.7
movements
Presence of minimum of potential pedestrian
4.94 0.293 1540 3.2
conflict points
Provision of adequate freedom for selecting
4.21 0.998 1313 2.7
pedestrian speed
Presence of reasonable pedestrian volume along
4.91 0.394 1531 3.1
travel line width
Absence of conflict points with vehicles 4.97 0.218 1551 3.2

Adjacent
Traffic
Limitation of vehicle speed adjacent to footpath 4.91 0.356 1532 3.1

Presence of reasonable vehicle volume adjacent to
4.92 0.392 1534 3.1
footpath
48835 100.0
3.6. Measuring PLOS vales using point system method

The proposed model for assessing PLOS was adopted using a disgruntlement approach with a
point system method. Factors were selected to meet the need for detailed information about
the footpath environment with a measurement of PLOS. Each factor was built to assess
individual elements of the walking environment with respect to PLOS, as well as for the
overall PLOS value. Factors include sections on various aspects of footpath features at
microscale level for different types of walking environments.
3.6.1. Determining pedestrian level of service scale
to targeted footpaths. On the basis of LOS grades adopted in the US Highway Capacity
Manual (HCM) scale, in this study, a PLOS range is allocated to one of the grades from A to
F based on a point system method. These grades are related to the current situation for
155
footpaths regarding the adequacy of footpaths’ aspects to provide a high level of comfort,
safety and mobility, and are intrinsically related to all 33 factors affecting PLOS. The range of
pedestrian level of service values might be between 1 and 6 points, based on the current
situation by the value of the obtained points. Table 8 presents the footpath conditions, the
proposed criteria and the range of scores, acceptability level, and the required remedial
actions assigned to each PLOS grade from A to F. Based on a point system method, the PLOS
grades can be defined based on a different range of scores as follows:
• PLOS A: indicates that the footpath is in excellent condition, provides highest level of
comfort, safety and mobility for pedestrian, and all the factors positively affect PLOS. PLOS
here can be assigned as 10% or less of the pedestrian disgruntlement rating for each factor.
• PLOS B: indicates that the footpath is in good condition, and few aspects of the footpath
environment (factors affecting PLOS) negatively affect pedestrian safety, comfort and
mobility. Limited procedures are needed for improvement. PLOS can be assigned as 10% -
20% of the pedestrian disgruntlement rating and is considered acceptable.
• PLOS C: indicates that the footpath is regular condition, and some aspects of the footpath
environment (factors affecting PLOS) negatively affect pedestrian safety, comfort and
mobility. Some procedures and efforts are needed for improvement. PLOS can be assigned as
20% - 40% of the pedestrian disgruntlement rating and is considered fairly acceptable.
• PLOS D: indicates that the footpath is in basic condition, and significant aspects of the
footpath environment (factors affecting PLOS) negatively affect pedestrian safety, comfort
and mobility. Many procedures and efforts are needed for improvement. PLOS can be
assigned as 40% - 60% of the pedestrian disgruntlement rating and is considered to be
unacceptable.
• PLOS E: indicates that the footpath is in poor condition, and too many aspects of the
footpath environment (factors affecting PLOS) negatively affect pedestrian safety, comfort
and mobility. Too much effort is needed here for improvement. PLOS can be assigned as 60%
- 80% of the pedestrian disgruntlement rating and is considered highly unacceptable.
• PLOS F: indicates that the footpath is totally inappropriate for walking, and the almost all
the aspects of the footpath environment (factors affecting PLOS) negatively affect pedestrian
safety, comfort and mobility. Total improvement and rehabilitation or rebuilding are needed
in this case. PLOS can be assigned as 80% or more of the pedestrian disgruntlement rating
and is considered a totally unacceptable or invalid.
Table 8 Pedestrian level of service (PLOS) grade scale.
156
PLOS
Conditions Proposed criteria Remedial action
Grade
A Excellent 10% or less of disgruntlement Very limited
B Good 10% 20% of disgruntlement Limited
C Regular 20% - 40% of disgruntlement Medium
D Basic 40% - 60% of disgruntlement High
E Poor 60% - 80% of disgruntlement Unlimited
F Appalling 80% or more of disgruntlement Rebuilding
3.6.2. Calculating PLOS value

Two main streets, including Adelaide and George Streets in Brisbane’s CBD, were chosen as
the setting for applying the methodology described in the previous section. A total of 8
footpath segments in Adelaide Street of about 2 km length, and 4 footpath segments in
George Street of about 1 km length were included to measure PLOS values in this study. To
calculate the PLOS grade, the overall PLOS score should first be calculated using a point
system method. The PLOS score for each factor can be achieved by multiplying the points
allocated to each factor by the relative importance of that factor. Table 10 shows one example
of calculating the PLOS grade for the entire sample through the point system method.
Table 9 The subcategories of PLOS
PLOS Category Total PLOS grade vis PLOS score

PLOS F 1 ≤ F- >1.5 F = 1.5 1.5 < F+ ≥ 1.9
PLOS E 2 ≤ E- > 2.5 E =2.5 2.5 < E+ ≥ 2.9
PLOS D 3 ≤ D- > 3.5 D = 3.5 3.5 < D+ ≥ 3.9
PLOS C 4 ≤ C- > 4.5 C = 4.5 4.5 < C+ ≥ 4.9
PLOS B 5 ≤ B- > 5.5 B = 5.5 5.5 < B+ ≥ 5.9
PLOS A A=6
Also, appropriate points can be allocated to each factor based on the grade assigned for that
factor. The grade of each factor was determined based on the pedestrian disgruntlement
percentage obtained from Table 6 and according to the scale developed in Table 8. Table 11
shows a comparison in PLOS grade between all factors affecting PLOS among various ages
and genders. The overall PLOS score was calculated by summing all PLOS scores for all the
factors. Then, the overall PLOS score was allocated to an appropriate grade from A to F based
on the PLOS score result. According to Table 9, the overall PLOS score of 5.2 as shown in
Table 11 reflects PLOS B for the entire sample, represented by 5.3 (B-) for young people, 5
(B) for elderly, 5.3 (B-) for males, and 5.1 (B-) for females.
157
Table 10 An example of calculating pedestrian level of service (PLOS) grade.
Disgruntlement PLOS Allocated PLOS
Category Factors affecting PLOS RWi %
ratings grade points score
Footpath width 40.7% D 3 0.032 0.096
Crossing opportunities 10.9% B 5 0.032 0.16

Quality of footpath surface 1% A 6 0.032 0.192
Footpath slope & grade 1.3% A 6 0.031 0.186
Buffer area and width 50% D 3 0.032 0.096
Pedestrian-traffic separation 17.6% B 5 0.031 0.155
Ramp 4.2% A 6 0.031 0.186
Clearance over footpath 0.3% A 6 0.019 0.114
Footpath connectivity 22% C 4 0.031 0.124

Interruption
Obstruction
Footpath continuity 0.6% A 6 0.032 0.192

s and
Ease of movement 7.7% A 6 0.032 0.192

Permanent obstructions 15.4% B 5 0.032 0.16
Awnings 0.3% A 6 0.029 0.174

Shade 10.2% A 6 0.032 0.192
Type of crossing facility 4.2% A 6 0.03 0.18
Activation of side facades 0.6% A 6 0.025 0.15
Facilities for the disabled 17.9% B 5 0.032 0.16

Footpath maintenance 0% A 6 0.032 0.192
Footpath cleanliness 0% A 6 0.032 0.192
Footpath surveillance 0.3% A 6 0.023 0.138
Garbage facilities 0.3% A 6 0.032 0.192
Guidance signs 2.2% A 6 0.032 0.192
Lighting 20.8% C 4 0.032 0.128
Public transportation 0.3% A 6 0.032 0.192
Trees and plants 41.3% D 3 0.032 0.096
Wayfinding guidance 13.8% B 5 0.031 0.155
Evasive movements 23% C 4 0.027 0.108

Pedestrian
Traffic
Pedestrian conflict points 6% A 6 0.032 0.192

Pedestrian speed 19.6% B 5 0.027 0.135
Pedestrian volume 26.6% C 5 0.031 0.155
Vehicle conflict points 6% A 6 0.032 0.192

Adjace
Traffic
nt
Vehicle speed 8.7% A 6 0.031 0.186

Vehicle volume 45.2% D 3 0.031 0.093
PLOS value B- 5.2
158
Table 11 Pedestrian level of service (PLOS) grade for various ages and genders.
Catego Factors affecting PLOS grade
ry PLOS Entire
Young Elderly Male Female
sample
Footpath width D D D C D
Crossing opportunities B A B A B
Quality of footpath
A A A A A
surface
Footpath slope & grade A A A A A
Buffer area and width D D D D D
Pedestrian-traffic
B B C B C
separation
Ramp A A A A A
Clearance over
A A A A A
footpath
Obstructions and
Footpath connectivity C C C C C
Interruptions
Footpath continuity A A A A A
Ease of movement A A B A A
Permanent obstructions B B C B B
Awnings A A A A A
Shade A A B A B
Type of crossing
A A A A A
facility
Activation of side
A A A A A
facades
Facilities for the

B A C B B
disabled
Footpath maintenance A A A A A
Footpath cleanliness A A A A A
Footpath surveillance A A A A A
Garbage facilities A A A A A
Guidance signs A A A A A
Lighting C B C B C
Public transportation A A A A A
Trees and plants D C D C D
Wayfinding guidance B B B B B
Evasive movements C C A C C
Pedestrian
Traffic
Pedestrian conflict
A A A A A
points
Pedestrian speed B C A C B
Pedestrian volume C C C C C
Adjacent
Traffic
Vehicle conflict points A A B A A

Vehicle speed A A B A B
Vehicle volume D D D D D
Overall PLOS score and grade 5.2, B- 5.3, B- 5, B 5.3, B- 5.1, B-
159
3.7. Prioritising the potential aspects of footpath environment for remediation
A scatter-graph method was adopted to identify the potential aspects of the footpath
environment for prioritisation by city authorities with responsibility for the study area (Figure
1). The factors affecting PLOS were plotted using Microsoft Excel by graphing the
disgruntlement percentage against the importance percentage for each factor, as shown in
Figure 1. Prioritising potential areas for remediation for each aspect was identified by
dividing the plot into four action zones (A1, A2, A3 and A4) around the centroid of the data,
enabling the identification of those factors which pedestrians perceived as a priority (Table 7).
The factors placed in Zone 1 are considered top priority with high degree of disgruntlement
and importance for pedestrians.
Table 12 Definition of zone conditions and prioritisation of factors from A1 to A4.

Zone Condition Priority for remedial action
A1 High importance + high disgruntlement High
A2 Low importance + high disgruntlement High/medium
A3 Low importance + low disgruntlement Low
A4 High importance + low disgruntlement Medium
The analysis revealed the main sources of dissatisfaction for pedestrians regarding these
footpaths. Specific factors relating to each of the older and younger users were identified.
Figure 1 presents the scatter graph of the disgruntlement percentage against the importance
ratings of each factor. For the whole sample, the main factors having high levels of
disgruntlement are vehicle volume, buffer area, footpath width, trees and plants, footpath
connectivity, pedestrian volume and lighting (Fig. 1). All the factors plotted in Zone A1 with
a high level of pedestrians’ disgruntlement obtained a low level of service with PLOS D. This
reflects an urgent need for remedial action.
Although factors relating to crossing opportunities, facilities for the disabled, traffic
separation, permanent obstruction, wayfinding and crossing facilities were plotted in Zone 4
for all of the groups, they caused concern for the users both in terms of importance and
disgruntlement percentage, however, the PLOS reflects an acceptability level of B.
160
Fig. 1. Scatter graph of disgruntlement vs. importance for all factors affecting PLOS for the entire
sample.
4. Results
The results of the survey provided the level of service value for a total of 12 footpath
segments in Adelaide and George Streets in Brisbane’s CBD. The overall PLOS score of 5.2
as shown in Table 11 truly shows that the level of service is equal to level B for the entire
sample, 5.3 (B-) for young people, 5 (B) for elderly, 5.3 (B-) for male, and 5.1 (B-) for
female.
Although the majority of the aspects of the walking environment were satisfied by pedestrians
with a PLOS rating of A, or B, pedestrians, including disabled people, should walk within a
narrow width in comparison to the extreme pedestrian volume, especially at peak hours.
Footpath width and buffer area were assigned a low level of service value by pedestrians of
different ages and genders within the category of geometric characteristics. However, elderly
people complained more than others about the lack of pedestrian-traffic separation.
Within the category of obstructions and interruptions, the consensus among different
pedestrian ages and genders regarding footpath connectivity was absolute, and this occupied
the lowest value with a PLOS score of C. Improving the connectivity of footpaths can result
in substantial improvement to the walkability of an area and an enlargement of the walkable
catchment for activities and activity centres, thus providing a high level of service for
pedestrians. Also, the more elderly users voiced their disgruntlement regarding permanent
obstructions, giving the lowest value of PLOS as C in comparison with others. The existence
161
of permanent obstructions can affect a clear access route for pedestrians. Street furniture, such
as garbage facilities, guidance signs, lighting, transportation facilities, trees, wayfinding
guidance, and so on can be permanent obstructions if randomly distributed on the path, or if
sufficient width for the buffer area is not allocated. Permanent obstructions might also make
the path very narrow and unattractive for high pedestrian volumes and for some users.
PLOS for facilities for the disabled obtained the worst value by elderly users in comparison to
other users, which reflects the need for improvements to footpaths to cater for the elderly.
Lighting is related to personal safety, and helps to improve the visibility and overall ambience
of pedestrian pathways, while trees, shrubs, and other landscaping along pedestrian footpaths
provide protection from bad weather, making an area more comfortable and attractive for
walking. Elderly people and female respondents assigned the factors of lighting and trees and
plants to score the PLOS as C and D, respectively.
Most pedestrians evaluated evasive movements and pedestrian volume as the worst aspects
within the category of pedestrian traffic. Consequently, a PLOS score of C was assigned to
both factors. Our opinion is that the low values given to these factors were as a result of
impacts which occurred because of insufficient footpath width, which cause many difficulties
in bypassing slower pedestrians or moving in reverse flows. However, it seems that the
elderly do not care about evasive movements and therefore, based on their ratings, a PLOS
score of A was recorded. Also, increasing pedestrian volumes complicated situations when
combined with other factors. Pedestrian volume plays a key role in designing the pedestrian
environment and directly affects walking safety, comfort, and mobility, and then, therefore,
PLOS.
The worst PLOS value within the adjacent traffic category was recorded by various ages and
genders for the vehicle volume factor with a PLOS score equaling D. The interaction between
pedestrians and traffic is mainly affected by the number of vehicles passing adjacent to
pedestrian facilities. This interaction is directly related to pedestrian safety, comfort and
mobility. The volume of vehicular traffic adjacent to/passing through pedestrian routes is one
of the main reasons why people do not feel safe or do not find a route attractive, especially, at
the end of each footpath segment, through intersections, or at turning points to another street
through midblock crossings. Therefore, it is necessary that the volume of motorised traffic is
acceptable for pedestrians.
162
5. Discussion
This paper presents a new methodology for measuring the factors that affect PLOS, based on
a new indicator to identify PLOS using a disgruntlement approach. Although this approach
was applied on footpaths in commercial streets in Brisbane, the approach and methods could
be used across a range of different facility types and in differing urban contexts.
The main contribution of this paper is methodological, using the six-step approach, and in
harnessing the gap analysis technique used in other fields and applying it to PLOS
measurement for the first time. As noted from studying the literature, previous PLOS
measurement approaches relied far too heavily on experts alone, often on the judgement of
only one or two experts, and had little in the way of testing with ordinary pedestrians.
Building on previous research, our approach using a sizeable group of experts (20 in the main
walkshop) and then field tested with 312 pedestrians is significantly more robust. The
scientific and practical benefits of this study were enhanced by involving the measurement of
importance of each factor affecting PLOS and combining this with a satisfaction measurement
to create a new model to measure pedestrian disgruntlement. Determining the percentage of
respondents who are currently disgruntled is key to identifying priority issues and, eventually,
to improving the streetscape. These methods open pathways for others to further refine and
improve on PLOS measures.
The pedestrian disgruntlement approach considers a wide range of the qualitative and
quantitative factors affecting PLOS which are impossible to be included together using the
HCM method. Our method provides a visual inspection of how dissatisfaction and importance
are distributed amongst the various PLOS measures, and then identifies the weaknesses in
level of service for any aspect on targeted footpaths. Further, as applied in this study, the
modified gap analysis approach, which has previously been used in measuring consumers’
perceptions of quality of service (QOS) in the service industries can help to identify the
aspects of the walking environment most in need of improvement. This technique also allows
for determination of the perceived ‘quality gaps’ of each aspect of a walking environment,
and it is not difficult to conceive how the approach could be used with cyclists and others. It
may also be used for different transport services and modes, different trip purposes, and
different population sub-groups.
There are many applied contributions in this paper that are of use to city authorities in
Brisbane in terms of planning the streets, footpaths and the urban environment of commercial
streets in the city. Statistical significance appears to have been obtained on most of the key
163
measures in this study. These procedures confirm that the described approach is
methodologically appropriate, and also that the results produced truly reflect the quality and
environment of city footpaths.
However, there is a key methodological limitation has been identified, which is that the
results obtained from pedestrians about the level of importance of the factors affecting PLOS
are at times too homogeneous, as compared to the level of disgruntlement. The choice of a
five-point Likert scale, instead of a seven-point Likert scale may have influenced this
outcome. Future researchers might wish to adopt a finer-grained scale. Although the
pedestrian survey sample was sizeable, obtaining 312 participants for such a lengthy survey, it
is not feasible to split the resulting dataset down to key sub-populations. Using sub-
population enquiries to explore whether key results are similar for younger and much older
pedestrians is important, especially given the need to create age-friendly cities. Future
research efforts, particularly those with access to professional survey support, could overcome
this limitation, using these methods over a more diverse set of times-of-day and targeting key
sub-sets of users.
References
Asadi-Shekari, Z., Moeinaddini, M., & Shah, M. Z. (2013). Disabled Pedestrian Level of
Service Method for Evaluating and Promoting Inclusive Walking Facilities on Urban
Streets. Journal of Transportation Engineering-Asce, 139(2), 181-192.
doi:10.1061/(Asce)Te.1943-5436.0000492
Asadi-Shekari, Z., & Zaly-Shah, M. (2011). Practical Evaluation Method for Pedestrian Level
of Service in Urban Streets' International Transport Research Conference. Penang,
Malaysia.
Boodlal, L. (2004). Accessible sidewalk and street crossing-AN informational guide.
Retrieved from
Carreno, M., Willis, A., & Stradling, S. (2002). Quality of service for pedestrians: closing the
gaps in knowledge. In Traffic And Transportation Studies (2002) (pp. 326-333).
Celik, E., Aydin, N., & Gumus, A. T. (2014). A multiattribute customer satisfaction
evaluation approach for rail transit network: A real case study for Istanbul, Turkey.
Dixon, L. (1996). Bicycle and pedestrian level-of-service performance measures and
standards for congestion management systems. Transportation Research Record:
Journal of the Transportation Research Board(1538), 1-9.
Ewing, R., Meakins, G., Hamidi, S., & Nelson, A. C. (2014). Relationship between urban
sprawl and physical activity, obesity, and morbidity – Update and refinement. Health
& Place, 26, 118-126. doi:http://dx.doi.org/10.1016/j.healthplace.2013.12.008
Forsyth, A., Hearst, M., Oakes, J. M., & Schmitz, K. H. (2008). Design and Destinations:
Factors Influencing Walking and Total Physical Activity. Urban Studies, 45(9), 1973-
1996. doi:10.1177/0042098008093386
Frank, L., Engelke, P., & Schmid, T. (2003). Health and community design: The impact of the
built environment on physical activity: Island Press.
164
Friman, M., Edvardsson, B., & Gärling, T. (2001). Frequency of negative critical incidents
and satisfaction with public transport services. I. Journal of Retailing and Consumer
Services, 8(2), 95-104.
Friman, M., Fujii, S., Ettema, D., Gärling, T., & Olsson, L. E. (2013). Psychometric analysis
of the satisfaction with travel scale. Transportation Research Part A: Policy and
Practice, 48, 132-145.
Friman, M., & Gärling, T. (2001). Frequency of negative critical incidents and satisfaction
with public transport services. II. Journal of Retailing and Consumer Services, 8(2),
105-114.
Fruin, J. J. (1971). Pedestrian planning and design. Retrieved from
Fujiyama, T., Childs, C., Boampomg, D., & Tyler, N. (2005). Investigation of Lighting Levels
for Pedestrians-Some questions about lighting levels of current lighting standards.
Gallin, N. (2001). Quantifying pedestrian friendliness--guidelines for assessing pedestrian
level of service. Road & Transport Research, 10(1), 47.
Guo, H., Zhao, F., Wang, W., Zhou, Y., Zhang, Y., & Wets, G. (2014). Modeling the
Perceptions and Preferences of Pedestrians on Crossing Facilities. Discrete Dynamics
in Nature and Society, 2014, 8. doi:10.1155/2014/949475
Heath, G. W., Brownson, R. C., Kruger, J., Miles, R., Powell, K. E., Ramsey, L. T., &
Services, T. F. o. C. P. (2006). The effectiveness of urban design and land use and
transport policies and practices to increase physical activity: a systematic review.
Journal of Physical Activity & Health, 3, S55.
Hensher, D. A. (1990). Hierarchical stated response designs-an application to bus user
preferences. Logistics and Transportation Review, 26(4), 299.
Hidayat, N., Choocharukul, K., & Kishi, K. (2012). Important factors on sidewalk with
vendor activities based on pedestrian perception by gender and age Jurnal
Transportasi, 12(3).
Hidayat, N., Choocharukul, K., & Kishi, N. (2010). Analysis of Sidewalk Level of Service
Incorporating Pedestrian’s Perceptions’ Towards Traffic and Transport Safety. Paper
presented at the The 7th National Transport Conference: NTC7, 15th October 2010.
Hidayat, N., Choocharukul, K., Nakatsuji, T., & Kishi, K. (2009). Determination of factors
related to sidewalk performance based on pedestrian perception. Paper presented at
the 40th Conference of Infrastructure Planning Japan Society of Civil Engineers,
Kanazawa University.
Hidayat, N., Choocharukul, K., & Kishi, K. (2011). Pedestrian Level of Service Model
Incorporating Pedestrian Perception for Sidewalk with Vendor Activities. Journal of
the Eastern Asia Society for Transportation Studies, 9, 1012-1023.
doi:10.11175/easts.9.1012
Jaskiewicz, F. (2000). Pedestrian level of service based on trip quality. Transportation
Research Circular, TRB.
Kang, L., Xiong, Y. G., & Mannering, F. L. (2013). Statistical analysis of pedestrian
perceptions of sidewalk level of service in the presence of bicycles. Transportation
Research Part a-Policy and Practice, 53, 10-21. doi:10.1016/j.tra.2013.05.002
Kerridge, J., Hine, J., & Wigan, M. (2001). Agent-Based Modelling of Pedestrian
Movements: The Questions That Need to Be Asked and Answered. Environment and
Planning B: Planning and Design, 28(3), 327-341. doi:10.1068/b2696
Khisty, C. J. (1994). Evaluation of pedestrian facilities: beyond the level-of-service concept
(0309055199). Retrieved from
165
Kim, S., Choi, J., & Kim, Y. (2011). Determining the sidewalk pavement width by using
pedestrian discomfort levels and movement characteristics. KSCE Journal of Civil
Engineering, 15(5), 883.
Landis, B., Vattikuti, V., Ottenberg, R., McLeod, D., & Guttenplan, M. (2001). Modeling the
roadside walking environment: pedestrian level of service. Transportation Research
Record: Journal of the Transportation Research Board(1773), 82-88.
Lee, B. J., Jang, T. Y., Wang, W., & Namgung, M. (2009). Design criteria for an urban
sidewalk landscape considering emotional perception. Journal of Urban Planning and
Development, 135(4), 133-140.
Maghelal, P. K., & Capp, C. J. (2011). Walkability: a review of existing pedestrian indices.
URISA Journal, 23(2), 5-20.
Mehta, V. (2008). Walkable streets: pedestrian behavior, perceptions and attitudes. Journal of
Urbanism: International Research on Placemaking and Urban Sustainability, 1(3),
217-245. doi:10.1080/17549170802529480
Muraleetharan, T., Adachi, T., Hagiwara, T., & Kagaya, S. (2005). Method to determine
pedestrian level-of-service for crosswalks at urban intersections. Journal of the
Eastern Asia Society for Transportation Studies, 6, 127-136.
Nasar, J. L., & Hong, X. (1999). Visual preferences in urban signscapes. Environment and
Behavior, 31(5), 671-691.
Øvstedal, L., & Ryeng, E. (2004). PROMPT Pedestrian Comfort Synthesis Report. In:
Trondheim: SINTEF Teknologi og samfunn.
Pikora, T., Giles-Corti, B., Bull, F., Jamrozik, K., & Donovan, R. (2003). Developing a
framework for assessment of the environmental determinants of walking and cycling.
Social Science & Medicine, 56(8), 1693-1703.
Polus, A., Schofer, J. L., & Ushpiz, A. (1983). Pedestrian Flow and Level of Service. Journal
of Transportation Engineering-Asce, 109(1), 46-56. doi:doi:10.1061/(ASCE)0733-
947X(1983)109:1(46)
Pushkarev, B., & Zupan, J. (1975). Urban space for pedestrian. A Report of the Regional
Planning Association, The MIT Press, Cambridge, Mass.
Raad, N., & Burke, M. I. (2018). What Are the Most Important Factors for Pedestrian Level-
of-Service Estimation? A Systematic Review of the Literature. Transportation
Research Record, 0361198118790623.
Randall, T. A., & Baetz, B. W. (2001). Evaluating pedestrian connectivity for suburban
sustainability. Journal of Urban Planning and Development, 127(1), 1-15.
Saelens, B. E., & Handy, S. L. (2008). Built environment correlates of walking: a review.
Medicine and science in sports and exercise, 40(7 Suppl), S550.
Sarkar, S. (1993). Determination of service levels for pedestrians, with European examples.
Transportation Research Record(1405).
Sarkar, S. (2003). Qualitative evaluation of comfort needs in urban walkways in major
activity centers. Transportation quarterly, 57(4), 39-59.
Shen, W., Xiao, W., & Wang, X. (2016). Passenger satisfaction evaluation model for Urban
rail transit: A structural equation modeling based on partial least squares. Transport
Policy, 46, 20-31.
Singh, K., & Jain, P. (2011). Methods of assessing pedestrian level of service. Journal of
Engineering Research and Studies, 2(1), 116-124.
166
Sousa, L., Sequeira, C., & Ferré-Grau, C. ‘Living together with dementia’ – Conceptual
validation of training programme for family caregivers: Innovative practice.
Dementia, 0(0), 1471301218762565. doi:10.1177/1471301218762565
Stradling, S. G., Anable, J., & Carreno, M. (2007). Performance, importance and user
disgruntlement: A six-step method for measuring satisfaction with travel modes.
Transportation Research Part A: Policy and Practice, 41(1), 98-106.
Tanaboriboon, Y., & Guyano, J. A. (1989). Level of service standards for pedestrian facilities
in Bangkok: A case study. ITE journal, 59(11), 39-41.
Todorova, A., Asakawa, S., & Aikoh, T. (2004). Preferences for and attitudes towards street
flowers and trees in Sapporo, Japan. Landscape and Urban Planning, 69(4), 403-416.
von der Gracht, H. A. (2012). Consensus measurement in Delphi studies: Review and
implications for future quality assurance. Technological Forecasting and Social
Change, 79(8), 1525-1536. doi:https://doi.org/10.1016/j.techfore.2012.04.013
Wang, W., Li, P., Wang, W., & Namgung, M. (2011). Exploring determinants of pedestrians’
satisfaction with sidewalk environments: Case study in Korea. Journal of Urban
Planning and Development, 138(2), 166-172.
Williams, K. (2002). Exploring resident preferences for street trees in Melbourne, Australia.
Journal of Arboriculture, 28(4), 161-170.
167
7 Developing and testing pedestrian level of service (PLOS) audit tool for
footpaths in commercial streets
7.1 Chapter summery
The main aim of this chapter is to develop a comprehensive and reliable PLOS model using
an audit tool method to measure the overall PLOS value. This chapter provides a description
of a unique attempt to develop and establish an appropriate methodology that leads to a new
PLOSAT model appropriate for the Australian context. A mixed-methods design is applied in
the research to collect and analyse data, including a systematic quantitative literature review
using the PRISMA technique, a Delphi technique, a pedestrian intercept survey, a point
system method, and an inter- and intra-rater reliability test. The PLOS audit tool model was
developed to provide a systematic, reliable and empirical method to measure factors affecting
PLOS, and to efficiently evaluate PLOS for footpaths along commercial streets. This model
consists of two parts. The first part is a fieldwork form of the factors affecting PLOS called
the Pedestrian Level of Service Audit Tool (PLOSAT). PLOSAT consists of 33 items or
factors within five categories. The second part is a weighted assessment form (WAF) which is
used for calculating the overall pedestrian level of service involving the same 33 factors
included within the fieldwork form. All items were selected based on the results of Chapter 4
(a systematic literature review) and Chapter 5 (the Delphi process). Both the WAF and the
PLOSAT parts of this model were built based on the results of Chapter 6 (pedestrians’
perceptions) and the comprehensive review conducted for the Australian Guidelines,
Standards and previous audit tools. Some factors or items included in this model are evaluated
by assessing the current conditions for each item compared to the standard requirement in the
pedestrian guidelines. Other items are evaluated based on eligibility criteria developed
according to the previous attempts of PLOS models. In this context, the subjective assessment
of raters may play a role in the assessment process, as to how they measure each item in the
field based on eligibility criteria. For this reason, a reliability test was required for the
development of a robust version of PLOSAT. The reliability or validity of the new audit tool
was assessed using inter- and intra-rater reliability. Inter-/intra-rater reliability was estimated
from data that were observed by six participants who rated footpath segments. A measure of
agreement between raters leads to measuring the degree of reliability of the PLOSAT.
Furthermore, kappa (κ) statistics were used to overcome any level of agreement which could
occur by chance. The key aim of this test was to obtain the degree of agreement among raters.
It gives a score of how much homogeneity, or consensus, there was in the ratings in order to
168
assess the degree of reliability of a new model. The proposed tool adequately evaluates all
factors related to issues of safety, comfort and mobility in the CBD area to ensure the
provision of a high level of service for all pedestrians.

Primary questions:
RQ4: How could a more optimal pedestrian level of service (PLOS) audit tool for footpaths
in commercial streets be developed which would be appropriate for Australian cities?
Sub-questions:

1. The methodological contributions of this chapter:
This chapter presents a combination of methodological methods to establish a PLOS audit
tool model which includes the PRISMA technique, the Delphi technique, a pedestrian
intercept survey, road guidelines and standards, a point system method, and an inter- and
intra-rater reliability test. It is considered to be the first comprehensive and reliable audit tool
established to objectively quantify PLOS value based on pedestrians’ perceptions of safety
and comfort for footpaths in commercial areas. Another main contribution of the proposed
PLOS model in this study is that it has the right combination of all quantitative and qualitative
factors involved to assess PLOS. Further, adopting this approach in measuring PLOS using
qualitative and quantities factors can supplement the results obtained through the quantitative
PLOS of flow, speed, and density, as elaborated in the HCM approach.
2. Practical and applied contributions of the chapter:
 This chapter produces a new weighted model based on a systematic review, a rating of
a panel of experts and pedestrians’ perceptions to evaluate PLOS.
169
 The outcome of this study provides a reliable, empirical and comprehensive method to
collect data, measure factors affecting PLOS and evaluate PLOS for footpaths along
commercial streets.
 The proposed PLOS model provides a good means of evaluating a wide range of
footpath facilities and pedestrian users, including disabled people.
 This study can be considered a guideline for evaluating pedestrian level of services in
Australian cities using systemic, reliable and empirical data.
 The proposed PLOS model in this study can be used as an indicator of the current
conditions of footpaths and other pedestrian facilities.
 The proposed PLOS model in this study can be used to improve existing conditions by
comparing existing conditions with standards, while considering pedestrians’
perceptions.
 The proposed model could be used by urban designers to design new, inclusive,
pedestrian friendly streets, in addition to improving existing streets.
 This study is harmonized with what The Department of Transport and Main Roads in
Queensland has recommended as an efficient way to provide safe and accessible
pedestrian facilities to accommodate all pedestrians, including those with mobility
difficulties.
 The proposed PLOS audit tool could be used to create a clear image to guide decision
makers in evaluating, monitoring and comparing the performance and quality of
pedestrian facilities.
 The proposed PLOS model could be most useful in guiding the allocation of funds for
changes and improvements to facilities.
170
Developing and testing pedestrian level of service (PLOS) audit tool for
footpaths in commercial streets
Abstract
Planners and engineers seeking to facilitate walking are increasingly trying to measure
pedestrian level of service (PLOS). It is critical that existing walking facilities are evaluated
in order to develop a safe, pleasant walking environment for pedestrians through measuring
the factors affecting PLOS on footpaths. In this chapter, the PLOS audit tool (PLOSAT) is
proposed to assess PLOS on footpaths in commercial areas. The methodology used to
evaluate the pedestrian level of service mainly involves determining the various factors which
influence pedestrians in terms of perceived safety and comfort. The proposed tool was
developed and tested based on expert workshops and pedestrian perception surveys, as shown
in Chapter 4-6. Thirty-three factors affecting PLOS are now taken forward into the proposed
tool to reflect different aspects of the walking environment. The instrument is then subjected
to an inter-rater reliability test using the percentage of agreement and the kappa statistic. The
results of PLOSAT reliability indicated a moderate to substantial inter-rater reliability over
the majority of the factors affecting PLOS. The reliability of overall PLOS results indicated a
high reliability. The raters also reported that the audit tool was easy to use.
171
1. Introduction
Streets are prominent parts of public space and reflect the liveability of communities. Many
streets are designed primarily for cars and neglect pedestrian requirements. Yet pedestrians
are a major users of streets, especially in activity centres and commercial areas. There is also
significant tension identified between subjective and objective measures, and uncertainty
about how to conceive of and include streetscape items (Raad & Burke, 2017).
In recent years, much more attention has been paid to active transport, including walking and
cycling, and the needs and opinions of pedestrians have become an increasing consideration
in road design. Efforts have been made toward improving pedestrian facilities and operational
characteristics in order to create appropriate environments for users and to promote walking.
Objectives include reducing traffic, improving safety, economic development and creating a
better environment. It is now recognised that streets must meet the needs of users and provide
a convenient environment for pedestrians, while at the same time, discouraging dependence
on vehicles (Sisiopiku & Akin, 2003). Much of the effort towards improving public spaces
has focused on streets, given their importance in the public realm (Mehta, 2007).
Studies on the relationship between the built environment and walking have led to the
creation of numerous tools and methods for measuring walkability (Saelens & Handy, 2008).
Improving the built environment and street configurations, Based on how pedestrians perceive
these issues, improvement of the built environment and street configurations is a priority, and
such improvement will promote walking in urban precincts (Koohsari et al., 2013). However,
there may be many factors that influence pedestrian behaviour in the built environment that
are yet to be discovered (Praveen K. Maghelal & Cara Jean Capp, 2011). A sub-set of the
research on built environments and walking is focused on the micro-scale of urban design.
These studies consider how an individual link, such as a section of footpath or an individual
crossing, provides for pedestrians. It is at this scale that we most commonly use the concept of
pedestrian level of service (PLOS).
2. PLOS approach
A level of service can be defined as an assessment of the quality of service or an overall
infrastructure. More technically, it can be described as “a quantitative stratification of a
performance measure or measures that represent quality of service” (L. Kang, Y. Xiong & F.
172
L. Mannering, 2013). A level of service (LOS) method can indicate existing conditions for
street users or forecast possible conditions for hypothetical designs to be built in the future. In
this sense, Gallin defined pedestrian level of service as “an overall measure of walking
condition on a route, path and facility” (Gallin, 2001). Pedestrian environments are becoming
more important in an urbanising world where walking is increasingly being encouraged.
Engineers and planners determine a link’s performance using measures codified as level of
service, which provide guidance on acceptable or desirable standards.
Many studies have been conducted to evaluate streets using LOS in relation to motorised
vehicles. However, until the 1970s this conceptual approach was not taken across into the
field of pedestrian studies. The needs of pedestrians were mostly neglected. Only recently has
research accelerated in the field of PLOS, with more attention being given to pedestrian
design (Raad & Burke, 2018). The main approach to PLOS has changed over time. As will be
shown in detail, initially the concept was very much about volume of pedestrians vs. capacity
and the freedom of movement for pedestrians in a given environment. This conception of
PLOS was narrow and engineering focused. As with LOS for motorised traffic, PLOS tools
tended to use a six-level scale using a scale of letters to suggest whether a facility offered
good service or not. Facilities scored as an ‘A’ offered the best PLOS; facilities scoring an ‘F’
the worst (Lei Kang et al., 2013). This may be seen as providing both objectivity in
measurement and an easy system for road managers to help prioritise investment. Over time
the narrow conception of PLOS as being mostly about volume vs. capacity has been
challenged. In a set of evolutionary steps, and despite much debate, the concept has been
broadened and expanded to encompass a greater set of issues from a wider range of
disciplines.
In our study, a wide range of factors affecting PLOS for different users has been considered
using an audit tool. This tool will aid designers and engineers in evaluating, developing or
retrofitting footpaths in commercial streets in the Australian context.
3. Developing and measuring PLOS using point system method

The proposed model for assessing PLOS adopted an audit tool approach using a point system
method. The pedestrian level of service model was developed to provide a systematic, reliable
and empirical method to measure factors affecting PLOS and efficiently evaluate PLOS for
footpaths along commercial streets. This model consists of two parts, including i) a fieldwork
form of the factors affecting PLOS called the Pedestrian Level of Service Audit Tool
173
(PLOSAT); and ii) a weighted assessment form (WAF) for calculating PLOS values. The two
parts of a pencil and paper PLOS model are shown in Tables 1 and 2 respectively. Both parts
were designed to meet the need for detailed information about the footpath environment. Each
audit item/factor was built to assess the individual elements of the walking environment with
respect to PLOS, as well to overall PLOS value. Audit items include sections on various
aspects of footpath features at the microscale level for different types of walking
environments, with many borrowed from previous audit tools duly acknowledged here.
Based on the hierarchical framework of the factors affecting PLOS, the PLOSAT items were
themed in two groups of factors based on the nature of influence. The first group consists of
factors related to the pedestrian environment, and involves three categories: 1) geometric
characteristics; 2) obstructions and interruptions; and 3) aesthetics and amenities. The second
group contains factors related to the traffic system and users, and includes two categories: 1)
pedestrian traffic; and 2) adjacent traffic. The framework determines the ‘groups’ at the top
level (defined as the overall factors affecting PLOS), ‘categories’ at the second level (defined
as the factors that are the components of groups), and ‘items’ at the bottom level that
influence the categories (defined as micro-factors that have the potential to be changed in
order to improve categories). For example, under the ‘factors related to the pedestrian
environment’ groups, ‘geometric characteristics’ is one of three influential categories, while
‘footpath width’ and ‘slope’ are two of the seven items that are key attributes of the geometric
characteristics which affect PLOS on footpaths. PLOSAT consists of 33 items in total under
these five categories: 7 in the geometric characteristics category, 4 in the obstruction and
interruption category, 14 in the aesthetic and amenity category, 4 in the pedestrian traffic
category, and 3 in the adjacent traffic category.
3.1. Pedestrian level of service audit tool (PLOSAT) (a fieldwork form)
A fieldwork form was developed to measure the factors affecting PLOS on footpaths, as
shown in Table 3. Some factors or items included in this part are evaluated by assessing the
current conditions for each item compared with the standard requirement in the pedestrian
guidelines. Other items are evaluated based on eligibility criteria developed according to the
previous attempts to create PLOS models. A paper form of the audit tool shown in the Table 3
was transferred to Google form to be used by phone or tablet.
A four point scale is allocated for each item, starting from point zero representing the worst
condition scenario (i.e., very unsafe/uncomfortable), to point 3 representing the best condition
scenario (i.e., very safe/comfortable). The allocated scale reflects how each item was built in
174
the real environment. The scale makes the comparison according to the standard requirement
and meets the needs of pedestrians for safety, comfort and mobility. The general information
for each audited segment, such as auditor ID, date of auditing, footpath segment ID and so on,
is also recorded. Based on the proposed method, the pedestrian level of service increases
whenever the point scale increases, reflecting the increase in footpath environment quality.
The allocated points for each item are then entered into the second part of the audit tool,
which is a ‘weighted assessment form’, used to calculate the value of PLOS.
3.2. Weighted assessment form
The weighted assessment form is the second part of the PLOS model which is used to
calculate the overall pedestrian level of service (PLOS). This part contains all 33 factors
included within the fieldwork form. The realistic weight (RW) of the factors affecting PLOS
derived from pedestrians by an intercept survey is calculated according to Eq.1, and includes
each factor in this form. Then, the factor assessment scores (FASs) gathered through a
fieldwork form on the audited footpath segments are transferred to the next column on this
form. The overall weighted assessment score (OWAS) is calculated using the following
equation (equation 2):
𝑛
𝑂𝑊𝐴𝑆 = ∑ 𝑅𝑊𝑖 × 𝐹𝐴𝑆𝑖

𝑖=1
Where, OWAS is the overall weighted assessment score, RW is the realistic weight for each
factor i, FAS is the factor assessment score assigned to each factor i, and n is the total number
of factors (in this study 33 factors). Finally, the overall PLOS is calculated for an audited
footpath by allocating the result of the overall weighted assessment score (OWAS) to the
PLOS grade from A (the best grade) to E (the worst grade) based on the scale developed for
this purpose (see Table 5).
175
Table 1 Pedestrian level of service audit tool (PLOSAT)
Auditor name -----------------------------, Auditor ID --------------, Date: -----------------------, Time of day: -----------------, Weather: ----------------
City: ----------------------, Suburb: -----------------------, Street name: ----------------------, Footpath side: ----------------, Segment ID: ---------------
Point System Method

Category Factor
0 1 2 3
Clear footpath width Less than 1.35m 1.35m- 2.25m 2.25m- 3.15 More than 3.15m
No crossings provided (hard to Insufficient crossings but in Insufficient crossings but in Sufficient crossings in logical
Crossing opportunities
cross) illogical locations logical locations locations

Quality of footpath surface Poor quality Moderate quality Good quality Excellent quality
Footpath slope & grade Extremely steep slope/gradient Steep slope/gradient Moderate slope/gradient Flat or gentle slope/gradient
Buffer area and width Less than 1.35m 1.35m- 1.65m 1.65m- 2.15m More than 2.15m
Sufficient separation in
Existing somewhere in
Pedestrian-traffic No separation and the segment Existing in the reasonable reasonable locations at
unreasonable locations in the
separation is needed locations but not sufficient adequate frequency or the path
segment
not needed
Insufficient kerb ramps and Sufficient kerb ramps but does
Sufficient kerb ramps, meets all
Kerb ramp No kerb ramps provided does not meet all required not meet all required
required specifications
specifications specifications
Clearance over footpath Less than 2.0m Equal to 2.0m 2.0m- 2- 2.5m Greater than 2.5m
Footpath connectivity No connectivity Low connectivity Moderate connectivity High connectivity

Obstructions and
interruptions
Footpath continuity Less than 50% completed 50%- 75% completed 75- 90% completed More than 90 % completed
Pedestrian crowding > Pedestrian crowding > Pedestrian crowding > 8ppmm
Pedestrian crowding ≤ 8ppmm,
26ppmm, movement is 100% 17ppmm≤ 26ppmm, movement ≤ 17ppmm, movement < 18% ≥
Easy of movement movement < 9% restricted so
restricted so very restricted ≤ 26% ≥ 18% restricted so 9% restricted so normal
free movement
movement restricted movement movement
Many obstructions (more than
Permanent obstructions Some obstruction (22-8 per km) Few obstructions (7-1per Km) No obstructions (0 per Km)
22 per km (11/500m))
176
Points System Method
Category Factor
0 1 2 3
Few awnings (less than 50% is Sufficient awnings (50-75% is Fully sufficient awnings (more
Awnings No awnings
covered) covered) than 75% is covered)
Sufficient trees/ plants along Fully sufficient trees/ plants
Few trees/ plants along the
the audited segment (more than along the audited segment
No or very few trees/ plants audited segment (25-75% is
Trees and plants 75% of the segment is covered) (more than 75% of the segment
(less than 25% is covered) covered) but does not meet all
but does not meet all required is covered) and meets all
specifications required specifications
Insufficient pedestrian scale Minimum sufficient pedestrian Full sufficient pedestrian scale
Shade No shade
shade scale shade shade
Type of crossing facility Poor Moderate Good Excellent
Activation of side facades Non-active frontage Poor active frontage Good active frontage Excellent active frontage
Sufficient TGSI and meets all

Insufficient TGSI and does not Sufficient TGSI but does not
Facilities for the disabled No TGSI provided required specifications or it is
meet all required specifications meet all required specifications
not needed
Footpath maintenance Poor Moderate Good Excellent
Footpath cleanliness Poor Fair Good Excellent
Fully sufficient surveillance
Footpath surveillance No surveillance means Few surveillance means Sufficient surveillance means
means
Existing somewhere but does Existing sufficient frequency Full sufficient frequency and
Garbage facilities No rubbish bin not meet all required but does not meet all required meets all required
specifications specifications specifications
Existing somewhere but Sufficient frequency but does Full sufficient frequency and
Pedestrian signs No signage insufficient and does not meet not meet all required meets all required
all required specifications specifications specifications
Insufficient pedestrian scale Sufficient pedestrian scale Full sufficient road oriented
Lighting No artificial lighting
lighting lighting lighting
Insufficient bus stops and does Sufficient bus stops but does
Public transportation Sufficient bus stops and meets
No bus stops provided not meet all required not meet all required
facilities all required specifications
Existing somewhere but does Sufficient frequency but does Full sufficient frequency and
Wayfinding guidance No wayfinding facility not meet all required not meet all required meets all required
177
Point System Method
Category Factor
0 1 2 3
Pedestrian evasive
Massive evasive manoeuvres Moderate evasive manoeuvres Normal evasive manoeuvres No evasive manoeuvres
Pedestrian
manoeuvres
traffic
Pedestrian conflict points Poor Moderate Reasonable Perfect

Pedestrian speed Very restricted walking speed Restricted walking speed Normal walking speed Free walking speed
Pedestrian volume Very crowded path Moderately crowded path Uncrowded path Very uncrowded path
Total traffic-pedestrian conflict Total traffic-pedestrian conflict

Adjacent
Vehicle conflict points points more than 16 per km points 5- 1 per km as

traffic
points 16- 6 as moderate points = 0 as perfect

(8/500m) as poor reasonable
Vehicle speed More than 60 km/h 60 km/h- 40 km/h 40 km/h- 30 km/h Less than 30 km/h
Vehicle volume High traffic volume Moderate traffic volume Normal traffic volume Low traffic volume
178
Table 2 The weight assessment form (WAF) for the factors affecting PLOS
Realistic Factor Weighted
Categoris
Group Category Factor weight assessment assessment
ed PLOS
(RW) score (FAS) score (WAS)
Footpath width 3.2
Geometric characteristics
Crossing opportunities 3.2
Quality of footpath surface 3.2
Footpath slope & grade 3.1
24%
Buffer area and width 3.2

Pedestrian-traffic separation 3.1
Ramp 3.1
Factor related to the pedestrian environment
Clearance over footpath 1.9

Footpath connectivity 3.1
interruption
Obstruction
12.6%
Footpath continuity 3.2

s and
Easy of movement 3.2

Permanent obstructions 3.2
Awnings 2.9
Shade 3.2
Type of crossing facility 3.0
Activation of side facades 2.5
Aesthetics and amenities
Facilities for the disabled 3.2

Footpath maintenance 3.2
Footpath cleanliness
42.5%
3.2
Footpath surveillance 2.3
Garbage facilities 3.2
Pedestrian signs 3.2
Lighting 3.2
Public transportation facilities 3.2
Trees and plants 3.2
Wayfinding guidance 3.1
Evasive movements 2.7
traffic system and user
Pedestrian
Factors related to the
Pedestrian conflict points

11.6%
traffic
3.2
Pedestrian speed 2.7
Pedestrian volume 3.1
Vehicle conflict points 3.2
Adjacen
t traffic
9.3%
Vehicle speed 3.1

Vehicle volume 3.1
Overall weighted assessment score (OWAS)
Results
Overall PLOS
PLOS grade
179
3.3. Final report of auditing pedestrian level of service (PLOS)
The overall PLOS score calculated at the end of this process is then graded from A to F, as
shown in Table 3. A number of possible recommendations and remedial actions are also
shown. Figures 2 and 3 present a copy of the final report template and a table of a summary
of the current situation, recommendations and remedial action plans required for the audited
footpaths.
Figure 1 Final report for auditing pedestrian level of service (PLOS)
180
Figure 2 Summary of the current situation, recommendations and remedial action plan required for
audited footpath (only the first page shown here as an example).
3.4. Overall pedestrian level of service scale

to audited footpaths. On the basis of the LOS grades adopted in the HCM scale, in this study a
PLOS range is allocated to one of the grades from A to F based on the point system method.
These grades are related to the current situation of footpaths regarding the adequacy of
footpaths’ aspects to provide a high level of comfort, safety and mobility, and are intrinsically
related to all 33 factors affecting PLOS. Table 3 presents the footpath conditions, the
proposed criteria and the range of scores, and the acceptability level. The range of maximum
and minimum weighted assessment score might be between 0 and 300, based on the current
situation drawn by the value of the obtained points.
The formulation of Table 3 to identify the intervals for each range of PLOS grades can be
derived by dividing the maximum value of the weighted assessment score (in this study 300)
by the total number of PLOS grades (in this study 6). Based on the point system method, the
PLOS grades can be defined based on a different range of scores as follows:
 PLOS A: (score ≥ 250) indicates that the footpath is likely in excellent condition,
provides a high level of comfort, safety and mobility for pedestrians, and key factors
positively affect PLOS.
 PLOS B: (score 200-249) indicates that the footpath is likely in good condition, with
only a few aspects of the footpath environment (factors affecting PLOS) negatively
affecting pedestrian safety, comfort and mobility. Limited effort is likely to be needed
for improvement.
181
 PLOS C: (score 150-199) indicates that the condition of the footpath is moderate, and
some aspects of the footpath environment (factors affecting PLOS) negatively affect
pedestrian safety, comfort and mobility. Some procedures and efforts are needed for
improvement.
 PLOS D: (score 100-149) indicates that the footpath is in basic condition, and
significant aspects of the footpath environment (factors affecting PLOS) negatively
affect pedestrian safety, comfort and mobility. Significant effort is likely to be needed
for improvement.
 PLOS E: (score 50- 99) indicates that the footpath is in poor condition, with many
aspects of the footpath environment (factors affecting PLOS) negatively affecting
pedestrian safety, comfort and mobility. Much effort is likely to be needed for
improvement.
 PLOS F: (score < 50) indicates that the footpath is in extremely poor condition, with
almost all the aspects of the footpath environment (factors affecting PLOS) negatively
affect pedestrian safety, comfort and mobility.
Table 3 Overall Pedestrian Level of Service (PLOS) grade scale

PLOS Grade PLOS Score Conditions
A ≥ 250 Excellent
B 200-249 Good
C 150-199 Regular
D 100-149 Basic
E 50-99 Poor
F < 50 Appalling
Based on the way used to identify the overall PLOS grades, Table 4 shows a range of PLOS
grades for each category of the factors affecting PLOS.
Table 4 PLOS grade scale of each category of the factors affecting PLOS
Categorical PLOS grade
PLOS factors’ category
A B C D E F
Geometric characteristics > 62 51-62 37-50 25-36 13-24 ≤ 12
Obstructions and interruptions > 32 27-32 21-26 15-20 9-14 ≤8
Aesthetics and amenities > 107 87-107 66-86 45-65 24-44 ≤ 23
Pedestrian traffic > 32 26-32 18-26 12-17 6-11 ≤5
Adjacent traffic > 25 19-25 12-18 8-12 4-8 ≤3
182
4. Guidance manual
This section displays the guidance manual which was prepared for raters to provide specific
information relating to the evaluation of pedestrian level of service (PLOS) on footpaths in
commercial streets. The manual provides instructions on how to organise and conduct an
onsite audit, the standards for pedestrian facilities and how to use the forms. Many national
and international policies, guidelines, specifications and audit tools were reviewed and
considered before developing this guideline in order to achieve the best standards for planning
and designing pedestrian facilities. All numbers, specifications and criteria mentioned in this
guideline are based on Australian standards and are appropriate for the Australian context.
4.1. Footpath selection
Before beginning the pedestrian level of service audit, the specific footpaths that are required
to be audited should be identified. The auditable footpaths should be drawn on a map of a
targeted area which can be produced simply via Google Maps. The auditable footpaths should
be subdivided into a set of segments with almost similar characteristics, in order to assess and
rate each segment individually. Auditors can gather the basic information about the auditable
area by driving around the area or viewing Google Maps. It is recommended that the
segments be split at logical points, such as a section of a street or a road between two
intersections or the end of a footpath, and each segment must have a unique number that is
used to identify it. The auditing process should be conducted by foot to discover all the issues
associated with the area. Ideally, footpaths on both sides of the street should be audited, given
that conditions can differ greatly on either side.
4.2. Required equipment
The auditor should ensure that he or she has the following tools required to conduct the
pedestrian level of service audit:
 Mobile phone or tablet to use online Google form of the tool.
 Two maps – one with street names and one with segment numbers, clearly showing
the audit footpath segments.
 Quick guidelines to use PLOSAT in the field work.
 Identification card.
 Water bottle.
 Tape measure.
 Camera.
 Stopwatch/timer.
183
 Comfortable shoes.
 Sun protection (eg., hat, sun screen, sunglasses).
4.3. General rules for the audit process
 Ensure that auditors are familiar and comfortable with the forms, and practice using
the forms before auditing begins.
 Ensure that two maps are provided for each footpath audited, one with street names
and the other with segment numbers. Both maps of each suburb have a unique ID.
 Ensure auditors begin at the starting point identified on map before leaving the office.
 Identify which side of the road the audit will begin before leaving the office. Code the
first side as number 1; code the opposite side as number 2.
 Clearly tick the targeted choice in blue biro to avoid later misunderstandings.
 In case of accidentally ticking the wrong choice, make sure to clearly cross out the
incorrect answer and tick the correct one.
 If any item is skipped, make sure to return to it by writing a reminder note in the notes
field. If the item is crossed out, indicate why in the notes field.
 Record any additional comments in the comments field. This can include such things
as the difficulties faced, problems with using the audit tool, modifications which
should be done on the audit tool, and so on.
 Try to complete the auditing of each segment in one shift. If that is impossible,
accurately identify the stopping point and record this point on the map, so as to be
clear about where to start when next in the field.
 Ensure complete and correct rating of the audit forms to obtain high quality results.
 The audit should be conducted at times when pedestrian volumes are at their highest
 If the audit process needs multiple shifts for more than one day, check each audit form
at the end of each day and check that all parts are complete, clear and legible.
 If the audit process takes a lot of time to be completed for a large area, deliver all
relevant documents related to the audit process weekly to the authorised person for
checking and processing to avoid any mistakes at the early stage of the audit process.
 Accurately transfer the data collected using the PLOSAT form for all segments to the
weighted assessment form (WAF) to calculate an overall pedestrian level of service of
the audited footpath.
184
 Organise the completed PLOSAT forms for each segment in numerical order (starting
from the first segment audited to the last one) along with relevant maps, the relevant
weighted assessment form (WAF) for each segment (in segment number order) and
the overall assessment audit forms.
 If any problems are encountered during the audit trip, contact the authorised person in
the organisation.
4.4. Guidelines for defining and measuring each factor affecting PLOS at audited sites
The following section of the protocol describes each PLOS factor, how to measure it in the
field and how to identify the response category to aid the assessor in dealing with variations in
the environment. The raters are encouraged to read through this section and use it as a
reference during the process of auditing the footpath segments.
4.4.1. Footpath width (pedestrian through zone)
This factor may reflect different attitudes to pedestrian comfort level associated with the level
of crowding. It is clear that footpath width directly relates to pedestrian flow/volume.
Footpath width influences the majority of other factors affecting PLOS and is directly related
to aspects of safety (personal and traffic), comfort (psychological and physical), and mobility
in the pedestrian environment. Footpaths should be constructed with/at a suitable width to
provide the required level of safety, comfort, and mobility for pedestrians and other eligible
road users. The minimum required path width varies based on the location, purpose, and
expected demand or pedestrian volume (Western Australia Department of Transport, Planning
& Public Transport Authority, 2011). Also, a suitable footpath width not only provides a clear
access route for pedestrians but provides space for all street furniture (such as garbage
facilities, guidance signs, lighting, public transportation facilities, facilities for the disabled,
trees and plants, wayfinding guidance, etc.). These features work as temporary or permanent
obstructions that reduce the effective width for pedestrians. Therefore, the total footpath width
must have enough space to accommodate pedestrian trips alongside these other elements. In
general, the road guidelines dividing the total footpath width for four distinct zones within
footpaths include: 1) a kerb zone, which defines the limit of the pedestrian environment and
is a major tactile cue for visually impaired pedestrians; 2) a street furniture zone or buffer area
which is used to accommodate all street furniture elements such as soft landscaping,
vegetation, light poles, traffic signal poles, seats, etc.; 3) a property frontage zone which
contains retaining walls, fences, space for pedestrians emerging from buildings, ‘window
shoppers’ or overhanging vegetation and cannot be used for pedestrians; and 4) clear footpath
185
width or a through-route zone which is used for pedestrians and which should be kept free
from obstructions, particularly for those using wheelchairs and those with low vision
(Austroads, 2017).
It is important to distinguish between the total width and the width of the zone likely to be
used by pedestrians (travel footpath width or through route) when auditing footpaths to
achieve accurate PLOS results. Only the clear footpath width or the through-route zone with
the property frontage width are being assessed, according to the standard width in the
Australian road guidelines. The other zones were checked within the next part of the buffer
area factor.
Figure 3 Examples of footpath zones.

Source: (NZ Transport Agency, 2009)
Answering the following questions will help in making decisions:
 What is the average clear footpath width along the audited footpath segment?
 Is the present footpath width clear for the pedestrian volumes and type of users?
These steps should be followed on the audit form:
 Measure a clear footpath width for each 20m length along the audited segments.
 Calculate the average width for the audited segment (using a tape measure).
 Consider the pedestrian volume for each audited area, using dependable statistics if
available or a field survey if needed.
 Compare the obtained results of the average width with the standard width provided
by the Australian road guidelines, based on the location, purpose, and expected
demand or pedestrian volume of the audited area.
Notes:
 The minimum and desired width for the pedestrian through-route zones is 2.4m (or
higher based on demand). In commercial or shopping environments the maximum
pedestrian flow is 80 pedestrians/ minute.
186
 In busy alfresco dining areas such as the central city area, a minimum through-route of
3m - 4m should be provided, reduced to 2.5m in areas with less pedestrian traffic.
 The minimum and desired width for a property frontage zone is 0.75m which is being
added to the average of the clear footpath width.
Select the choice of the clear footpath width condition for the audited footpath segment based
on the following criteria:
 If the total clear footpath width+ frontage width is ≤ 1.2+0.15, select 0.
 If the total clear footpath width+ frontage width is < 1.8+ 0.45≥ 1.2+0.15, select 1.
 If the total clear footpath width + frontage width is < 2.4+0.75≥ 1.8+ 0.45, select 2.
 If the total clear footpath width+ frontage width is ≥ 2.4+0.75, select 3.
4.4.2. Crossing opportunities
Crossing opportunities refer to the number of facilities or available opportunities such as
median refuges, pedestrian crossings, guarded crossings, crosswalks, underpasses, and
overpasses provided along the footpath to assist pedestrians to safely cross roads and paths.
One of the main reasons people are discouraged from walking is because they are unable to
safely cross roads, or because of delays in crossing the roads. Pedestrians usually prefer to
travel using the quickest and most direct means to cross roads. Limited crossing opportunities
can make pedestrians walk outside the provided pedestrian infrastructure at crossings, even if
these crossings do not provide the required level of safety (Gallin, 2001).
Figure 4 Crossing opportunity

Source: (Western Australia Department of Transport et al., 2011).
Answering the following questions will help in making decisions:

 Are sufficient crossing opportunities provided along the audited footpath?
 Are crossing opportunities located at logical locations along the audited footpath?
187
 Do pedestrians correctly use the crossing opportunity provided along the audited
footpath?
Then the crossing opportunity conditions should be estimated for the audited footpath
segment based on the following criteria:
 If there are no crossings provided (hard to cross) in the segment, select 0.
 If insufficient crossings are placed in illogical locations in the segment, select 1.
 If insufficient crossings are placed in logical locations, select 2.
 If sufficient crossings are placed in logical locations, select 3.
4.4.3. Quality of footpath surface
This factor refers to the quality of the surface of the path in terms of general design (firm,
stable and slip resistant even when wet), decorative surfacing and materials. The quality of
footpath surface relates both to the original design of the footpath and the maintenance factor
(whether enough repairs have been undertaken). A footpath should be designed in such a way
that the surface is smooth, slip resistant, free of uneven levels, tripping hazards, minor uplifts,
and lips with paving are of high quality material (Western Australia Department of Transport
et al., 2011) (Transport, 2005) (NZ Transport Agency, 2009).
To identify choices on the audit form, the following questions should be answered:
 Is the walking surface of the footpath appropriately designed?
 Are the pedestrian crossovers smooth?
 Is the texture of the surface traversable by people who use a wheelchair and those with
an ambulant or sensory disability?
To identify choices on the audit form, the following conditions should be understood:
 Design transition shall be 0mm. Construction tolerances shall be as follows:
1. 0-3mm vertical
2. (b) 0-5mm, provided the edges have a bevelled or rounded edge to reduce the likelihood
of tripping. Any gaps, lips, joins or changes in height cannot exceed 5mm as these can
create trip hazards.
Figure 5 Measuring the maximum deviation of the surface

Source: (NZ Transport Agency, 2009).
188
 A poor path is one with a massive roughness and non-slip resistance which contains a
lot of uneven levels, many tripping hazards, major uplifts and lips with paving using
poor quality material.
 A moderate path is one with a little smoothness and low slip resistance which
contains few uneven levels, few tripping hazards, few uplifts, and has lips with paving
with moderate quality materials.
 A good path is one with a smooth and slip resistant surface which contains very few
uneven levels, very few tripping hazards and very few uplifts, and has lips with paving
with good quality materials.
 An excellent path is one which is perfectly smooth and which has high slip
resistance. It does not contain uneven levels, tripping hazards or uplifts, and has lips
with paving with excellent quality materials.
Excellent quality Poor quality

Figure 6 Quality of footpath surface
The quality of the surface conditions of the audited footpath segment should then be estimated
based on the following criteria:
 If the segment condition seems of poor quality, select 0.
 If the segment condition seems of moderate quality, select 1.
 If the segment condition seems of good quality, select 2.
 If the segment condition seems of excellent quality, select 3.
4.4.4. Footpath slope and grade
The gradient of a through route is the slope parallel to the direction of travel where the
movement becomes more difficult as the gradient increases. In contrast, the slope or crossfall
refers to a slope of the footpath at right angles to the direction of travel (NZ Transport
Agency, 2009). Slope and gradient are required for engineering design purposes, such as
facilitating the drainage processes, the natural topography of the area, and so on, but steep
walking surfaces (both gradients and slopes) in the through footpath can affect pedestrians’
189
stability and control, especially those who use wheelchairs and walking frames, forcing them
to expend more energy to resist the sideways forces (Western Australia Department of
Transport et al., 2011). Therefore, footpaths should be designed to minimize slopes or provide
intermittent level landings (U.S. Department of Transportation, 2007).
Figure 7 Correct and incorrect provision of crossfall

To identify choices on the audit form, these steps should be followed:
 Check the slope condition of the footpath segment paralleled to the roadway.
 Check the gradient condition of the footpath segment perpendicular to the roadway.
Often this becomes a problem at driveways, especially for pedestrians in wheelchairs
who must maintain a straight path.
 Assess the slope and gradient based on the conditions of the majority of the segment.
For example, if the beginning of a segment has a moderate slope but the majority is
flat, score it as flat.
 Assess the slope and gradient based on the prevalent conditions if the segment is
evenly divided between two gradients. For example, if the segment has both a slight
and a moderate slope, then score it as moderate.
Then estimate the slope and gradient conditions for the audited footpath segment based on the
following criteria:
 If the audited segment has an extremely steep slope/gradient, select 0 (an extremely
steep slope is one with an extremely sharp or extremely rapid incline).
 If the audited segment has a steep slope/gradient, select 1 (a steep slope is one with a
sharp or rapid incline).
 If the audited segment has a moderate slope/gradient, select 2 (a moderate condition is
one with a medium incline).
190
 If the audited segment has a flat or gentle slope/gradient, select 3 (a flat condition is
one with a gradual incline).
4.4.5. Buffer width
The buffer area or street furniture zone is the distance between a road and a pedestrian
footpath. It is used to accommodate all street furniture elements and plants. The buffer width
is measured from the back of the curb to the edge of the footpath closest to the road
(Austroads, 2017). A sufficient buffer width between roadways and pedestrian pathways can
contribute to making pedestrians feel more comfortable and safer by providing lateral
separation between pedestrians and vehicular traffic. Buffers can also improve accessibility
by providing space for level sidewalk crossings across driveways. When landscaped, buffers
may also deter pedestrians from crossing roads at unsafe locations. Wide buffers reduce
pedestrians’ exposure to vehicle spray (as well as puddle splash) during wet weather, and also
provide space for snow storage, both of which promote pedestrians’ use of the sidewalk.
The auditor should check the adequacy of the width of the buffer. In this part, the buffer width
or street furniture zone width alongside the kerb zone width are being assessed according to
the standard width in the Australian road guidelines.
Figure 8 Buffer width

 Measure a buffer width for each 20m length along the audited segment.
 Calculate the average width for the audited segment.
 Compare the obtained results of the average width with the standard width provided
by the Australian road guidelines based on the location, purpose, and expected
demand or pedestrian volume for the audited area.
Notes:
 The desired width for the buffer zones is 2m for commercial and shopping
environments or city streets.
 The desired width for the kerb zone is 0.15m which is being added to the average of
the buffer width.
191
Then, select the buffer width condition for the audited footpath segment based on the
following criteria:
 If the total buffer width+ kerb zone width is ≤ 1.2+0.15 or there is no buffer zone,
select 0.
 If the total buffer width+ kerb zone width is < 1.5+ 0.15≥ 1.2+0.15, select 1.
 If the total buffer width + kerb zone width is < 2m+0.15m ≥ 1.5+ 0.15, select 2.
 If the total buffer width+ kerb zone width is ≥ 2m+0.15m, select 3.
4.4.6. Pedestrian traffic separation
Pedestrian traffic separation is the separation of pedestrian traffic from vehicle traffic lines.
Vehicular traffic should be restricted in pedestrian facilities using various elements such as
fences, bollards on street parking or buffer areas. For example, using fencing within a
commercial streetscape can enhance a safety feature or be used as a functional support rail for
people to lean against. Railings and fences can help define the street space. Also, buffers can
positively enhance both the perceived and actual safety of footpaths by providing lateral
separation between pedestrians and vehicular traffic. Bicycle lanes and on street parking areas
may also help to separate pedestrians from vehicular traffic. Footpaths are occasionally
designed with only a curb to separate pedestrians from the vehicular traffic. This measure
alone is often inadequate, as the curb does not form an adequate barrier between vehicular and
pedestrian traffic, especially in areas where vehicles travel at speeds over 40kph (U.S.
Department of Transportation, 2007).
Sufficient separation NO separation
192
Separation by bollards
Figure 9 Pedestrian traffic separation
Source: (City of Sydney Council, 2013; NZ Transport Agency, 2009; U.S. Department of
Transportation, 2007)
To identify choices on the audit form, answers to the following questions will help in decision
making:
 Is separation urgently needed at the audited segment?
 Is any kind of separation provided between motorists and pedestrians?
 Does the pathway provide enough separation from high-speed, high-volume traffic?
 Is there adequate separation between vehicular traffic and pedestrians at the expected
conflict points between pedestrians, cyclists, and motorists?
Then the pedestrian traffic separation conditions for the audited footpath segment should be
estimated, based on following criteria:
 If there is no separation and it is needed in a segment, select 0.
 If there is separation somewhere in unreasonable locations in the segment, select 1.
 If there is separation in reasonable locations but they are insufficient, select 2.
 If there is a sufficient separation in reasonable locations at adequate frequency, or a
footpath is not needed, select 3.
4.4.7. Kerb ramp
Kerb ramps are sloped sections which link footpaths to roads by providing a smooth transition
from the footpath height to street level. Kerb ramps are imperative in increasing the safety and
accessibility of footpaths and crossing facilities by providing a continuous path of travel for
all pedestrians, especially for people with mobility impairment, with wheelchairs or prams
(Western Australia Department of Transport et al., 2011).
Ramps should be located perpendicular to the direction of travel on either side of a road and
should be aligned with each other to ensure the provision of a continuous pedestrian network.
The use of specific colours to identify kerb ramps is a way of signalling the preferred crossing
193
location for all pedestrians on footpaths. Appropriate materials should be used in building the
most suitable surface for a kerb ramp to ensure that a firm, even, stable, slip resistant surface
is provided. The uneven surface of a kerb ramp, or even level changes between the street and
the kerb ramp, may leave gaps or cracks that cause a hazard or obstruction for wheelchairs,
scooters or prams. These gaps or cracks also increase the efforts of pedestrians with
disabilities to get over or around them, causing delays in the crossing time, thereby increasing
the pedestrian’s exposure to oncoming traffic (Austroads, 2017; Western Australia
Department of Transport et al., 2011).
Figure 10 Kerb ramp

Source: (Western Australia Department of Transport, 2011; Western Australia Department of
Transport et al., 2011).
Answers to the following questions will help in making decisions:
 Are curb ramps provided for each crossing area?
 Are ramps aligned with the crossing area, so that pedestrians (especially those with
low vision) are correctly oriented when entering the crossing area from the ramp?
 Is a level area of sufficient size at the top of the ramp to allow wheelchair users (or
people using walking aids) to wait or manoeuvre without the risk of rolling into the
roadway?
 Are kerb ramps flush with the street surfaces (i.e., no lip or gap)?
 Are gaps, cracks and discontinuities between kerb ramps and street surfaces avoided
at each crossing?
 Is water pooling on pedestrian ramps, making walking uncomfortable and unsafe,
especially if pools of water stand for prolonged periods?
To identify choices on the audit form, kerb ramps having the following specifications should
be considered:
 There should not be there any level differences between a kerb ramp on a footpath and
an adjacent roadway.
194
 A smooth change in the level between the footpath and road pavement should be
provided.
 A minimum footpath width of 1.33m beyond the top of the ramp should be provided.
 A gradient of 1:8 should not be exceeded, however, a more desirable gradient is 1:10
or less.
 Total length of 1.5m should not be exceeded.
 The kerb ramp should be flush with the adjacent footpath, street surfaces, and gutter
(i.e., no lip or gap).
 The ramp should be located perpendicular to the direction of travel.
 The ramp should have a warning style tactile ground surface indicator (TGSI) at each
crossing point.
 The ramp should be at least 0.9m wide.
 The ramp should have a firm and stable surface.
 The ramp should be located within the corner area of the footpath, aligned with the
street wall, and set at a straight angle to the street alignment.
Then, the kerb ramp conditions for the audited footpath segment should be estimated, based
 If no kerb ramps are provided in the segment, select 0.
 If there are insufficient kerb ramps and they do not meet all required specifications,
select 1.
 If there are sufficient kerb ramps but they do not meet all required specifications,
select 2.
 If there are sufficient kerb ramps which meet all required specifications, select 3.
4.4.8. Height clearance
Height clearance is a vertical (overhead) clearance over the entire footpath width (including
the street furniture and frontage zones) to prevent head injuries to pedestrians (NZ Transport
Agency, 2009). Adequate vertical clearance should be free of all obstructions, such as
vegetation, signs, shop awnings and low verandahs. An absolute minimum height clearance of
2.0m above the accessible path of travel is required and the ideal height clearance equal to
2.5m is preferred (Western Australia Department of Transport et al., 2011).
195
Figure 11 Height or clearance above footpaths
To identify choices on the audit form, check the height clearances of shop awnings and low
verandahs along the audited footpath segment.
Then estimate the height clearance conditions for the audited footpath segment based on the
following criteria:
 If the height clearance is less than 2.0m, select 0.
 If the height clearance is = 2.0m, select 1.
 If the height clearance is 2.0m- 2.5m, select 2.
 If the height clearance is greater than 2.5m, select 3.
4.4.9. Footpath connectivity
Footpath connectivity is defined as the directness of links and the density of connections in a
footpath network. It refers to the degree to which the segment footpath provides a useful,
direct and logical link between crosswalks and other footpaths (Clifton et al., 2007).
The connectivity of the footpath to other footpaths or crosswalks and key destinations in the
area to form a pedestrian network is very important for the walkability of an area. A
discontinuous network might be unpleasant for wheelchairs, scooters, or for all users if a
footpath suddenly ‘ends’ because it may either force users to retrace their path and find
another route, or cross the road at an unsafe crossing point. This might cause decreasing
pedestrian level of service on a footpath, making regular users unwilling to use this kind of
path as a result of poor pedestrian access along the route (Western Australia Department of
Transport, 2011).
196
Low connectivity High connectivity
Figure 12 An example of connectivity for two suburb types
Source: http://www.walkscore.com/walkable-neighborhoods.shtml
To identify choices on the audit form, check the number of connections on the footpath
segment to crosswalks and other footpaths in the following way:
 Stop at the beginning of the targeted segment and look backward, right and left. Then
check and identify how many footpaths or crosswalks are connected.
 Stop at the middle of the targeted segment and look right and left. Then check and
identify how many footpaths or crosswalks are connected.
 Walk to the end of the targeted segment and look forward, right and left. Then check
and identify how many footpaths or crosswalks are connected.
 Stop signs at the end of the segment can be treated as crosswalks.
 Count the number of connections to other footpaths and crosswalks and calculate the
total connectivity score by summing the number of connections of all parts.
Figure 13 Number of connections to other footpaths and crosswalks

Source: (Clifton et al., 2007).
Answers to the following questions will help in making decisions:
 Is the footpath connected to other pedestrian facilities to form a pedestrian network?
197
 Is the footpath connected to key destinations along the route (like public transport
stops, schools, shops, parks, community centres, offices)?
Then, estimate the connectivity condition for the audited footpath segment based on the
following criteria:
 If the total connectivity score is 1 or fewer, select 0 as no connectivity.
 If the total connectivity score is 2- 4, select 1 as low connectivity.
 If the total connectivity score is 5- 7, select 2 as moderate connectivity.
 If the total connectivity score is more than 7, select 3 as high connectivity.
4.4.10. Footpath continuity/ completeness
Footpath continuity refers to the completeness of the footpath within a road network or within
the segment. A footpath is considered complete if it is continuous along one side of a segment
without any gaps, missing pieces or breaks within the segment. Otherwise, if there is a gap or
missing pavement within the segment, it should be categorized as incomplete or partly
complete. Missing sections or unpaved areas might prevent passage by pedestrians with
mobility impairment and may be impassable for other pedestrians. That might cause
pedestrians to walk outside footpaths onto the roadway, where they may come into conflict
with motorists and cyclists, or cause them to walk through paved parking areas, where they
may come into conflict with parking, entering or departing vehicles (Pikora et al., 2002)
(Moudon & Lee, 2003).
Complete path Incomplete path Incomplete path

Figure 14 Footpath continuity/ Completeness
To identify choices on the audit form, answer the following questions:
 Does a footpath have any breaks within the segment?
 Does a footpath have any gap or missing pavement pieces within the segment?
 Is the footpath continuous; does it have any missing sections?
198
Then, estimate the continuity/ completeness conditions of the footpath segment based on the
following criteria:
 If segment condition seems ≤ 50% continuous/complete (e.g., incomplete), select 1.
 If segment condition seems > 50% ≤ 75% continuous/complete, select 2.
 If segment condition seems > 75 ≤ 99% continuous/complete, select 3.
 If segment condition seems 100% continuous/complete, select 4.
4.4.11. Ease of movement
Ease of movement refers to the ability of a pedestrian to walk on a path without interruptions
or delays, for example, as experienced at signalized and unsignalized intersections, or when
pedestrians do not travel in clear ‘lanes’ although they may sometimes do so in heavy flow
conditions. The method of measuring ease of movement is based on the restriction indicator
of pedestrians’ movement, established by London Transport to measure pedestrian comfort
(Transport for London, 2010).
Movement < 9% restricted Movement < 18% ≥ 9% restricted
Movement ≤ 26% ≥ 18% restricted Movement is 100% restricted

Figure 15 Easy of movement condition.
Source: (Transport for London, 2010).
Answers to the following questions will help in decision making:
 Is there ample space for pedestrians to walk on their chosen route?
 Does the current path provide enough space for some choice in routes taken?
 Do pedestrians have very little personal space and restricted movement? Do
pedestrians experience extreme difficulty bypassing slower pedestrians or moving in
reverse flows?
199
To identify choices on the audit form:
 Examine the results of the pedestrian volume factor in each segment.
 Use subjective assessment to determine the restriction indicator in personage for the
audited segment.
 Make a decision taking into consideration previous conditions, based on the next
criteria.
Then, estimate the ease of movement conditions for the audited footpath segment based on
the following criteria:
 If the pedestrian crowding is > 26ppmm, movement is 100% restricted, select 0 as
very restricted movement.
 If the pedestrian crowding is > 17ppmm≤ 26ppmm, movement is ≤ 26% ≥ 18%
restricted, select 1 as restricted movement.
 If the pedestrian crowding is > 8ppmm ≤ 17ppmm, movement is< 18% ≥ 9%
restricted, select 2 as normal movement.
 If the pedestrian crowding is ≤ 8ppmm, movement is < 9% restricted, select 3 as free
movement.
4.4.12. Permanent obstructions
Measurement of the number of obstructions per width per distance segment on a footpath
being assessed is essential in determining whether access is available and does not impede
travel along pedestrian pathways for all users, especially for those people with disabilities or
in wheelchairs (U.S. Department of Transportation, 2007). Permanent obstructions are, for
example, street signs, bus stop shelters, street furniture such as rubbish bins and benches,
trees, bushes, letter boxes, poles, parking meters, lamp columns or street art (Western
Australia Department of Transport, 2011). This factor negatively affects the level of service of
footpaths by decreasing the effective width of the footpath, making the path insufficient to
accommodate pedestrian trips. However, some permanent obstructions do not cause any
impediment to users if the remaining width of the footpath is still sufficient.
200
Figure 16 Permanent obstructions
To identify choices on the audit form, follow these steps:
 Identify the existence of items that are considered to be permanent obstructions on the
audited footpath segment.
 At the location of each obstruction, measure the minimum effective width of the path.
 Evaluate whether the effective width is still sufficient for the pedestrian volumes and
types of users on the path, according to the recommended footpath width identified in
the footpath width factor.
 If the effective width is insufficient, count this item as being a permanent obstruction;
otherwise do not consider this item to be an obstruction.
 Add the number of items that are considered as permanent obstructions on the audited
footpath segment and save the score to calculate the total score of permanent
obstructions for the audited segment.
Then estimate the permanent obstruction conditions for the footpath segment based on the
following criteria:
 If the total permanent obstructions score is more than 11 pieces per segment, select 0.
 If the total permanent obstructions score is 5- 11 pieces per segment, select 1.
 If the total permanent obstructions score is 1- 4 pieces per segment 8 per km, select 2.
 If the total permanent obstructions score is = 0 (no obstructions), select 3.
4.4.13. Awnings
Awnings are constructed from private or public buildings and extend to the public domain and
the urban block along footpaths. Buildings along footpaths such as shops, cafes, hotels, and
other buildings are encouraged to provide protective awnings to create shade, protection from
201
rain and snow for pedestrians and frontages, and to otherwise add colour and attractiveness to
the street. Awnings are especially important in hot climates. In addition to these functions,
awnings also contribute to reducing the perceived scale of tall buildings and focus pedestrian
views on street frontages and ground level activities.
Figure 17 Awnings
Source: (Western Australia Department of Transport, 2011).
To identify choices on the audit form, estimate the awning conditions for the audited footpath
 If there are no awnings along the audited segment, select 0.
 If there are few awnings along the audited segment (less than 50% is covered), select
1.
 If there are sufficient awnings along the audited segment (50-75% is covered), select
2.
 If there are fully sufficient awnings along the audited segment (more than 75% is
covered), select 3.
4.4.14. Tree and plants
Trees, plants, shrubs, and other well-tended landscaping along pedestrian pathways can create
an attractive visual environment and a ‘buffer’ to create comfortable separation between
vehicular traffic and footpaths, forming a safe, intimate, and attractive environment for
pedestrians. They also provide enhancement of the air quality and water run-off, creating
important environmental benefits, as well as a protecting pedestrians from weather by
providing shade and shelter, making walking more comfortable and attractive (Western
Australia Department of Transport et al., 2011; NZ Transport Agency, 2009).
202
Figure 18 Trees and plants
Source: (City of Sydney Council, 2013; NZ Transport Agency, 2009).
To identify choices on the audit form, consider whether the trees and plants have the
following specifications:
 Trees and plants do not intrude into the clear continuous path of travel. Ideally, they
should be placed in the street furniture zone within the footpath cross-section in order
to keep the pedestrian through route free (preferably 1.5m or wider) from obstruction.
 Trees and plants should be placed in a location that does not obscure pedestrian or
driver visibility when installed or when mature.
 Trees should be five metres tall at installation.
 Trees and plants should be placed within four metres of the edge of the nearest traffic
lane.
Then, estimate the conditions of the trees and plants on the audited footpath segment based on
 If there are no/very few trees/plants along the audited segment (less than 25% is
covered), select 0.
 If there are few trees/plants along the audited segment (between 25% and 75% of the
segment is covered) but they not meet all required specifications, select 1.
 If there are sufficient trees/ plants along the audited segment (more than 75% of the
segment is covered) but they not meet all required specifications, select 2.
 If there are fully sufficient trees/ plants along the audited segment (more than 75% of
the segment is covered) and they meet all required specifications, select 3.
4.4.15. Shade
Shade refers to the sufficiency of shade and shelter provided on a path, helping to make the
walkable environment more comfortable and more accessible by protecting pedestrians from
heat, rain and other elements. The attractiveness of an area for walking also depends on the
203
provision of shaded areas (Western Australia Department of Transport, 2011). The auditor
should check whether sufficient shade is provided for the footpath and for waiting and resting
areas. Large trees provide excellent shade; verandahs, shade cloth or sails and other types of
shelters are useful in places where trees are not appropriate.
Figure 19 Shade by trees and/or awnings

making:
 Do trees or built structures provide shade along the footpath?
 Do trees or built structures provide shade at resting places and places with street
furniture?
Then, estimate the shade conditions for the footpath segment based on the following criteria:
 If there is no shade in the area, select 0.
 If there is insufficient pedestrian scale shade (less than 25% of the area is shaded),
select 1.
 If there is minimum sufficient pedestrian scale shade (25% to75% of the area is
shaded), select 2.
 If there is full sufficient pedestrian scale shade (more than 75% of the area is shaded),
select 3.
4.4.16. Crossing facility type
Crossing facility type refers to the type of crossing facilities provided to assist in the safe
crossing of roads and paths by pedestrians. The type of pedestrian crossings provided should
be appropriate for the size of the road and the amount of traffic at peak or off-peak periods
(Western Australia Department of Transport, 2011).
204
School crossing Unmarked crosswalks Marked crosswalks
Figure 20 Crossing facility type
To identify choices on the audit form, take into consideration whether kerb ramps have the
following specifications:
 Sufficient for the type of road, size of road, volume of traffic, pedestrian volumes and
type of pedestrians.
 Note whether ramps are clearly visible to pedestrians, motorists, and other road users.
 Kerb ramps should be provided with the required crossing aids, including pedestrian
refuge islands, curb extensions, pedestrian crossing warning signs, pedestrian signals,
pedestrian countdown signals and ‘yield to pedestrian’ signs.
Then estimate the crossing facility type conditions for the audited footpath segment based on
 If the crossing facilities are unmarked crosswalks without crossing aids, select 0 as
poor.
 If the crossing facilities type are unmarked crosswalks with insufficient crossing aids,
select 1 as moderate.
 If the crossing facilities are marked crosswalks with insufficient crossing aids, select 2
as good.
 If the crossing facilities are marked crosswalks with sufficient crossing aids, select 3
as excellent.
4.4.17. Active frontage of buildings
As cities and towns evolve, more demands are made on the walking environment. In order to
achieve high amenity, community safety and visual interest, the relationship between the
walking environment and adjacent buildings becomes critical. To maximise these qualities, it
is important that buildings have an ‘active frontage’ to the footpath environment. An active
frontage adds interest and vitality to the walking environment by creating active building
edges on the street. Buildings having active frontages increase the walkability of an area,
205
relying in part on their attractiveness. The sides of buildings can be made more attractive by
adopting public art that captures and reinforces the unique character of a place (Moreland City
Council, 2010).
 Are most buildings situated close to the street with entrances accessible from the
footpath (without having to cross a parking lot)?
 Do buildings have multiple windows that allow people to see into and out of the
buildings?
 Do buildings with several windows close to the street create a safer and more
interesting environment for pedestrians?
To identify choices on the audit form, take into consideration whether the active frontages of
buildings have the following specifications:
 Several doors and windows with few blank walls.
 The frontage of buildings gives a vertical rhythm to the street scene.
 Facades with projections such as bays and porches provide a welcoming feeling.
 Lively internal uses are visible from the outside, or spill onto the street.
Then estimate the active frontage of building conditions for the audited footpath segment
 If the frontage of buildings along the audited segment is non active, select 0.
 If the frontage of buildings along the audited segment is poorly active, select 1.
 If the frontage of buildings along the audited segment is fairly active, select 2.
 If the frontage of buildings along the audited segment is excellently active, select 3.
4.4.18. Facilities for the disabled (directional TGSI)
Tactile Ground Surface Indicators (TGSI) are physical and tactile facilities provided for
people with disabilities, such as those who are blind or have low vision, in the absence of
enough existing cues to help them move independently and safely through the pedestrian
environment. There are two types of TGSI, including warning TGSI that are used to identify
hazards and targets, and directional TGSI which are used to identify a safe path. Generally,
these indicators are used to guide pedestrians in changing directions (to access a bus stop or
building), or to warn of an upcoming pedestrian hazard (a road or steps). However, the use of
TGSI is only preferred where necessary, being installed in a manner that minimises
difficulties for other pedestrians, particularly those using wheelchairs or other walking aids.
Here, only directional TGSI will be checked as a factor related to the facilities for the disabled
206
that affect PLOS, because other facilities for the disabled factors and TGSI types are checked
within other factors (Western Australia Department of Transport et al., 2011).
Warning TGSI Directional TGSI

Figure 21 Disabled facilities.
To identify choices on the audit form, directional TGSI should have the following
specifications:
 Should only be used where necessary. Check whether they are needed on the audited
segment; then check the next specifications and record the score for this factor.
Otherwise, leave this factor and record it as 4.
 Should be colour contrasted against the surrounding pavement.
 Should be used across a path of travel to identify the point at which a pedestrian needs
to turn to access a particular target such as a mid-block pedestrian crossing, bus stop
or point of entry to a significant public facility. In these situations, TGSI should:
1. Be set perpendicular to the flow of pedestrian traffic.
2. Be 0.8m deep.
3. Extend across the entire width of the walkway.
4. Be aligned to the left hand side of the adjoining pathway.
 Should be used as a dedicated guidance strip to identify a safe path of travel through
an otherwise complicated site, with few or no alternative cues, from point A to point
B. In these situations, TGSI should:
1. Be 0.3m to 0.4m wide.
2. Be laid along the midline of the safe path of travel.
3. Allow for minimum clearance of at least 0.6m on both sides.
Then the directional TGSI conditions for the audited footpath segment should be estimated
based on following criteria:
 If there are no TGSI provided in the segment, select 0.
 If there are insufficient TGSI which do not meet all required specifications, select 1.
207
 If there are sufficient TGSI but they do not meet all required specifications, select 2.
 If there are sufficient TGSI which meet all required specifications or if they are not
needed, select 3.
4.4.19. Footpath maintenance
Footpath maintenance refers to the overall condition of the footpath and how all footpath
facilities should be regularly maintained. A well maintained footpath is free from cracking,
buckling, holes, damaged pavement, weeds, overgrown vegetation, protruding tree roots,
abrupt changes in surface level, drainage problems, uplifts or lips in the pavement surface.
The existence one of these issues can create several problems for pedestrians, such as tripping
hazards, and can significantly affect accessibility for pedestrians using wheelchairs, canes, the
elderly, or pedestrians with limited visibility (NZ Transport Agency, 2009).
To identify choices on the audit form, the following conditions should be understood (Pikora
et al., 2002):
 Poor maintenance is with the existence of a lot of bumps, cracks, holes and weeds
growing in the surface or between the cracks of footpaths, abrupt changes in surface
level, drainage problems, uplifts or lips in the pavement surface.
 Moderate maintenance is with the existence of some bumps, cracks, holes and weeds
 Good maintenance is with the existence of very few bumps, cracks, holes and weeds
 Excellent maintenance is with the absence of bumps, cracks, holes and weeds
Figure 22 Poorly maintained paths

Source: (U.S. Department of Transportation, 2007).
208
Then the maintenance condition for the audited footpath segment should be estimated based
on following criteria:
 If the segment condition seems less than 25% well maintained (a lot of bumps, cracks,
holes and weeds), select 0 as poor maintenance.
 If the segment condition seems between 25-50% well maintained (some bumps,
cracks, holes and weeds), select 1 as moderate maintenance.
 If the segment condition seems between 75-90% well maintained (very few bumps,
cracks, holes and weeds), select 2 as well maintained.
 If the segment condition seems more than 90% well maintained (no bumps, cracks,
holes and weeds), select 3 as excellent maintenance.
4.4.20. Footpath cleanliness
Cleanliness contributes to an increase in a footpath’s attractiveness and pleasantness and so
promotes the walkability of the whole area. A pedestrian route is attractive if the path is free
of litter, dumped rubbish, discarded items, overgrown vegetation, fallen leaves, broken glass,
trash, and other unsightly and hazardous debris (Clifton et al., 2007) (Pikora et al., 2002).
To identify choices on the audit form, estimate the cleanness condition of the footpath
 If there is a large amount (a very noticeable amount) of litter, rubbish, on the segment,
select 0 as poor.
 If there is small amount of litter, rubbish, etc., on the segment, select 1 as fair.
 If there is very little litter, rubbish, etc., on the segment, select 2 as good.
 If there is no little litter, rubbish, etc., on the segment, select 3 as excellent.
4.4.21. Footpath surveillance
Surveillance (or ‘eyes on the street’) is a way of enhancing personal safety or the perception
of personal safety, and is an opportunity for people to be seen by other people or devices (e.g.,
surveillance cameras) (Pikora et al., 2002). Pedestrians should feel safe in public spaces such
as footpath areas where a lack of safety, or even a feeling of unease, will dramatically
decrease the attractiveness and hence the utilisation of an area. Having people around and
having buildings in proximity to the path increases the ‘passive surveillance’ of the area.
Passive surveillance is one of the best ways of helping people feel safe, both during the day
and at night, when pedestrians feel safer if other people are walking nearby (Western
Australia Department of Transport, 2011). The rater should check the personal safety of the
209
area both during daytime and night-time, as the characteristics of the area might change
significantly after dark.
making:
 Are many pedestrians walking along the footpath?
 Do people feel safe walking on this route section?
 Is the path visible? Is there an opportunity for surveillance from adjacent land uses and
buildings?
 Are there enough people around to make people feel safe?
 Are there enough police patrols or security cameras along the footpath?
Then estimate the surveillance conditions for the audited footpath segment based on the
following criteria:
 If there are no surveillance means along the audited area, select 0.
 If there are few surveillance means along the audited area, select 1.
 If there are sufficient surveillance means along the audited area, select 2.
 If there are fully sufficient surveillance means along the audited area, select 3.
4.4.22. Garbage facilities

The attractiveness of an area in which to walk also depends on the existence of rubbish bins
as part of the street furniture. These must be carefully placed to be unobtrusive yet effective,
and should be easily accessible for most pedestrians and trash collection. Good walking routes
have an adequate supply of rubbish bins (Western Australia Department of Transport et al.,
2011). The raters should check to see whether these facilities are provided along the route
section and whether they are well maintained.
To identify choices on the audit form, consideration should be given as to whether rubbish
bins have the following specifications:
 Are they in areas where litter may be generated, such as bus stops, transport
interchanges and fast food outlets?
 Are they provided along the route segment?
 Are they in good condition, i.e., are they well maintained (no damage or graffiti)?
210
 Rubbish bins should not intrude into the clear continuous path of travel. Ideally they
should be placed in the street furniture zone within the footpath cross-section so that
the pedestrian through route is kept free (preferably 1.5 m or wider) from obstruction.
 Bins should be approximately 0.4m-1m high.
 Bins should be placed within four metres of the edge of the nearest traffic lane.
 Bins should be accessible to all users.
 Bins should be colour contrasted from background surfaces to ensure that they are
visible to the vision impaired.
Then estimate the rubbish bin conditions for the audited footpath segment based on the
following criteria:
 If there are no rubbish bins in the audited segment, select 0.
 If there are bins somewhere but they do not meet all required specifications, select 1.
 If there is sufficient frequency of bins but they not meet all required specifications,
select 2.
 If there are sufficient rubbish bins meeting all required specifications, select 3.
4.4.23. Pedestrian signage
Pedestrian signage provides provide general information to guide pedestrians, and also to
warn motorists of approaching pedestrian activity and crossings. Two kinds of signage should
be provided on a path, including both guidance and warning signs. The guidance signage
provides general information about a path, such as the type of path (shared use path or
separated path) present on the route section, or important information about what is permitted
or prohibited in this facility. Warning signs and pavement markings warn motorists about the
presence of pedestrians in the area to increase safety for pedestrians at pedestrian activity
locations and crossings. Generally, pedestrian signage should also be clearly visible and
effective for all likely conditions, e.g., day, night, rain, fog, rising or setting sun, oncoming
headlights, light coloured pavement surface, and poor lighting (Western Australia Department
of Transport et al., 2011).
211
Figure 23 Pedestrian signage
 Is guidance signage sufficient on the path?
 Is important information clearly written on signs? Are signs visible to pedestrians?
 Are there sufficient warning signs on the path?
 Are marked and signed pedestrian crossings clearly visible to pedestrians?
 Is the type of pedestrian infrastructure clearly marked?
 Are pedestrian routes and crossings clearly visible to motorists, with pedestrian
warning signs and pavement markings?
 Is the condition of signage and markings adequate (for example, well painted, non-
slippery material, visible during day and night, not damaged or worn)?
During the audit process and before making any decisions, consider whether the guidance
signage meets the following specifications:
 The signs do not intrude into the clear continuous path of travel (ideal locations).
 The signs are appropriately oriented.
 The signs are visible, legible and understandable.
 The signs are placed above a pathway at a height of 2.5m.
 The signs are regularly maintained and in good condition.
Then estimate the guidance signage conditions for the footpath segment based on the
following criteria:
 If there is no signage in the area, select 0.
 If there are some signs but they are insufficient for the purpose and do not meet all
required specifications, select 1.
 If there is sufficient frequency of signs but they not meet all required specifications,
select 2.
 If there is fullly sufficient frequency of signs meeting all required specifications, select
3.
212
4.4.24. Footpath lighting
Most footpaths require street lighting to improve pedestrian visibility and create a sense of
security and safety. Lighting for lateral movements of pedestrians is critical for safety.
Lighting on roads helps motorists to see pedestrians at marked and unmarked crossing points,
where pedestrians are more likely to cross mid-block in commercial areas (Nabors et al.,
2007). Lighting is related to the personal safety of pedestrians, enabling them to perceive
hazards such as unusual or uneven surfaces or obstacles. Footpath lighting should be bright
enough to enable pedestrians to recognise potential threats from other people in time to take
appropriate action. Lighting helps to improve the visibility and overall ambience of pedestrian
pathways, ensuring the personal security of pedestrians, and making an area more attractive
for walking. The issues of personal safety and security are particularly important for elderly
people and people with impaired vision who may be more vulnerable to trip hazards or feel
insecure or uncomfortable in poorly lit environments (Austroads, 2017). Designers should
consider all aspects of the design that may influence its effectiveness, such as the presence of
overhanging trees and low-profile hedges which may create significant shadowing, and
which, when combined with adjacent headlights (from the roadway), could make the
silhouettes of pedestrians extremely difficult to see. It is preferred that the raters visit audited
sites at night to evaluate the adequacy of the lighting along the roadway with respect to
pedestrian volumes.
Figure 24 Lighting
Source: (City of Sydney Council, 2013).
In identifying choices on the audit form and making decisions, keep in mind the lighting
requirements for pedestrian pathways outlined in MRWA Lighting Design Guideline for
213
Roadway and Public Spaces (based on the AS/NZS1158.3.1 – 2005 requirements) (Western
Australia Department of Transport et al., 2011). These requirements include:
 At locations of potential conflict and high pedestrian usage, such as intersections of
paths and railway stations, the point horizontal and vertical illumination is a minimum
of 20 lux (although this is not always practical to achieve).
 Ideally the location of lamp posts should be located in the street furniture zone within
the footpath cross-section, in order to keep the pedestrian through route free from
obstruction. This width can vary between 0.9m - 1.2m. When street furniture is
located on the footpath then a clear path of travel of at least 1.2m (preferably 1.5m or
wider) must be maintained along the building or property line. This will allow a
navigational path for those with low vision, and allow a wheelchair to pass along the
path.
Following these guidelines, lighting conditions on the footpath segment based on the
following criteria should be estimated:
 If there is no artificial lighting in the area, select 0.
 If there is insufficient pedestrian scale lighting, select 1.
 If there is sufficient pedestrian scale lighting, select 2.
 If there is fully sufficient road oriented lighting, select3.
4.4.25. Public transportation bus stops
Public transport bus stops are distributed throughout walking environments, as are footpaths
to provide comfortable and safe accessibility to all kinds of public transport services. Well-
designed bus stops, including their interfaces on footpaths, are essential to improve the level
of service along the pedestrian network system. Footpath width needs to be considered
carefully at public transport bus stops where a large number of pedestrians are expected to
board or exit, such as at railway stations. Also, the footpath surface of bus stops should be
designed to serve all users, including pedestrians who are vision impaired or disabled, by
identifying the public transport access areas using environmental cues or tactile paving (NZ
Transport Agency, 2009). Both directional and warning tactile ground surface indicators
(TGSI) should be installed at all new bus stops to guide pedestrians to and from the bus entry
point (Western Australia Department of Transport et al., 2011). Standard shelters should be
also placed at bus stops in a widened street furniture zone where there is a clear need. Shelters
should not affect the standard footpath width (through route zone) if installed in the frontage
214
zone creating kerb extensions. Alternatively, bus shelters should be located in kerb extensions
on footpaths (NZ Transport Agency, 2009).
Figure 25 Public transport bus stop layout

 Are public transport bus stops needed along the audited segment?
 Are sufficient public transport bus stops provided along the audited segment?
 Are public transport bus stops located at logical locations along the audited segment?
To identify choices on the audit form, the following specifications of the public transport bus
stops should be taken into consideration:
 Bus stops should have both directional and warning tactile ground surface indicators
(TGSI).
 Bus stops should have standard shelters if there is a clear need.
 Bus stops should maintain a continuous path with a desired width of 1.54m (absolute
minimum width required is 1.2m) between the bus stop and the kerb.
 Bus stops should be placed at footpath kerb extensions zones if there is insufficient
width.
Then, estimate the public transport bus stop conditions for the audited footpath segment based
 If there are no bus stops provided in the segment, select 0.
 If there are insufficient bus stops which do not meet all required specifications, select
1.
 If there are sufficient bus stops but they do not meet all required specifications, select
2.
 If there are sufficient bus stops and they meet all required specifications, select 3.
215
4.4.26. Wayfinding facilities
Wayfinding refers to the type and sufficiency of navigational systems provided on a footpath
to help pedestrians to determine where they are and where they need to go to reach a
destination. Pedestrians use different landmarks to find their way, such as landforms,
watercourses, views, surrounding land uses and street layouts to orientate themselves as they
walk around an area (Transport, 2005). Wayfinding facilities provide information to assist
pedestrians in identifying and locating a particular amenity or facility, or to provide
information on the use of a facility. In busy activity centres, maps identifying key destinations
can assist with improving the walkability of an area, particularly where there are large
numbers of tourists. Clear visual and physical wayfinding facilities make an area easy to read,
as well as increasing the legibility and walkability of area (Western Australia Department of
Transport et al., 2011).
Ideally the wayfinding facilities should be placed in the street furniture zone within the
footpath cross-section in order to keep the pedestrian through route free from obstruction, and
the wayfinding facilities should face the oncoming pedestrian flow. To accommodate
children, people in wheelchairs and people standing, the most appropriate height is in the
range of 1.2m to 1.6m, however where this space is unavailable, the absolute minimum height
is 1m (Western Australia Department of Transport et al., 2011).
Figure 26 Illustrations of the city’s wayfinding sign

Source: (City of Sydney Council, 2013).
Answers to following questions will help in decision making:
 Is signage provided to guide and direct pedestrians to the key destinations in the area?
 Is information, such as street names, clearly written on signage and visible to
pedestrians?
216
To identify choices on the audit form, the following specifications of wayfinding facilities
should be considered:
 Wayfinding facilities should not intrude into the clear continuous path of travel.
Ideally, these facilities should be placed in the street furniture zone within the footpath
cross-section so as to keep the pedestrian through route free from obstruction.
 Wayfinding facilities should be appropriately oriented (should face the oncoming
pedestrian flow).
 Wayfinding facilities should be visible, legible and understandable (have a clear visual
and physical presence).
 Wayfinding facilities should use clear font and simple language (use both plain
English and symbols that are easy to read).
 Wayfinding facilities should be usable for all users, and be of appropriate height - in
the range of 1.2m to 1.6m.
Then, the wayfinding facility conditions for the audited footpath segment based on the
following criteria should be estimated:
 If there is no wayfinding facility in the area, select 0.
 If there is a facility somewhere but it does not meet all required specifications, select
1.
 If there is sufficient frequency of wayfinding facilities, but they do not meet all
required specifications, select 2.
 If there is full sufficient frequency of wayfinding facilities and they meet all required
specifications, select 3.
4.4.27. Pedestrian evasive manoeuvres
A pedestrian’s evasive movement refers to the ability of a pedestrian to cross another
pedestrian’s traffic stream, to walk in the reverse direction of a major pedestrian flow, to
manoeuvre generally without conflict and changes in walking speed. The pedestrian flow may
be unidirectional, bidirectional, or multi directional. It is human nature for pedestrians try to
keep within their own personal space when faced with a situation where other pedestrians are
expected to occupy a part of their personal space, making evasive movements to protect these
spaces. These evasive movements include a change in walking speed and/or walking path.
Various evasive movements were determined in the real world, particularly when considering
pedestrians walking on footpaths. These kinds of evasive manoeuvres include, for example,
avoiding pedestrians coming from the opposite direction, passing pedestrians in the same
217
direction, and walking abreast (Kim, Choi, Kim & Tay, 2014). The number of evasive
movements can be used to capture pedestrian level of service where the higher the number of
evasive movements undertaken, the lower the pedestrian level of service on a footpath.
making:
 Are many evasive manoeuvres undertaken by pedestrians on the footpath?
 Are there a massive number of evasive movements on the audited segment?
 Are there many attempts by pedestrians to avoid other pedestrians coming from the
opposite direction, passing other pedestrians in the same direction, or walking abreast?
Then, pedestrians’ evasive manoeuvres conditions for the audited footpath segment based on
the following criteria should be estimated:
 If there are many evasive manoeuvres undertaken by pedestrians in the segment, select
0 as massive evasive manoeuvres.
 If there are a moderate amount of evasive manoeuvres undertaken by pedestrians on
the segment, select 1 as moderate evasive manoeuvres.
 If there are few evasive manoeuvres undertaken by pedestrians in the segment, select 2
as normal evasive manoeuvres.
 If there are no evasive manoeuvres undertaken by pedestrians in the segment, select 3
as no evasive manoeuvres.
4.4.28. Pedestrian conflict points
Pedestrian conflict points are the points that could arise as a result of users’ behaviour and/or
failure to adequately plan, design, build and maintain a facility to fully account for the
diversity of users. This factor is directly related to pedestrian safety. It appears that it is
difficult to provide one facility that offers the appropriate level of service with no potential
conflict points in users’ movements. However, consideration of the potential for conflict
among users on existing paths, or when new facilities are being planned and designed, can
address these issues. The pedestrian environment can become increasingly unsafe and
uncomfortable, and a majority of people can experience conflict or closeness with other
pedestrians and bi-directional movement, if conflict points arise.
 Are there frequent conflict points and, in retail areas, do people begin to avoid these
areas?
218
 Do the majority of people experience a conflict or closeness with other pedestrians
and does bi-directional movement become difficult?
 Estimate the potential points that are considered as pedestrian to pedestrian conflict
points or pedestrian to bicycle conflict points on the audited footpath segment.
 Focus on pedestrians in retail areas and the areas of closeness with other pedestrians
and bi-directional movement during observations of pedestrian conflict points.
Then, estimate the pedestrian conflict points’ conditions on the footpath segment, based on
 If there are many pedestrian conflict points, select 0 as poor.
 If there are a moderate number of pedestrian conflict points, select 1 as moderate.
 If there are few pedestrian conflict points, select 2 as reasonable.
 If there are no pedestrian conflict points, select 3 as perfect.
4.4.29. Pedestrian speed
Pedestrian speed is the walking speed at which pedestrians prefer to walk, which can be
improved by providing adequate footpath width for normal walking. However, normal
pedestrian speeds can frequently cause conflict. The environment becomes very safe and
comfortable when pedestrians are provided with a wide space so that they can walk at the
speed and along the route they choose. Pedestrian speeds are restricted and reduced with
insufficient footpath width, and many difficulties appear in bypassing slower pedestrians or
moving in reverse flows. All these factors, with others, become more complicated with
increasing pedestrian volumes on a footpath. The average walking speed for a fit adult is 1.5
metres per second (m/s). Walking speeds for elderly pedestrians or pedestrians with mobility
impairments can be 1.0m/s to 1.2m/s. When assessing crossing sites, a walking speed of 1.2
m/s is generally adopted, covering 85% of pedestrian walking speeds (Western Australia
Department of Transport et al., 2011).
Pedestrians’ speed can be assessed in two ways. The first is an objective method by
measuring the pedestrian speed in the field and comparing it to the standard speed of 1.2m/s.
The second way is a subjective method, by observing the comfortable walking speed on the
audited segment based on personal space provided for pedestrians to walk at the comfortable
walking speed chosen. Here, we adopt the second method, as it is cheap and efficient.
To identify choices on the audit form, answer the following questions to make decisions:
 Is there plenty of space for pedestrians to walk at the speed chosen?
219
 Does the current path provide enough space for the walking speed taken?
 Do pedestrians have very little personal space, and is their speed very restricted? Do
they experience extreme difficulty in passing slower pedestrians?
Then, estimate the pedestrian speed conditions for the audited footpath segment based on the
following criteria:
 If the path provides very little personal space for pedestrians to walk at the speed
chosen, select 0 as very restricted walking speed.
 If the path provides reasonable personal space for pedestrians to walk at the speed
chosen, select 1 as restricted walking speed.
 If the path provides sufficient personal space for pedestrians to walk at the speed
chosen, select 2 as normal walking speed.
 If the path provides plenty of personal space for pedestrians to walk at the speed
chosen, select 3 as free walking speed.
4.4.30. Pedestrian volume
Pedestrian volume refers to the estimate of the number of pedestrians using the footpath per a
specific period. Pedestrian volume also plays a key role in designing the pedestrian
environment and directly affects walking safety, comfort, and mobility, and then, therefore,
PLOS. We adopt the pedestrian comfort method established by Transport for London to count
pedestrian volume, providing appropriate pedestrian comfort levels as presented below
(Transport for London, 2010).
To identify choices on the audit form, conduct the pedestrian volume counts using the
following steps:
 Stand in a location that does not disrupt normal activity on the audited footpaths.
 Draw an imaginary line perpendicular to the footpaths that are targeted for a
pedestrian volume count.
 Count the number of pedestrians who cross the imaginary line for a specific period (15
minutes at each segment) using a manual method. It is advisable to use tally counters
to record this information, particularly in busy sites.
 Pedestrian crowding is measured in pedestrians per metre of clear footway width per
minute (ppmm) and is calculated using the following formula:
Pedestrians per 15 minutes ÷ 15 ÷ clear footpath width in metres.
220
Then, estimate the pedestrian volume conditions for the audited footpath segment based on
 If the pedestrian crowding is > 26ppmm, select 0 as very crowded path.
 If the pedestrian crowding is > 17ppmm≤ 26ppmm, select 1 as moderately crowded
path.
 If the pedestrian crowding is > 8ppmm ≤ 17ppmm, select 2as an uncrowded path.
 If the pedestrian crowding is ≤ 8ppmm, select 3 as a very uncrowded path.
4.4.31. Traffic-pedestrian conflict points
A conflict between traffic and pedestrians is defined as an event involving two or more road
users, in which the action of one user causes the other user to make an evasive manoeuvre to
avoid a collision (Wu, Radwan & Abou-Senna, 2018). Recently, the total number of conflicts
has been used as a measurement of safety issues, taking the severity of the conflicts into
consideration as another element of the safety measurement. The main idea behind identifying
potential traffic-pedestrian conflict points on footpaths is that it is possible to reduce the
incidence of accidents by reducing the less severe, yet more frequent, traffic conflicts to
increase pedestrian’s safety (U.S. Department of Transportation, 2007). Traffic-pedestrian
conflict points can be calculated by counting the number of potential vehicle conflict points
along the path measured per path kilometre (Gallin, 2001). The vehicle conflict points
include, for example, signalized or unsignalized intersections, driveways, left or right-turning
traffic points and mid-block crossings (U.S. Department of Transportation, 2007). Pedestrian
safety decreases with increased traffic-pedestrian conflict points’ frequency and that
negatively affects pedestrian level of service.
 Are there many traffic-pedestrian conflict points along the audited segment?
 How many traffic-pedestrian conflict points are there?
 Do you feel safe when you walk in these conflict points?
 Identify the existence of points that are considered as traffic-pedestrian conflict points
in the audited footpath segment. This includes points in the following areas:
unsignalized intersections, driveways, left or right-turning traffic points and random
mid-block crossings.
221
 Add the number of points that are considered to be traffic-pedestrian conflict points in
audited footpath segments and save the score to calculate the total score of traffic-
pedestrian conflict points for the audited footpath.
Then, estimate the traffic-pedestrian conflict points’ conditions for the footpath segment
 If the total traffic-pedestrian conflict points are more than 8 per segment, select 0 as
poor.
 If the total traffic-pedestrian conflict points are 4-8 per segment, select 1 as moderate.
 If the total traffic-pedestrian conflict points are 1-3 per segment, select 2 as
reasonable.
 If the total traffic-pedestrian conflict points = 0 (no potential conflict points), select 3
as perfect.
4.4.32. Vehicle speed
Vehicle speed refers to the preferred average speed of motor vehicle traffic (km/hr). Vehicle
speed is related to issues about the interaction between pedestrians and traffic, and affects
pedestrian safety and comfort (U.S. Department of Transportation, 2007). Even in the
presence of well-designed and well-maintained pedestrian footpaths, the speed of traffic may
compromise pedestrian safety. The free vehicle speed is one of the main reasons why people
do not feel safe or do not find a route attractive, particularly on roads where there is greater
potential for conflict between cars and unprotected road users (pedestrians and cyclists).
Therefore, there is a need to reduce the speed of traffic through busy activity centres where
there are large numbers of pedestrians crossing the road, adopting the Towards Zero Road
Safety Strategy to reduce the traffic speed to a 30 km/h safe speed threshold (Western
Australia Department of Transport et al., 2011).
Figure 27 Desired traffic speed

 Check the current allowed traffic speed in the audited area.
222
 Consider this speed as the speed of the traffic flow, regardless of offending drivers.
 Compare this speed with the criteria determined below and identify selection.
Then, estimate the vehicle speed conditions for the footpath segment based on the following
criteria:
 If the traffic speed is more than 60 km/h, select 0 as poor conditions.
 If the traffic speed is 41 km/h - 60 km/h, select 1 as moderate conditions.
 If the traffic speed is 31 km/h- 40 km/h, select 2 as good conditions.
 If the traffic speed is 30 km/h or lesser, select 3 as perfect conditions.
4.4.33. Vehicle volume
Vehicle volume refers to the estimate of the average number of motor vehicles during a
specific period. The presence of a massive number of vehicles leads to severe traffic
congestion, which can often trap pedestrians causing a delay in pedestrian flow. High traffic
volumes can interfere with pedestrian movements, especially at signalized and unsignalized
intersections and mid-block crossings, affecting pedestrian safety during peak or off-peak
periods (U.S. Department of Transportation, 2007).
To identify choices on the audit form, estimate the traffic-pedestrian conflict points’
conditions for the footpath segment based on the following criteria:
 If there is a high traffic volume, select 0 as poor conditions.
 If there is moderate traffic volume, select 1 as moderate conditions.
 If there is normal traffic volume, select 2 as good conditions.
 If there is low traffic volume, select 3 as excellent conditions.
5. Reliability test
In this context, the subjective assessments of raters may play a role in the assessment process,
because of the ways in which they measure each item in the field based on eligibility criteria.
For this this reason, in order to develop a robust version of the PLOSAT, it is essential that a
reliability test be created. The reliability test reflects the stability in response between the
multiple measurements of the same concept. Inter-rater reliability of an audit tool refers to the
variation or convergence of answers between two or more raters who measure the same group
of subjects (Pikora et al., 2002). The reliability can be assessed using percent agreement
measurement when the unit of measurement is on a categorical scale (Clifton et al., 2007;
Griew et al., 2013; Troped et al., 2006). Percent agreement is a simple measurement of
agreement by calculating the proportion of times that raters agree on the same scores.
223
However, this measurement does not take into account the level of agreement that could occur
by chance. Consequently, the Fleiss’ kappa statistic is used to overcome this limitation by
presenting a chance-corrected measure of agreement between raters (Griew et al., 2013). A
kappa coefficient of 1.0 indicates a perfect agreement, and a kappa coefficient of 0 refers to
an agreement corresponding to that expected by chance, while a kappa coefficient less than 0
indicates an agreement less than that expected by chance (Clifton et al., 2007). A training
session with six raters was held to provide a good understanding of each of the audit items.
Raters who had suitable experience in the transport planning field were recruited from the
transport PhD cohort at the Cities Research Institute, Griffith University. First, the guideline
manual was distributed to participants one week before holding the training session, and
participants were asked to read it carefully. The session comprised of three main parts: i) an
overview of the audit; ii) a detailed description of each audit element; and iii) targeted
segments.
A comprehensive review of the audit tool item by item was conducted. Then, each item and
selected option was reviewed in depth, with photographic examples. Raters were given a
chance to ask questions and discuss any problems or concerns regarding the audit process
during the training session. Then, the location in the field of the targeted segment for the audit
was determined and the maps identifying the required segments were provided to each rater.
The training session and protocol materials provide detailed background information so that
once trained, raters could manage and conduct the audit reliably.
5.1. Data collection

A total of 2 km of footpath was audited during August 2019 consisting of six footpath
segments with near-equal lengths. All factors were audited and measured in the field using the
PLOS audit tool.
All raters were asked to conduct an audit process during morning peak hours, as this is a
critical time period in which to calculate vital PLOS values. The required equipment and
materials, such as a fieldwork forms, internet data, phones, power banks for phone chargers,
user manuals, and detailed maps were provided to each rater, and this included all necessary
information for field measurement. Maps of targeted footpath segments were prepared and
provided to the raters with all required information such as street names, segments ID, auditor
ID, and so on. The maps helped to ensure that the same segments were audited by different
raters. At the end of data collection, the raters manually transferred the obtained results from
224
the fieldwork form to a weighted assessment form to calculate the PLOS value. Finally, the
completed audit form and maps were collected from the raters for analysis purposes.
5.2. Data analysis
Tests of reliability explored: 1) the items involved in the use of the tool; 2) the overall inter-
rater reliability of the audit; and 3) the overall inter-rater reliability of the PLOS grades. The
reliability of the items in the instrument was assessed using Fleiss’ kappa statistics for each
item and the percentage of agreement between the raters.
Kappa values were interpreted as follows: 0.01-0.20 (slight agreement); 0.21-0.40 (fair
agreement); 0.41-0.60 (moderate agreement); 0.61-0.80 (substantial agreement); 0.81-0.99
(almost perfect agreement) (Griew et al., 2013). The percentage of agreement indicated was
more appropriate for data which exhibited little variation, and this was also a good way to
measure reliability (Boarnet, Day, Alfonzo, Forsyth & Oakes, 2006). Each audited footpath
was rated from the best ‘A’ to the worst ‘F’, for overall PLOS and for each of the five PLOS
categories based on information gathered in the field from the associated factors.
Most factors demonstrated high agreement and acceptable kappa scores, as shown in Tables 5
and 6. The reliability results of only two (pedestrian-traffic separation within geometric
characteristics category, and pedestrian evasive manoeuvres within the pedestrian traffic
category) were low for both percentage agreement (<60%) and the kappa coefficient (<0.4).
The factors requiring subjective assessment demonstrated less reliability than those
demanding objective assessment. Only a small number of factors in the audit consistently
garnered kappa scores below 0.40. As expected, these were mainly more subjective factors,
such as pedestrian-traffic separation, type of crossing facility, pedestrian signs, wayfinding
guidance and pedestrian evasive manoeuvres. On the other hand, most factors in the audit
such as footpath width, buffer width, connectivity and permanent obstruction received very
high kappa scores with perfect agreement. Assessment of the lighting factor was difficult
because the audit was administered during the day, and this is a limitation of the study.
Generally, the PLOSAT reliability recorded a moderate to substantial inter-rater reliability
over the majority of the factors affecting PLOS with an average of 82.5% (k = 0.5), and for
the categorical PLOS results with an average of 83% of agreement, while, the reliability of
overall PLOS results indicated a high reliability with an average of 97% of agreement.
Table 5 Kappa statistics and percentage of agreement for each factor affecting PLOS by 6 raters across
6 segments.
225
% Percent
Category Factor Kappa score
agreement
Footpath width N/A 100
Geometric characteristics
Crossing opportunities -0.13 78
Quality of footpath surface 0.5 75
Footpath slope & grade -0.06 89
Buffer area and width N/A 100
Pedestrian-traffic separation 0.27 57
Ramp 0.22 88
Clearance over footpath N/A 100
Footpath connectivity 0.87 94

interruption
Obstruction
Footpath continuity N/A 100

s and
Easy of movement N/A 100

Permanent obstructions 1 100
Awnings 0.63 80
Trees and plants 0.63 76
Shade 0.60 77
Type of crossing facility -0.07 65
Aesthetics and amenities
Activation of side facades 0.57 74

Facilities for the disabled 0.46 73
Footpath maintenance 0.77 90
Footpath cleanliness -0.09 83
Footpath surveillance 0.64 80
Garbage facilities -0.03 94
Pedestrian signs 0.44 65
Lighting 0.40 72
Public transportation facilities -0.09 83
Wayfinding guidance 0.03 72
Pedestrian evasive manoeuvres 0.22 58

Pedestrian
Pedestrian conflict points 0.40 76

traffic
Pedestrian speed 0.30 68

Pedestrian volume N/A 100
Vehicle conflict points 0.5 82

Adjacen
t traffic
Vehicle speed N/A 100

Vehicle volume 0.62 74
Overall agreement 82.5
226
Table 6 percentage of agreement for PLOS results by 6 raters across 6 segments.
% Percent agreement
PLOS results
Among raters across the 6 segments Overall PLOS
For the whole categories 78% 100%
For the geometric characteristics category 69% 100%
For the obstructions and interruptions category 94% 100%
For the aesthetics and amenities category 91% 100%
For the pedestrian traffic category 78% 83%
For the adjacent traffic category 89% 100%
6. Discussion
In summary, PLOSAT is not perfect an audit tool, but it appears to be a robust and reliable
tool that focuses on the factors both practitioners and users agree are important. PLOSAT
focuses on the following areas: i) consideration of a wide variety and range of footpath
elements and contexts within Australia; ii) measurement of an overall pedestrian level of
service on footpaths in commercial areas; iii) integration with handheld technology; iv)
detailed training materials, supporting documentation and guidelines; and v) testing the
reliability of the tool.
While the reliability and useability of the tool were high overall, a limitation is that only six
segments in total were assessed by six raters. Increasing the number of segments and the
number of raters would allow more precise estimation of reliability.
References
Austroads. (2017). Guide to road design part 6A: paths for walking and cycling.
Boarnet, M. G., Day, K., Alfonzo, M., Forsyth, A., & Oakes, M. (2006). The Irvine–Minnesota
inventory to measure built environments: reliability tests. American Journal of Preventive
Medicine, 30(2), 153-159. e143.
Brownson, R. C., Hoehner, C. M., Day, M. K., Forsyth, A., & Sallis, J. F. (2009). Measuring the Built
Environment for Physical Activity. American Journal of Preventive Medicine, 36(4), S99-
S123.e112.
City of Sydney Council. (2013). Sydney Streets Design Code Street Design Coordination Part E.
Sydney: City of Sydney.
Clifton, K. J., Smith, A. D. L., & Rodriguez, D. (2007). The development and testing of an audit for
the pedestrian environment. Landscape and Urban Planning, 80(1), 95-110.
Croft, P., Elazar, N., & Levasseur, M. (2013). Guide information for pedestrian facilities.
Gallin, N. (2001). Quantifying pedestrian friendliness--guidelines for assessing pedestrian level of
service. Road & Transport Research, 10(1), 47.
Griew, P., Hillsdon, M., Foster, C., Coombes, E., Jones, A., & Wilkinson, P. (2013). Developing and
testing a street audit tool using Google Street View to measure environmental supportiveness
for physical activity. International Journal of Behavioral Nutrition and Physical Activity,
10(1), 103. doi:10.1186/1479-5868-10-103
Kang, L., Xiong, Y., & Mannering, F. L. (2013). Statistical analysis of pedestrian perceptions of
sidewalk level of service in the presence of bicycles. Transportation Research Part A: Policy
and Practice, 53, 10-21.
227
Kim, S., Choi, J., Kim, S., & Tay, R. (2014). Personal space, evasive movement and pedestrian level
of service. Journal of Advanced Transportation, 48(6), 673-684. doi:10.1002/atr.1223
Koohsari, M. J., Karakiewicz, J. A., & Kaczynski, A. T. (2013). Public Open Space and Walking: The
Role of Proximity, Perceptual Qualities of the Surrounding Built Environment, and Street
Configuration. Environment and Behavior, 45(6), 706-736. doi:10.1177/0013916512440876
Maghelal, P. K., & Capp, C. J. (2011). Walkability: a review of existing pedestrian indices. URISA
Journal, 23(2), 5-20.
Mehta, V. (2007). Lively Streets: Determining Environmental Characteristics to Support Social
Behavior. Journal of Planning Education and Research, 27(2), 165-187.
doi:10.1177/0739456x07307947
Moreland City Council. (2010). Brunswick atructure plan lygon street. Moreland City Council.
Moudon, A. V., & Lee, C. (2003). Walking and bicycling: an evaluation of environmental audit
instruments. American Journal of Health Promotion, 18(1), 21-37.
Nabors, D., Gibbs, M., Sandt, L., Rocchi, S., Wilson, E., & Lipinski, M. (2007). Pedestrian road
safety audit guidelines and prompt lists. Retrieved from
NZ Transport Agency. (2009). Pedestrian planning and design guide. New Zealand: NZ Transport
Agency
Pikora, T. J., Bull, F. C. L., Jamrozik, K., Knuiman, M., Giles-Corti, B., & Donovan, R. J. (2002).
Developing a reliable audit instrument to measure the physical environment for physical
activity. American Journal of Preventive Medicine, 23(3), 187-194.
Raad, N., & Burke, M. (2017). Pedestrian Level of Service Tools: problems of conception, factor
identification, measurement and usefulness. Paper presented at the 39th Australian Transport
Research Forum (ATRF), Auckland, New Zealand
Raad, N., & Burke, M. (2018). What Are the Most Important Factors for Pedestrian Level-of-Service
Estimation? A Systematic Review of the Literature. Transportation Research Record,
0361198118790623.
Saelens, B. E., & Handy, S. L. (2008). Built environment correlates of walking: a review. Medicine
and science in sports and exercise, 40(7 Suppl), S550.
Sisiopiku, V. P., & Akin, D. (2003). Pedestrian behaviors at and perceptions towards various pedestrian
facilities: an examination based on observation and survey data. Transportation Research Part
F: Traffic Psychology and Behaviour, 6(4), 249-274.
doi:http://dx.doi.org/10.1016/j.trf.2003.06.001
Transport for London. (2010). Pedestrian Comfort Level Guidance. London: Transport for London.
Transport, Q. (2005). Easy Steps: A toolkit for planning, designing and promoting safe walking.
Brisbane, Australia.
Troped, P. J., Cromley, E. K., Fragala, M. S., Melly, S. J., Hasbrouck, H. H., Gortmaker, S. L., &
Brownson, R. C. (2006). Development and reliability and validity testing of an audit tool for
trail/path characteristics: the Path Environment Audit Tool (PEAT). Journal of Physical
Activity & Health, 3, S158.
U.S. Department of Transportation. (2007). Pedestrian Road Safety Audits Guidelines and Promplists.
U.S. Department of Transportation: Federal Highway Administration.
Western Australia Department of Transport. (2011). WALKABILITY AUDIT TOOL. Western Australia:
Department of Transport Retrieved from www.transport.wa.gov.au/walking.
Western Australia Department of Transport, Planning, & Public Transport Authority. (2011). Planning
and designing for pedestrians: guidelines. In. Western Australia: Department of Transport.
Wu, J., Radwan, E., & Abou-Senna, H. (2018). Assessment of pedestrian-vehicle conflicts with
different potential risk factors at midblock crossings based on driving simulator experiment. Advances
in transportation studies, 44, 33-46.
228
8 Discussion
This chapter begins with a general summary of the thesis, including its objectives, methods
and outcomes. A brief overview of the key findings of the study is provided, followed by the
contributions this research makes, its limitations and recommendations for future research.

This study established a comprehensive methodology for identifying and measuring
pedestrian level of service (PLOS) in a systematic and reliable manner. A set of inter-related
aims were achieved in this study which include: a) identifying the various factors affecting
PLOS in different pedestrian facilities using systematic methods; b) selecting the appropriate
factors for the Australian context which mainly influence the PLOS on footpaths in
commercial areas; c) examining the importance level of a wide range of factors affecting
PLOS, and their relationship to each other; d) developing two PLOS models, the first based
on our new approach labeled ‘pedestrian disgruntlement’, and the second based on a
satisfaction approach; and e) developing a comprehensive PLOS model using an audit tool
method to measure the overall PLOS value.
8.2 Response to research questions

This study had four primary research questions:
RQ1: What were the previous approaches, methodologies and factors used to evaluate the
This study found that the growth in the development of different approaches to measure
PLOS since 2013 (i.e., Jensen, 2013; Kim et al., 2014; Asadi-Shekari et al., 2014; Lazou et
al., 2015; Daniel et al., 2016) is indicative of increased policy and planning interest in
provisions for walking. The systematic quantitative review of the past literature conducted in
this study (Raad & Burke, 2018) explains data collection and analysis methods being used,
and identifies a wide variety of factors which may be considered in order to adequately
capture pedestrians’ walking experiences. The systematic literature review shows that PLOS
studies initially adopted capacity based modelling approaches, following Fruin’s (Fruin,
1970) approach. But though this approach continues in the work of those we call the
geometricians, it has been challenged by the alternative approaches of the experientialists,
who include Khisty (1994), Sarkar (2003) and Hidayat et al. (2011). These two competing
camps fundamentally conceive PLOS in different ways: the former as a problem of facility
229
capacity, flow and geometric design; the latter as about the more subjective and experiential
factors of the built environment which produce pedestrian amenity. Many researchers,
including Kang et al. (2013), Zhao et al. (2014), and Hasan et al. (2015), have sought to
bridge these two concepts, merging aspects of each in their work. But the litany of different
approaches and methods shows that the field remains in significant dispute as to how
authorities should proceed. There are increasing studies but increasing disagreement. Even
within each camp, the sets of factors proposed differs widely across authors, showing little
convergence.
The advantages of the geometricians’ approach are relatively straightforward. Their measures
of PLOS are fairly objective, based on the fundamentals of well-understood volume/capacity
ratios, and with variations allowing for factors such as turning movements or buffers from
moving vehicles. Their approaches are both empirical and relatively replicable. The
disadvantages their approaches have are also the reasons for the emergence of the competing
camp. The geometricians might rate a footpath that offers very good freedom of movement of
pedestrians very highly, even if it offers no shade, comfort or aesthetic amenity, and is located
immediately adjacent to a noisy and polluted, ten-lane, high-speed arterial.
The advantages of the experientialists’ approach are in its capacity to include almost any
factor via the use of a points system or another form of multi-criteria evaluation. They can
incorporate the most subjective factors, such as comfort and attractiveness. But the
experientialists cannot seem to agree on what those factors might be, or how one should
measure them. Too many of their studies rely on ‘best guess’ approaches from one or two
learned experts, rather than surveys with pedestrians to inform factor selection, albeit that the
best-guess approach is a beneficial method in qualitative research to open new perspectives
for further inquiry (Baxter & Eyles, 1997).
The factors others have used differ across each facility type (streets, footpaths, intersections
and midblock crossings). Factors included distinct objects in the built environment (i.e., the
presence of bollards, water fountains or pedestrian signals) collected from audits or
geographic information systems (GIS), and more subjective factors that represent an
environment’s effect on pedestrians (i.e., concepts of comfort, connectivity or security). The
results found that the most used factors were, in order: footpath width; obstructions to
pedestrian flow; motor vehicle speeds and volumes; shoulder widths; and buffers such as on-
street parking.
230
RQ2: What do experts think should be included for PLOS models appropriate for the
Australian context?
The study findings have demonstrated a novel, robust method for determining and ranking the
factors that affect PLOS using a Delphi process for two rounds, including an online survey
and a ‘walkshop’ with experts in Brisbane. This approach builds on previous approaches by
Khisty (1994), including by implementing this novel Dephi process. Inspection of the results
shows that the factors affecting PLOS can generally be classified into five categories
including geometric characteristics, obstructions and interruptions, aesthetics and amenities,
pedestrian traffic and adjacent traffic. In general, the factors with very high influence were
considered those with a relative importance index (RII) above 0.800, drawn from the Delphi
outputs. The factors the experts reported as having the highest RII at the end of the second-
round were from rank 1 (footpath connectivity) to rank 15 (crossing opportunities). The full
list of factors the experts suggested should be included is shown in Figure 3 and Table 2
(Chapter 5).
RQ3: What do pedestrians think about the importance level of a wide range of factors
affecting PLOS, and their relationship to each other?
As noted in our previous systematic review paper (Raad & Burke, 2018), previous PLOS
measurement approaches relied far too heavily on expert judgement alone, often of only one
or two experts, and had little in the way of testing with real-world pedestrians. Some
researchers, including Bian et al. (2009) and Kim et al. (2013), have sought pedestrian user
input. By taking the findings of the workshop and then field-testing them via the survey with
312 pedestrians in the street locations of interest, our research provided a relatively robust
method of answering the research question. The results revealed significant consistency
between the set of factors that were identified by experts through the Delphi process and the
perceptions of the pedestrians surveyed. These factors involve geometric characteristics,
obstructions and interruptions, aesthetics and amenities, pedestrian traffic and adjacent traffic.
Those surveyed were asked to score each factor regarding both the satisfaction (how satisfied
they were) and the importance (how important good performance was to them) using a five-
point Likert scale. The disgruntlement percentage for each factor was then calculated using
the cross-tabulating process of the percentage of respondents who were dissatisfied or
strongly dissatisfied, and who rated each aspect as important or extremely important based
upon the gap analysis approach developed by Stradling et al. (2007). The five factors the
231
pedestrians identified as causing most disgruntlement were the vehicle volume is not
reasonable, the buffer width is too narrow, there aren’t enough trees and plants, the footpath is
too narrow and there are too many other pedestrians. The full list is shown in Table 6
(Chapter 6).
RQ4: How could a more optimal pedestrian level of service (PLOS) audit tool for footpaths
in commercial streets be created which is appropriate for Australian cities?
In this study a reliable pedestrian level of service audit tool (PLOSAT) is developed using a
combination of methodological methods including the PRISMA technique, the Delphi
technique, a pedestrian intercept survey, road guidelines and standards, a point system
method, and an inter- and intra-rater reliability test. It is considered to be the first
comprehensive and reliable audit tool established to objectively quantify PLOS value based
on pedestrians’ perceptions of safety and comfort for footpaths in Australian commercial
streets. Further, adopting this approach for measuring PLOS using qualitative and quantitative
factors can supplement the results obtained through the quantitative PLOS of flow, speed, and
density, as elaborated in the Highway Capacity Manual (HCM) approach (HCM, 2010).
8.3 Contributions
The contributions of the thesis are summarised as being: i) conceptual or theoretical
contributions, ii) methodological contributions; and, iii) applied or practical contributions.
8.3.1 Conceptual/ theoretical contributions

There are only a few minor theoretical contributions made by this thesis. The most important
are:
 the theming and grouping of the alternative conceptions of PLOS measurement into
the geometric and experientialist approaches, and the lengthy exposition of their
respective strengths and weaknesses; and,
 The recognition that the concept of disgruntlement may be useful if picked up and
applied in this sub-field of research.
8.3.2 Methodological contributions

There are a larger set of methodological contributions made by this research, including that it:
 Is one of the first studies to use both expert opinions (from a large set of experts, not
just one or two) and surveys with a representative sample of pedestrian users of key
street segments, to refine and weight the list of factors for PLOS measurement;
232
 Is believed to be the first ever use of a two-stage Delphi process to elicit expert views
about the key factors to be used for PLOS measurement, and their relative importance;
 Is the first to make a disgruntlement approach work in this sub-field, in combination
with a point system method and a gap analysis approach;
 Provides a visual inspection of how dissatisfaction and importance are distributed
amongst the various PLOS measures; and,
 Includes a robust inter-rater reliability test to check for the reliability of the resultant
PLOSAT tool (a common failing of past efforts at audit tool development).
8.3.3 Applied/practical contributions

There are a number of applied/practical contributions. These include:
 The first ever systematic review of the literature in the sub-field of PLOS, which
brought attention to a large set of issues often hidden in the literature, and presented a
set of ways forward. That this paper won a best-paper prize from a group of
practitioners and academics such as the TRB Pedestrians Committee, was recognition
of its achievements;
 The spelling out of a significant research agenda for this sub-field, most notably in
Chapter 4;
 The identification of a few priority urban design and geometric issues in the Brisbane
CBD, brought to the attention of Brisbane City Council, through the workshop
processes; and,
 The PLOSAT is considered to be a specific method for checking the conditions of
pedestrian facilities and infrastructure as well as for evaluating the quality of service.
The PLOSAT tool itself is now being made available online to practitioners, and it can
be used as a reliable audit tool after the required training. An audit tool guideline
manual has been developed for field and desktop work to address general concerns,
explain audit methods and individual audit items, and present response scale
operational definitions. The audit protocol provides instructions and references for
data collection in the field and for calculating PLOS value. This guide consists of
general directions to run the auditing process as efficiently as possible. The major
tasks that need to be undertaken are detailed. The materials developed in this guide
include a detailed description of each audit element and how to measure each factor,
with illustrations.
233
The audit tool approach used in this study to evaluate PLOS on footpaths in commercial
streets in Australia can achieve other practical contributions including:
1. The provision of a reliable, empirical and comprehensive method to collect data,

measure factors affecting PLOS and evaluate PLOS for footpaths along commercial
streets.
2. The provision of a reliable method for evaluating and prioritising the need for footpath
retrofit construction on existing streets and also for designing new streets that promote
the transition to more sustainable environments.
3. The proposed audit tool aids planners/engineers in systematically assessing
pedestrians’ needs and in prioritising improvements, as well strengthening objectivity
in the decision making process.
4. The model can be used as an indicator for improving existing conditions by comparing
existing conditions with standards, while considering pedestrians’ perceptions.
5. The model provides a suitable tool for creating a clear image to guide decision makers
in evaluating, monitoring and comparing the performance and quality of pedestrian
facilities.
6. The model assists in the allocation of budgeting funds for facility changes and
improvements.
In terms of impact, a suggested change and reference to this research was included by the
Queensland Department of Transport and Main Roads in their feedback to a review of the
AUSTROADS Guide to Traffic Management (GTM) series and Guide to Road Design (GRD)
series in March 2019. As such, it is likely early research findings of this thesis will be
included in changes to the key guidance for state and local authorities for street design in
Australasia.
8.4 Limitations and future research directions

There are some limitations to this study which have implications for further research.
Limitations with the sample and the research methods include:
1. Although the results obtained from the Delphi process may look impressive, they do
not take into account the perceptions of a broader range of experts. The experts
involved in the Delphi process were predominantly white and middle-class, and, at
least for the initial survey, were predominantly male. All were physically able to walk
234
long distances. It is also important to take pedestrians of all ages and genders, from
young children to the elderly, into consideration. Though the broader pedestrian
surveys helped overcome this limitation, there appears a need to broaden the diversity
of the expert practitioners in Brisbane, Australia. This ‘stale, male and pale’ problem
has previously been recognised by professional associations, especially the local
engineering fraternity who have explicit programs to encourage female engineering
training.
2. Although the pedestrian survey sample was sizeable, obtaining 312 participants for
such a lengthy survey, it is not feasible to split the resulting dataset to disaggregate
sub-populations beyond the broad groupings that were analysed in Chapter 6. Using
sub-population enquiries to explore whether key results are similar for younger and
much older pedestrians is important, especially given the need to create age-friendly
cities. Future research efforts, particularly those with access to professional survey
support, could overcome this limitation, using these methods over a more diverse set
of times-of-day and targeting key sub-sets of users.
3. This study’s outputs, and PLOSAT itself, are solely limited to commercial streets in
the Australian setting. The tool could be used in a few other select locations, but it is
highly specific to this Australian context and does not take into account features of the
footpath environment that are found elsewhere (i.e., North American cities’ problems
with snow).
4. If there is a key methodological limitation identified it is that the results obtained from
pedestrians about the level of importance of the factors affecting PLOS in Chapter 6
are at times too homogeneous, as compared to the level of disgruntlement. The choice
of a five-point Likert scale, instead of a seven-point Likert scale may have influenced
this outcome. Future researchers might wish to adopt a finer-grained scale.
It is helpful to try to spell out a research agenda for this growing sub-field of research. As
highlighted most clearly in Chapter 4, it is suggested that:
1. It is likely that there are alternative formulations for selecting and including factors in
PLOS calculations that can go beyond this study’s approach and methods. And there
may be techniques identified that can obtain the input of both experts and users in
more efficient ways than undertaken in this study;
235
2. There is a lack of studies identifying the importance of specific factors, especially the
ways in which pedestrians’ opinions are treated. This issue goes well beyond PLOS,
per se, and is needed to inform a range of transport and land use planning policies and
programs. One especially under-researched factor is local pedestrian connectivity.
Whilst analysis methods such as Space Syntax offer interesting takes on connectivity
and are becoming widely used in academia and practice, more real-world analysis of
pedestrian networks based on sound theoretical conceptions is needed to identify the
key effects of network designs.
3. A wider research base covering more regions of the world is needed. The majority of
the literature is from North America. Walking rates in the USA are especially low by
world standards. PLOS is a critical issue for cities in Saudi Arabia, Iraq, Iran and
India, where the world’s largest pilgrimages occur, and where stampedes have resulted
in catastrophic loss of life (Aljohani, 2015; Haghighati & Hassan, 2013). While
sunlight may be considered a desirable pedestrian factor in the northern USA or
Canada, shade may be much more desirable in equatorial climates. The development
of factors which can be adapted across places within an overall framework would
assist.
4. There is little research examining a wide range of PLOS factors for disabled or special
needs users, with the exception of the study by Asadi-Shekari et al. (Asadi-Shekari,
Moeinaddini & Shah, 2013). An understanding of how the use of the PLOS for able-
bodied pedestrians might impact on those with disabilities would be useful.
5. Further disaggregation by facility type, particularly for different types of streets, may
make sense. Developing separate PLOS measures suitable for commercial streets, for
mixed-use streets or for suburban streets, may have significant merit.
6. Studies of the use of PLOS by agencies and the outcomes they obtain would be
desirable. When and where do these tools have most impact? What prevents agency
staff and developers adopting and using better approaches? It is not yet known quite
how influential different approaches to measuring PLOS are for real world transport
and built environment outcomes, or how design changes in the transport/land use
system influence user behaviour. Comparative studies may be possible where local
authorities under similar national/state policies may take different approaches to
236
measuring PLOS, creating different outcomes in street, link or intersection design,
eventually leading to changes in behaviour.
Bibliography
Abdalla, S. S., Elariane, S. A., & El Defrawi, S. H. (2015). Decision-Making Tool for
Participatory Urban Planning and Development: Residents’ Preferences of Their Built
Environment. Journal of Urban Planning and Development, 0(0), 04015011.
doi:10.1061/(asce)up.1943-5444.0000289
Alfonzo, M. A. (2005). To Walk or Not to Walk? The Hierarchy of Walking Needs.
Environment and Behavior, 37(6), 808-836. doi:10.1177/0013916504274016
Aljohani, A. M. (2015). Pilgrim crowd dynamics. University of Birmingham.
Andresen, E. M., Malmstrom, T. K., Wolinsky, F. D., Schootman, M., Miller, J. P., & Miller,
D. K. (2008). Rating neighborhoods for older adult health: results from the African
American Health study. BMC Public Health, 8(1), 1.
Asadi-Shekari, Z., Moeinaddini, M., & Shah, M. Z. (2013). Disabled Pedestrian Level of
Service Method for Evaluating and Promoting Inclusive Walking Facilities on Urban
Streets. Journal of Transportation Engineering-Asce, 139(2), 181-192.
doi:10.1061/(Asce)Te.1943-5436.0000492
Asadi-Shekari, Z., Moeinaddini, M., & Zaly Shah, M. (2012). Disabled pedestrian level of
service method for evaluating and promoting inclusive walking facilities on urban
streets. Journal of Transportation Engineering, 139(2), 181-192.
Asadi-Shekari, Z., Moeinaddini, M., & Zaly Shah, M. (2013). Non-motorised level of service:
addressing challenges in pedestrian and bicycle level of service. Transport reviews,
33(2), 166-194.
Asadi-Shekari, Z., & Zaly-Shah, M. (2011). Practical Evaluation Method for Pedestrian Level
of Service in Urban Streets' International Transport Research Conference. Penang,
Malaysia.
Austroads. (2017). Guide to road design part 6A: paths for walking and cycling.
Badland, H. M., Opit, S., Witten, K., Kearns, R. A., & Mavoa, S. (2010). Can Virtual
Streetscape Audits Reliably Replace Physical Streetscape Audits? Journal of Urban
Health : Bulletin of the New York Academy of Medicine, 87(6), 1007-1016.
doi:10.1007/s11524-010-9505-x
Badland, H. M., & Schofield, G. M. (2005). The built environment and transport-related
physical activity: what we do and do not know. Journal of Physical Activity and
Health, 2(4), 433-442.
Bain, L., Gray, B., & Rodgers, D. (2012). Living Streets : Strategies for Crafting Public
Space. Hoboken, NJ, USA: John Wiley & Sons.
Ballantyne, M., & Pickering, C. M. (2015). The impacts of trail infrastructure on vegetation
and soils: Current literature and future directions. Journal of Environmental
Management, 164, 53-64. doi:http://dx.doi.org/10.1016/j.jenvman.2015.08.032
Banister, D. (2002). Transport Planning : In the UK, USA and Europe Retrieved from
http://Griffith.eblib.com.au/patron/FullRecord.aspx?p=181816
Barker, L. (2012). Exploring the Relationship Between Walkability and the Built
Environment: A Case Study of Three Intersections in Seattle's University District.
(1508909 M.U.P.), University of Washington, Ann Arbor. ProQuest Dissertations &
Theses Global database.
237
Baxter, J., & Eyles, J. (1997). Evaluating qualitative research in social geography:
establishing ‘rigour’in interview analysis. Transactions of the Institute of British
geographers, 22(4), 505-525.
Bedimo-Rung, A. L., Gustat, J., Tompkins, B. J., Rice, J., & Thomson, J. (2006).
Development of a direct observation instrument to measure environmental
characteristics of parks for physical activity. Journal of Physical Activity & Health, 3,
S176.
Benz, G. P. (1986). PEDESTRIAN TIME-SPACE CONCEPT. A NEW APPROACH TO
THE PLANNING AND DESIGN OF PEDESTRIAN FACILITIES.
Bian, Y., Wang, W., Lu, J., Ma, J., & Tan, D. (2007). Pedestrian level of service for sidewalks
in China. Paper presented at the Transportation Research Board 86th Annual Meeting.
Boarnet, M. G., Day, K., Alfonzo, M., Forsyth, A., & Oakes, M. (2006). The Irvine–
Minnesota inventory to measure built environments: reliability tests. American
Journal of Preventive Medicine, 30(2), 153-159. e143.
Boodlal, L. (2004). Accessible sidewalk and street crossing-AN informational guide.
Retrieved from
Brancheau, J. C., & Wetherbe, J. C. (1987). Key issues in information systems management.
MIS quarterly, 23-45.
Brisbane City Council. (2014). Brisbane Inner City Investment Prospectus.
Brownson, R. C., Hoehner, C. M., Brennan, L. K., Cook, R. A., Elliott, M. B., & McMullen,
K. M. (2004). Reliability of 2 instruments for auditing the environment for physical
activity. J Phys Act Health, 1, 191-208.
Brownson, R. C., Hoehner, C. M., Day, M. K., Forsyth, A., & Sallis, J. F. (2009). Measuring
the Built Environment for Physical Activity. American Journal of Preventive
Medicine, 36(4), S99-S123.e112.
Carreno, M., Willis, A., & Stradling, S. (2002). Quality of service for pedestrians: closing the
gaps in knowledge Traffic And Transportation Studies (2002) (pp. 326-333).
Caughy, M. O., O’Campo, P. J., & Patterson, J. (2001). A brief observational measure for
urban neighborhoods. Health & Place, 7(3), 225-236.
Celik, E., Aydin, N., & Gumus, A. T. (2014). A multiattribute customer satisfaction
evaluation approach for rail transit network: A real case study for Istanbul, Turkey.
Chaturvedi, H., & Shweta, R. (2015). Evaluation of inter-rater agreement and inter-rater
reliability for observational data: An overview of concepts and methods. JIAAP, 41,
20-27.
City of Sydney Council. (2013). Sydney Streets Design Code Street Design Coordination
Part E. Sydney: City of Sydney.
Clifton, K. J., Smith, A. D. L., & Rodriguez, D. (2007). The development and testing of an
audit for the pedestrian environment. Landscape and Urban Planning, 80(1), 95-110.
Craig, C. L., Brownson, R. C., Cragg, S. E., & Dunn, A. L. (2002). Exploring the effect of the
environment on physical activity: a study examining walking to work. American
Journal of Preventive Medicine, 23(2), 36-43.
Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods
approaches: Sage publications.
Creswell, J. W., & Clark, V. L. P. (2007). Designing and conducting mixed methods research.
Creswell, J. W., & Miller, D. L. (2000). Determining validity in qualitative inquiry. Theory
into practice, 39(3), 124-130.
Croft, P., Elazar, N., & Levasseur, M. (2013). Guide information for pedestrian facilities.
238
Cunningham, G. O., Michael, Y. L., Farquhar, S. A., & Lapidus, J. (2005). Developing a
reliable senior walking environmental assessment tool. American Journal of
Preventive Medicine, 29(3), 215-217.
Dannenberg, A. L., Cramer, T. W., & Gibson, C. J. (2005). Assessing the Walkability of the
Workplace: A New Audit Tool. American Journal of Health Promotion, 20(1), 39-44.
doi:doi:10.4278/0890-1171-20.1.39
Davis, D., & Braaksma, J. (1987). Level-of-service standards for platooning pedestrians in
transportation terminals. ITE journal, 57(4), 31-35.
Day, K., Boarnet, M., Alfonzo, M., & Forsyth, A. (2006). The Irvine–Minnesota inventory to
measure built environments: development. American Journal of Preventive Medicine,
30(2), 144-152.
De Vet, E., Brug, J., De Nooijer, J., Dijkstra, A., & De Vries, N. K. (2004). Determinants of
forward stage transitions: a Delphi study. Health Education Research, 20(2), 195-205.
Denzin, N. K., & Lincoln, Y. S. (2005). The Sage handbook of qualitative research: Sage.
Dixon, L. (1996). Bicycle and pedestrian level-of-service performance measures and
standards for congestion management systems. Transportation Research Record:
Journal of the Transportation Research Board(1538), 1-9.
Doke, E. R., & Swanson, N. (1995). Decision variables for selecting prototyping in
information systems development: A Delphi study of MIS managers. Information &
Management, 29(4), 173-182.
Ehrenfeucht, R. (2006). Constructing the public in urban space: Streets, sidewalks and
municipal regulation in Los Angeles, 1880--1940. (3226054 Ph.D.), University of
California, Los Angeles, Ann Arbor. ProQuest Dissertations & Theses Global
database.
Emery, J., Crump, C., & Bors, P. (2003). Reliability and validity of two instruments designed
to assess the walking and bicycling suitability of sidewalks and roads. American
Journal of Health Promotion, 18(1), 38-46.
Ensley, J. O. (2012). Application of Highway Capacity Manual 2010 Level-of-Service
Methodologies for Planning Deficiency Analysis.
Ewing, R., & Handy, S. (2009). Measuring the Unmeasurable: Urban Design Qualities
Related to Walkability. Journal of Urban Design, 14(1), 65-84.
doi:10.1080/13574800802451155
Ewing, R., Handy, S., Brownson, R. C., Clemente, O., & Winston, E. (2006). Identifying and
measuring urban design qualities related to walkability. Journal of Physical Activity &
Health, 3, S223.
Ewing, R., Meakins, G., Hamidi, S., & Nelson, A. C. (2014). Relationship between urban
sprawl and physical activity, obesity, and morbidity – Update and refinement. Health
& Place, 26, 118-126. doi:http://dx.doi.org/10.1016/j.healthplace.2013.12.008
Ewing, R. H. (1996). Pedestrian-and transit-friendly design.
Flanders, F. B. (1988). Determining curriculum content for nursery/landscape course work in
vocational agriculture for the twenty-first century: A futures study utilizing the Delphi
technique. (8823794 Ed.D.), University of Georgia, Ann Arbor. ProQuest
Dissertations & Theses Global database.
Forsyth, A., Hearst, M., Oakes, J. M., & Schmitz, K. H. (2008). Design and Destinations:
Factors Influencing Walking and Total Physical Activity. Urban Studies, 45(9), 1973-
1996. doi:10.1177/0042098008093386
Foster, S., Giles-Corti, B., & Knuiman, M. (2011). Creating safe walkable streetscapes: Does
house design and upkeep discourage incivilities in suburban neighbourhoods? Journal
239
of Environmental Psychology, 31(1), 79-88.
doi:http://dx.doi.org/10.1016/j.jenvp.2010.03.005
Frackelton, A., Grossman, A., Palinginis, E., Castrillon, F., Elango, V., & Guensler, R.
(2013). Measuring walkability: Development of an automated sidewalk quality
assessment tool. Suburban Sustainability, 1(1), 4.
Frank, L., Engelke, P., & Schmid, T. (2003). Health and community design: The impact of the
built environment on physical activity: Island Press.
Friman, M., Edvardsson, B., & Gärling, T. (2001). Frequency of negative critical incidents
and satisfaction with public transport services. I. Journal of Retailing and Consumer
Services, 8(2), 95-104.
Friman, M., Fujii, S., Ettema, D., Gärling, T., & Olsson, L. E. (2013). Psychometric analysis
of the satisfaction with travel scale. Transportation Research Part A: Policy and
Practice, 48, 132-145.
Friman, M., & Gärling, T. (2001). Frequency of negative critical incidents and satisfaction
with public transport services. II. Journal of Retailing and Consumer Services, 8(2),
105-114.
Fruin, J. J. (1970). Designing for Pedestrians, a Level of Service Concept. Mechanical
Engineering, 92(12), 63-&.
Fruin, J. J. (1971). Pedestrian planning and design. Retrieved from
Fujiyama, T., Childs, C., Boampomg, D., & Tyler, N. (2005). Investigation of Lighting Levels
for Pedestrians-Some questions about lighting levels of current lighting standards.
Gallin, N. (2001). Quantifying pedestrian friendliness--guidelines for assessing pedestrian
level of service. Road & Transport Research, 10(1), 47.
Gauvin, L., Richard, L., Craig, C. L., Spivock, M., Riva, M., Forster, M., . . . Gagnon, H.
(2005). From walkability to active living potential: an “ecometric” validation study.
American Journal of Preventive Medicine, 28(2), 126-133.
Gillham, B. (2000). Case study research methods: Bloomsbury Publishing.
Griew, P., Hillsdon, M., Foster, C., Coombes, E., Jones, A., & Wilkinson, P. (2013).
Developing and testing a street audit tool using Google Street View to measure
environmental supportiveness for physical activity. International Journal of
Behavioral Nutrition and Physical Activity, 10(1), 103. doi:10.1186/1479-5868-10-
103
Guba, E. G. (1990). The paradigm dialog: Sage Publications.
Guba, E. G., & Lincoln, Y. S. (1994). Competing paradigms in qualitative research.
Handbook of qualitative research, 2(163-194), 105.
Guitart, D., Pickering, C., & Byrne, J. (2012). Past results and future directions in urban
community gardens research. Urban Forestry & Urban Greening, 11(4), 364-373.
doi:http://dx.doi.org/10.1016/j.ufug.2012.06.007
Guo, H., Zhao, F., Wang, W., Zhou, Y., Zhang, Y., & Wets, G. (2014). Modeling the
Perceptions and Preferences of Pedestrians on Crossing Facilities. Discrete Dynamics
in Nature and Society, 2014, 8. doi:10.1155/2014/949475
Haghighati, R., & Hassan, A. (2013). Modeling the flow of crowd during tawaf at Masjid Al-
Haram. Journal of Mek, 36, 2-18.
Hamed, M. M. (2001). Analysis of pedestrians’ behavior at pedestrian crossings. Safety
Science, 38(1), 63-82. doi:http://dx.doi.org/10.1016/S0925-7535(00)00058-8
Heath, G. W., Brownson, R. C., Kruger, J., Miles, R., Powell, K. E., Ramsey, L. T., &
Services, T. F. o. C. P. (2006). The effectiveness of urban design and land use and
transport policies and practices to increase physical activity: a systematic review.
240
Hensher, D. A. (1990). Hierarchical stated response designs-an application to bus user
preferences. Logistics and Transportation Review, 26(4), 299.
Henson, C. (2000). Levels of service for pedestrians. Institute of Transportation Engineers.
ITE Journal, 70(9), 26.
Hess, F., Salze, P., Weber, C., Feuillet, T., Charreire, H., Menai, M., . . . Oppert, J.-M. (2017).
Active Mobility and Environment: A Pilot Qualitative Study for the Design of a New
Questionnaire. PLoS One, 12(1), e0168986.
Hidayat, Choocharukul, K., & Kishi, K. (2012). Important factors on sidewalk with vendor
activities based on pedestrian perception by gender and age Jurnal Transportasi,
12(3).
Hidayat, Choocharukul, K., & Kishi, N. (2010). Analysis of Sidewalk Level of Service
Incorporating Pedestrian’s Perceptions’ Towards Traffic and Transport Safety. Paper
presented at the The 7th National Transport Conference: NTC7, 15th October 2010.
Hidayat, Choocharukul, K., Nakatsuji, T., & Kishi, K. (2009). Determination of factors
related to sidewalk performance based on pedestrian perception. Paper presented at
the 40th Conference of Infrastructure Planning Japan Society of Civil Engineers,
Kanazawa University.
Hidayat, N., Choocharukul, K., & Kishi, K. (2011). Pedestrian Level of Service Model
Incorporating Pedestrian Perception for Sidewalk with Vendor Activities. Journal of
the Eastern Asia Society for Transportation Studies, 9, 1012-1023.
doi:10.11175/easts.9.1012
Hoehner, C. M., Ivy, A., Brennan Ramirez, L. K., Handy, S., & Brownson, R. C. (2007).
Active neighborhood checklist: a user-friendly and reliable tool for assessing activity
friendliness. American Journal of Health Promotion, 21(6), 534-537.
Jacobs. (1993). Great streets (Vol. null).
Jaskiewicz, F. (2000). Pedestrian level of service based on trip quality. Transportation
Research Circular, TRB.
Jensen, S. (2007). Pedestrian and bicyclist level of service on roadway segments.
Transportation Research Record: Journal of the Transportation Research
Board(2031), 43-51.
Jones, D. L. (2012). The Behavioral Impacts of Urban Street Modifications: A Case Study of
East Blvd. in Charlotte, NC. (3538279 Ph.D.), North Carolina State University, Ann
Arbor. ProQuest Dissertations & Theses Global database.
Kang, L., Xiong, Y., & Mannering, F. L. (2013). Statistical analysis of pedestrian perceptions
of sidewalk level of service in the presence of bicycles. Transportation Research Part
A: Policy and Practice, 53, 10-21.
Kang, L., Xiong, Y. G., & Mannering, F. L. (2013). Statistical analysis of pedestrian
perceptions of sidewalk level of service in the presence of bicycles. Transportation
Research Part a-Policy and Practice, 53, 10-21. doi:10.1016/j.tra.2013.05.002
Kerridge, J., Hine, J., & Wigan, M. (2001). Agent-Based Modelling of Pedestrian
Movements: The Questions That Need to Be Asked and Answered. Environment and
Planning B: Planning and Design, 28(3), 327-341. doi:10.1068/b2696
Khisty, C. J. (1994). Evaluation of pedestrian facilities: beyond the level-of-service concept.
Transportation Research Record (1438), 45-50.
Kim, S., Choi, J., Kim, S., & Tay, R. (2014). Personal space, evasive movement and
pedestrian level of service. Journal of Advanced Transportation, 48(6), 673-684.
doi:10.1002/atr.1223
241
Kim, S., Choi, J., & Kim, Y. (2011). Determining the sidewalk pavement width by using
pedestrian discomfort levels and movement characteristics. KSCE Journal of Civil
Engineering, 15(5), 883.
Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and reporting intraclass correlation
coefficients for reliability research. Journal of chiropractic medicine, 15(2), 155-163.
Koohsari, M. J., Karakiewicz, J. A., & Kaczynski, A. T. (2013). Public Open Space and
Walking: The Role of Proximity, Perceptual Qualities of the Surrounding Built
Environment, and Street Configuration. Environment and Behavior, 45(6), 706-736.
doi:10.1177/0013916512440876
Lam, W. H., Morrall, J. F., & Ho, H. (1995). Pedestrian flow characteristics in Hong Kong.
Transportation Research Record(1487), 56-62.
Landis, B., Vattikuti, V., Ottenberg, R., McLeod, D., & Guttenplan, M. (2001). Modeling the
roadside walking environment - Pedestrian level of service. Transportation Research
Record(1773), 82-88.
Lawlor, D. A., Ness, A. R., Cope, A., Davis, A., Insall, P., & Riddoch, C. (2003). The
challenges of evaluating environmental interventions to increase population levels of
physical activity: the case of the UK National Cycle Network. Journal of
Epidemiology and Community Health, 57(2), 96-101.
Lawshe, C. H. (1975). A quantitative approach to content validity. Personnel psychology,
28(4), 563-575.
Lee, B. J., Jang, T. Y., Wang, W., & Namgung, M. (2009). Design criteria for an urban
sidewalk landscape considering emotional perception. Journal of Urban Planning and
Development, 135(4), 133-140.
Lee, R. E., Booth, K. M., Reese-Smith, J. Y., Regan, G., & Howard, H. H. (2005). The
Physical Activity Resource Assessment (PARA) instrument: evaluating features,
amenities and incivilities of physical activity resources in urban neighborhoods.
International Journal of Behavioral Nutrition and Physical Activity, 2(1), 13.
Lillebye, E. (1996). Architectural and functional relationships in street planning: an historical
view. Landscape and Urban Planning, 35(2–3), 85-105.
doi:http://dx.doi.org/10.1016/0169-2046(96)00307-6
Lincoln, Y. S., Lynham, S. A., & Guba, E. G. (2011). Paradigmatic controversies,
contradictions, and emerging confluences, revisited. The Sage handbook of qualitative
research, 4, 97-128.
Lo, R. H. (2009). Walkability: what is it? Journal of Urbanism: International Research on
Placemaking and Urban Sustainability, 2(2), 145-166.
doi:10.1080/17549170903092867
Loukaitou-Sideris, A. (1997). Inner-City Commercial Strips: Evolution, Decay: Retrofit? The
Town Planning Review, 68(1), 1-29.
Maghelal, P. K., & Capp, C. J. (2011). Walkability: A review of existing pedestrian indices.
Journal of the Urban and Regional Information Systems Association, 23(2), 5-19.
Maghelal, P. K., & Capp, C. J. (2011). Walkability: a review of existing pedestrian indices.
URISA Journal, 23(2), 5-20.
HCM 2010 : Highway Capacity Manual. National Research Council, Washington, DC, 113.
Mateo-Babiano, I. B., & Ieda, H. (2007). Street space sustainability in Asia: The role of the
Asian pedestrian and street culture. Paper presented at the Proceedings of the Eastern
Asia Society for Transportation Studies.
McDonough, S. (1997). Research methods as part of English language teacher education.
English Language Teacher Education and Development, 3(1), 84-96.
242
Mehta, V. (2006). Lively streets: Exploring the relationship between built environment and
social behavior. (3241522 Ph.D.), University of Maryland, College Park, Ann Arbor.
ProQuest Dissertations & Theses Global database.
Mehta, V. (2007). Lively Streets: Determining Environmental Characteristics to Support
Social Behavior. Journal of Planning Education and Research, 27(2), 165-187.
doi:10.1177/0739456x07307947
Mehta, V. (2008). Walkable streets: pedestrian behavior, perceptions and attitudes. Journal of
Urbanism: International Research on Placemaking and Urban Sustainability, 1(3),
217-245. doi:10.1080/17549170802529480
Mehta, V. (2009). Look Closely and You Will See, Listen Carefully and You Will Hear:
Urban Design and Social Interaction on Streets. Journal of Urban Design, 14(1), 29-
64. doi:10.1080/13574800802452658
Mehta, V., & Bosson, J. K. (2010). Third Places and the Social Life of Streets. Environment
and Behavior, 42(6), 779-805. doi:10.1177/0013916509344677
Miller, J., Bigelow, J., & Garber, N. (2000). Calibrating pedestrian level-of-service metrics
with 3-D visualization. Transportation Research Record: Journal of the
Transportation Research Board(1705), 9-15.
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & The, P. G. (2009). Preferred Reporting
Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS
Med, 6(7), e1000097. doi:10.1371/journal.pmed.1000097
Moreland-Russell, S., Eyler, A., Barbero, C., Hipp, J. A., & Walsh, H. (2013). Diffusion of
complete streets policies across US communities. Journal of Public Health
Management and Practice, 19, S89-S96.
Moreland City Council. (2010). BRUNSWICK STRUCTURE PLAN LYGON STREET.
Moreland City Council.
Morgan, M. S. (2012). Case Studies: One Observation or Many? Justification or Discovery?
Philosophy of Science, 79(5), 667-677.
Moudon, A. V., & Lee, C. (2003). Walking and bicycling: an evaluation of environmental
audit instruments. American Journal of Health Promotion, 18(1), 21-37.
Moughtin, J. C. (2003). Urban Design : Street and Square. Jordan Hill, GBR: Routledge.
Mozer, D. (1994). Calculating Multi-Modal Levels-of-Service (Abridged). International
Bicycle Fund.
Muraleetharan, T., Adachi, T., Hagiwara, T., & Kagaya, S. (2004 B). Method to determine
overall level-of-service of pedestrians on sidewalks based on total utility value. Paper
presented at the Proceedings of the 83rd Annual Meeting of Transportation Research
Board, Washington DC.
Muraleetharan, T., Adachi, T., Hagiwara, T., & Kagaya, S. (2005). Method to determine
pedestrian level-of-service for crosswalks at urban intersections. Journal of the
Eastern Asia Society for Transportation Studies, 6, 127-136.
Muraleetharan, T., Adachi, T., Hagiwara, T., Kagaya, S., & Uchida, K. (2004 A). Evaluation
of Pedestrian Level-of-Service on Sidewalks and Crosswalks Using Conjoint Analysis.
Paper presented at the Transportation Research board Annual Meeting, Washington
DC, Paper.
Muraleetharan, T., ADACHI, T., UCHIDA, K.-e., HAGIWARA, T., & KAGAYA, S. (2004).
A study on evaluation of pedestrian level of service along sidewalks and at crosswalks
using conjoint analysis. INFRASTRUCTURE PLANNING REVIEW, 21, 727-735.
Nabors, D., Gibbs, M., Sandt, L., Rocchi, S., Wilson, E., & Lipinski, M. (2007). Pedestrian
road safety audit guidelines and prompt lists. Retrieved from
243
Næss, P., & Jensen, O. B. (2002). Urban land use, mobility and theory of science: exploring
the potential for critical realism in empirical research. Journal of Environmental
Policy & Planning, 4(4), 295-311. doi:10.1002/jepp.114
Nasar, J. L., & Hong, X. (1999). Visual preferences in urban signscapes. Environment and
Behavior, 31(5), 671-691.
NZ Transport Agency. (2009). Pedestrian planning and design guide. New Zealand: NZ
Transport Agency
Oldenburg, R. (2001). Celebrating the third place: Inspiring stories about the" great good
places" at the heart of our communities: Da Capo Press.
Øvstedal, L., & Ryeng, E. (2004). PROMPT Pedestrian Comfort Synthesis Report:
Trondheim: SINTEF Teknologi og samfunn.
Petritsch, T., Landis, B., McLeod, P., Huang, H., Challa, S., Skaggs, C., . . . Vattikuti, V.
(2006). Pedestrian level-of-service model for urban arterial facilities with sidewalks.
Transportation Research Record: Journal of the Transportation Research
Board(1982), 84-89.
Petticrew, M. (2001). Systematic reviews from astronomy to zoology: myths and
misconceptions. BMJ : British Medical Journal, 322(7278), 98-101.
Pickering, C., & Byrne, J. (2014). The benefits of publishing systematic quantitative literature
reviews for PhD candidates and other early-career researchers. Higher Education
Research & Development, 33(3), 534-548. doi:10.1080/07294360.2013.841651
Pikora, T., Giles-Corti, B., Bull, F., Jamrozik, K., & Donovan, R. (2003). Developing a
framework for assessment of the environmental determinants of walking and cycling.
Social Science & Medicine, 56(8), 1693-1703.
Pikora, T. J., Bull, F. C. L., Jamrozik, K., Knuiman, M., Giles-Corti, B., & Donovan, R. J.
(2002). Developing a reliable audit instrument to measure the physical environment
for physical activity. American Journal of Preventive Medicine, 23(3), 187-194.
Polus, A., Schofer, J. L., & Ushpiz, A. (1983). Pedestrian Flow and Level of Service. Journal
of Transportation Engineering-Asce, 109(1), 46-56. doi:doi:10.1061/(ASCE)0733-
947X(1983)109:1(46)
Pushkarev, B., & Zupan, J. (1975). Urban space for pedestrian. A Report of the Regional
Planning Association, The MIT Press, Cambridge, Mass.
Raad, N., & Burke, M. (2017). Pedestrian Level of Service Tools: problems of conception,
factor identification, measurement and usefulness. Paper presented at the 39th
Australian Transport Research Forum (ATRF), Auckland, New Zealand
Raad, N., & Burke, M. (2018). What Are the Most Important Factors for Pedestrian Level-of-
Service Estimation? A Systematic Review of the Literature. Transportation Research
Record, 0361198118790623.
Raad, N., & Burke, M. (2019). Investigating relationship between the importance of the
factors affecting Pedestrian Level of Service PLOS and level of satisfaction on footpath. Paper
presented at the World Conference on Transport Research (WCTR), Mumbai.
Randall, T. A., & Baetz, B. W. (2001). Evaluating pedestrian connectivity for suburban
sustainability. Journal of Urban Planning and Development, 127(1), 1-15.
Raskin, M. S. (1994). The Delphi study in field instruction revisited: Expert consensus on
issues and research priorities. Journal of Social Work Education, 30(1), 75-89.
Rayens, M. K., & Hahn, E. J. (2000). Building consensus using the policy Delphi method.
Policy, politics, & nursing practice, 1(4), 308-315.
Roy, S., Byrne, J., & Pickering, C. (2012). A systematic quantitative review of urban tree
benefits, costs, and assessment methods across cities in different climatic zones.
244
Urban Forestry & Urban Greening, 11(4), 351-363.
doi:http://dx.doi.org/10.1016/j.ufug.2012.06.006
Ruzzene, A. (2014). Using Case Studies in the Social Sciences: Methods, Inferences,
Purposes. Erasmus Universiteit.
Saelens, B. E., Frank, L. D., Auffrey, C., Whitaker, R. C., Burdette, H. L., & Colabianchi, N.
(2006). Measuring physical environments of parks and playgrounds: EAPRS
instrument development and inter-rater reliability. Journal of Physical Activity &
Health, 3, S190.
Saelens, B. E., & Handy, S. L. (2008). Built environment correlates of walking: a review.
Medicine and science in sports and exercise, 40(7 Suppl), S550.
Saelens, B. E., Sallis, J. F., Black, J. B., & Chen, D. (2003). Neighborhood-Based Differences
in Physical Activity: An Environment Scale Evaluation. American Journal of Public
Health, 93(9), 1552-1558. doi:10.2105/AJPH.93.9.1552
Saelens, B. E., Sallis, J. F., & Frank, L. D. (2003). Environmental correlates of walking and
cycling: Findings from the transportation, urban design, and planning literatures.
Annals of Behavioral Medicine, 25(2), 80-91. doi:10.1207/s15324796abm2502_03
Saito, Y., Oguma, Y., Inoue, S., Tanaka, A., & Kobori, Y. (2013). Environmental and
individual correlates of various types of physical activity among community-dwelling
middle-aged and elderly Japanese. International journal of environmental research
and public health, 10(5), 2028-2042.
Sarkar, S. (1993). Determination of service levels for pedestrians, with European examples.
Transportation Research Record(1405).
Sarkar, S. (2003). Qualitative evaluation of comfort needs in urban walkways in major
activity centers. Transportation quarterly, 57(4), 39-59.
Sayer, A. (2000). Realism and social science: Sage.
Schmidt, R. C. (1997). Managing Delphi surveys using nonparametric statistical techniques.
decision Sciences, 28(3), 763-774.
Seagle, E. D., & Iverson, M. (2001). Characteristics of the turfgrass industry in 2020: a
Delphi study with implications for agricultural education programs. Citeseer.
Shash, A. A. (1993). Factors considered in tendering decisions by top UK contractors.
Construction Management and Economics, 11(2), 111-118.
doi:10.1080/01446199300000004
Shen, W., Xiao, W., & Wang, X. (2016). Passenger satisfaction evaluation model for Urban
rail transit: A structural equation modeling based on partial least squares. Transport
Policy, 46, 20-31.
Singh, K., & Jain, P. (2011). Methods of assessing pedestrian level of service. Journal of
Engineering Research and Studies, 2(1), 116-124.
Sisiopiku, V. P., & Akin, D. (2003). Pedestrian behaviors at and perceptions towards various
pedestrian facilities: an examination based on observation and survey data.
Transportation Research Part F: Traffic Psychology and Behaviour, 6(4), 249-274.
doi:http://dx.doi.org/10.1016/j.trf.2003.06.001
Sisiopiku, V. P., & Byrd, J. E. (2006). Comparison of Level-of-Service Methodologies for
Pedestrian Sidewalks. Paper presented at the Transportation Research Board 85th
Annual Meeting.
Siu, K. W. M., & Wong, K. S. L. (2015). Flexible design principles: Street furniture design
for transforming environments, diverse users, changing needs and dynamic
interactions. Facilities, 33(9/10), 588-621. doi:doi:10.1108/F-02-2014-0021
245
Skelton, C., Koplin, M., & Cipolla, V. (2011). Massively participatory urban planning and
design tools and process: the Betaville project. Paper presented at the Proceedings of
the 12th Annual International Digital Government Research Conference: Digital
Government Innovation in Challenging Times, College Park, Maryland, USA.
Smith, A. L., Troped, P. J., McDonough, M. H., & DeFreese, J. (2015). Youth perceptions of
how neighborhood physical environment and peers affect physical activity: a focus
group study. International Journal of Behavioral Nutrition and Physical Activity,
12(1), 80.
Smith, R., Reed, S., & Baker, S. (2010). Street design: Part 1. Complete streets. Public roads,
74(1), 12-17.
Soligo, M. J., Irwin, P. A., Williams, C. J., & Schuyler, G. D. (1998). A comprehensive
assessment of pedestrian comfort including thermal effects. Journal of Wind
Engineering and Industrial Aerodynamics, 77–78, 753-766.
Sousa, L., Sequeira, C., & Ferré-Grau, C. ‘Living together with dementia’ – Conceptual
validation of training programme for family caregivers: Innovative practice.
Dementia, 0(0), 1471301218762565. doi:10.1177/1471301218762565
Spittaels, H., Foster, C., Oppert, J.-M., Rutter, H., Oja, P., Sjöström, M., & De
Bourdeaudhuij, I. (2009). Assessment of environmental correlates of physical activity:
development of a European questionnaire. International Journal of Behavioral
Nutrition and Physical Activity, 6(1), 39. doi:10.1186/1479-5868-6-39
Spivock, M., Gauvin, L., & Brodeur, J.-M. (2007). Neighborhood-level active living buoys
for individuals with physical disabilities. American Journal of Preventive Medicine,
32(3), 224-230.
Steven, R., Pickering, C., & Guy Castley, J. (2011). A review of the impacts of nature based
recreation on birds. Journal of Environmental Management, 92(10), 2287-2294.
doi:http://dx.doi.org/10.1016/j.jenvman.2011.05.005
Stradling, S. G., Anable, J., & Carreno, M. (2007). Performance, importance and user
disgruntlement: A six-step method for measuring satisfaction with travel modes.
Transportation Research Part A: Policy and Practice, 41(1), 98-106.
Su, M., Du, Y.-k., Liu, Q.-m., Ren, Y.-j., Kawachi, I., Lv, J., & Li, L.-m. (2014). Objective
assessment of urban built environment related to physical activity—development,
reliability and validity of the China Urban Built Environment Scan Tool (CUBEST).
BMC Public Health, 14(1), 1.
Suminski, R. R., Heinrich, K. M., Poston, W. S., Hyder, M., & Pyle, S. (2008).
Characteristics of urban sidewalks/streets and objectively measured physical activity.
Journal of Urban Health, 85(2), 178-190.
Suminski, R. R., Petosa, R. L., & Stevens, E. (2006). A method for observing physical
activity on residential sidewalks and streets. Journal of Urban Health, 83(3), 434-443.
Sung, H., Go, D., Choi, C.-g., Cheon, S., & Park, S. (2015). Effects of street-level physical
environment and zoning on walking activity in Seoul, Korea. Land Use Policy, 49,
152-160.
Susam-Sarajeva, Ş. (2009). The case study research method in translation studies. The
Interpreter and Translator Trainer, 3(1), 37-56.
Tanaboriboon, Y., & Guyano, J. A. (1989). Level of service standards for pedestrian facilities
in Bangkok: A case study. ITE journal, 59(11), 39-41.
Teddlie, C., & Tashakkori, A. (2009). Foundations of mixed methods research: Integrating
quantitative and qualitative approaches in the social and behavioral sciences: Sage
Publications Inc.
246
Todorova, A., Asakawa, S., & Aikoh, T. (2004). Preferences for and attitudes towards street
flowers and trees in Sapporo, Japan. Landscape and Urban Planning, 69(4), 403-416.
Transport for London. (2010). Pedestrian Comfort Level Guidance. London: Transport for
London.
Transport, Q. (2005). Easy Steps: A toolkit for planning, designing and promoting safe
walking. Brisbane, Australia.
Troped, P. J., Cromley, E. K., Fragala, M. S., Melly, S. J., Hasbrouck, H. H., Gortmaker, S.
L., & Brownson, R. C. (2006). Development and reliability and validity testing of an
audit tool for trail/path characteristics: the Path Environment Audit Tool (PEAT).
U.S. Department of Transportation. (2007). Pedestrian Road Safety Audits Guidelines and
Promplists. U.S. Department of Transportation: Federal Highway Administration.
Van Cauwenberg, J., Van Holle, V., Simons, D., Deridder, R., Clarys, P., Goubert, L., . . .
Deforche, B. (2012). Environmental factors influencing older adults’ walking for
transportation: a study using walk-along interviews. International Journal of
Behavioral Nutrition and Physical Activity, 9(1), 85.
Viera, A. J., & Garrett, J. M. (2005). Understanding interobserver agreement: the kappa
statistic. Fam Med, 37(5), 360-363.
von der Gracht, H. A. (2012). Consensus measurement in Delphi studies: Review and
implications for future quality assurance. Technological Forecasting and Social
Change, 79(8), 1525-1536. doi:https://doi.org/10.1016/j.techfore.2012.04.013
Wang, W., Li, P., Wang, W., & Namgung, M. (2011). Exploring determinants of pedestrians’
satisfaction with sidewalk environments: Case study in Korea. Journal of Urban
Planning and Development, 138(2), 166-172.
Weich, S., Burton, E., Blanchard, M., Prince, M., Sproston, K., & Erens, B. (2001).
Measuring the built environment: validity of a site survey instrument for use in urban
settings. Health & Place, 7(4), 283-292. doi:http://dx.doi.org/10.1016/S1353-
8292(01)00019-3
Weliange, S. H. D. S., Fernando, D., & Gunatilake, J. (2014). Development and validation of
a tool to assess the physical and social environment associated with physical activity
among adults in Sri Lanka. BMC Public Health, 14(1), 423.
Western Australia Department of Transport. (2011). WALKABILITY AUDIT TOOL. Western
Australia: Department of Transport Retrieved from
www.transport.wa.gov.au/walking.
Western Australia Department of Transport, Planning, & Public Transport Authority. (2011).
Planning and designing for pedestrians: guidelines. In. Western Australia: Department
of Transport.
Williams, J. E., Evans, M., Kirtland, K. A., Cavnar, M. M., Sharpe, P. A., Neet, M. J., &
Cook, A. (2005). Development and use of a tool for assessing sidewalk maintenance
as an environmental support of physical activity. Health Promotion Practice, 6(1), 81-
88.
Williams, K. (2002). Exploring resident preferences for street trees in Melbourne, Australia.
Journal of Arboriculture, 28(4), 161-170.
Wu, J., Radwan, E., & Abou-Senna, H. (2018). Assessment of pedestrian-vehicle conflicts
with different potential risk factors at midblock crossings based on driving simulator
experiment. Advances in transportation studies, 44, 33-46.
www.activelivingresearch.org.
Yin, R. (1994). Case study research: Design and methods . Beverly Hills: CA: Sage
publishing.
247
Yin, R. (2012). A (very) brief refresher on the case study method. Applications of case study
research, 3-20.
Zadeh, F. A., & Sulaiman, A. B. (2010). Dynamic street environment. Local Environment,
15(5), 433-452. doi:10.1080/13549831003735403
Zainal, Z. (2007). Case study as a research method. Jurnal Kemanusiaan, 9.
248
Appendix A: Griffith University thesis preparation policy
1. Inclusion of papers within the thesis
Higher degree by research is a program of independent supervised study that produces
significant and original research outcomes, culminating in a thesis, exegesis or equivalent
(refer to Higher Degree by Research Thesis). Inclusion of papers within a thesis is not a
suitable thesis format for all research projects, for example: collaborative projects where there
may be several co-authors for each paper which may make it difficult for the examiner to
establish the independence of the candidates work; where primary data is not collected, or
results obtained, until late in the candidature; or where the research will not produce a logical
sequence of papers that are able to be presented as an integrated whole.
Candidates should also take into account whether this thesis format is an accepted practice
within their discipline and likely to be received well by the thesis examiners (refer also to the
examination requirements below). Candidates are required to consult with their supervisor(s)
early in their candidature to determine if this thesis format is appropriate.
It is expected that candidates will identify as part of the confirmation of candidature milestone
if their thesis is to be prepared in this format. Candidates should consult their Group specific
guidelines in addition to the requirements detailed below. Candidates are also encouraged to
attend the workshop: ‘Inclusion of papers within a thesis’ offered by the Griffith Graduate
Research School. Refer also to the Griffith University Code for the Responsible Conduct of
Research, specifically the sections pertaining to publication ethics and the dissemination of
research findings, and authorship.
Status of papers
A thesis may include papers that have been submitted, accepted for publication, or published.
Some disciplines may specify a variation to the status of papers requirement, refer to your
Group specific guidelines.
Type of papers
For the purpose of this requirement, papers are defined as a journal article, conference
publication, book or book chapter. Papers which have been rejected by a publisher must not
be included unless they have been substantially rewritten to address the reviewers’ comments,
or have since been accepted for publication. Some disciplines may specify a variation to the
type of papers requirement, refer to your Group specific guidelines.
249
Number of papers
A thesis may be entirely or partly comprised of papers. A paper maybe included as a single
chapter if the paper contributes to the argument of the thesis, or several papers may form the
core chapters of the theses where they present a cohesive argument. Where a thesis is entirely
comprised of papers, there is no minimum requirement for the number of papers that must be
included (except as noted below) and is a matter of professional judgment for the supervisor
and the candidate. Overall, the material presented for examination needs to reflect the
research thesis standard required for the award of the degree. For example, PhD candidates,
on the basis of a program of independent supervised study, must produce a thesis that makes a
significant and original contribution to knowledge and understanding in the relevant field of
study. This remains a matter of professional judgment for the supervisor and the candidate.
Where a thesis is entirely comprised of papers, some disciplines may specify a minimum
number of papers to be included, refer to your Group specific guidelines.
Authorship
The candidate should normally be principal author (that is, responsible for the intellectual
content and the majority of writing of the text) of any work included in the body of the thesis.
Where a paper has been co-authored, the candidate is required to have made a substantial
contribution to the intellectual content and writing of the text. Co-authored work in which the
candidate was a minor author can only be used and referenced in the way common to any
other research publication cited in the thesis. A signature from the corresponding author is
required in order to include co- authored material in the body of the thesis, refer to the
declarations section below.
For co-authored papers, the attribution of authorship must be in accordance with the Griffith
University Code for the Responsible Conduct of Research, which specifies that ‘authorship
must be based on substantial contributions in one or more of:
 Conception and design of the research project
 Analysis and interpretation of research data
 Drafting or making significant parts of the creative or scholarly work or critically
revising it so as to contribute significantly to the final output’.
Some disciplines may specify a variation to the authorship requirement, refer to your
Group specific guidelines.
250
Quality of papers
Candidates should endeavour to publish their research in high quality peer reviewed
publications. Papers to be included in the body of the thesis should be published (or submitted
for publication) in reputable outlets that are held in higher regard in the relevant field of
research. Candidates should consult their supervisor(s) for advice on suitable publications
specific to their research discipline. Some disciplines may specify quality standards that must
be met for papers to be included, refer to your Group specific guidelines.
The library also provides support and advice to candidates on choosing a journal. Candidates
are advised to note in particular advice in order to avoid ‘predatory’ publishers.
 Research Guide: Higher degree research candidates - Get Published
 Publishing in Open Access journals
Copyright
As copyright in an article is normally assigned to a publisher, the publisher must give
permission to reproduce the work in the thesis and put a digital copy on the institutional
repository. Information on how to seek permission is available at: Copyright and Articles in
thesis. If permission cannot be obtained, students may still include the publication in the body
of the thesis, however following examination the relevant chapter(s) will be redacted from the
digital copy to be held by the Griffith University Library so that the copyright material is not
made publicly available in the institutional repository. Students are required to advise the
copyright status of each publication included in the thesis via a declaration to be inserted in
the thesis, as detailed below.
Students requiring further advice regarding copyright issues can contact the Information
Policy Officer on (07) 3735 5695 or [email protected].
Group and discipline requirements
Some Groups or Elements may specify additional requirements for including papers within a
thesis, refer below:
 Arts, Education and Law
 Griffith Business School (PDF 214k)
 Griffith Health
 Griffith Sciences
251
2. Format of thesis
General
Consult the thesis preparation and formatting guidelines for general information about the
requirements for formatting the thesis. Some disciplines may specify a variation to the thesis
format requirements below, refer to your Group specific guidelines.
Structure of thesis and linking chapters
The structure of the thesis will vary depending on whether the thesis is partly or entirely
comprised of papers. Whatever the format, the thesis must present as a coherent and
integrated body of work in which the research objectives, relationship to other scholarly work,
methodology and strategies employed, and the results obtained are identified, analysed and
evaluated.
In general every thesis should include a general introduction and general discussion to frame
the internal chapters. The introduction should outline the scope of the research covered by the
thesis and include an explanation of the organisation and structure of the thesis. The general
discussion should draw together the main findings of the thesis and establish the significance
of the work as a whole, and should not just restate the discussion points of each paper.
It is important that candidates explicitly argue the coherence of the work and establish links
between the various papers/chapters throughout the thesis.
Linking text should be added to introduce each new paper or chapter, with a foreword which
introduces the research and establishes its links to previous papers/chapters.
Depending on the content of the paper(s) and nature of research, a research methods chapter
may also be necessary to ensure that any work that is not included in the paper(s), but is
integral to the research, is appropriately covered.
Any data omitted from a paper may also be included as an addendum to the thesis.
For further information on the thesis structure, refer to the following examples of acceptable
ways to format the thesis when including papers.
Format of papers
The papers may be rewritten for the thesis according to the general formatting guidelines; or
they can be inserted in their published format, subject to copyright approval as detailed above.
Pagination
Candidates may repaginate the papers to be consistent with the thesis. However, this is at the
discretion of the candidate.
252
Declarations
All theses that include papers must include declarations which specify the publication status
of the paper(s), your contribution to the paper(s), and the copyright status of the paper(s). The
declarations must be signed by the corresponding author (where applicable). If you are the
sole author, this still needs to be specified. The declaration will need to be inserted at the
beginning of the thesis, and for any co-authored papers, additional declarations will need to be
inserted at the beginning of each relevant chapter. You may wish to consult the declaration
requirements for inclusion of papers diagram to ensure that you insert the correct
declaration(s) within the thesis. Please note that completion of the declaration(s) does not
negate the need to comply with any other.
University requirement relating to co-authored works as outlined in the Griffith University
Code for the Responsible Conduct of Research.
3. Examination requirements
Assessment by examiners
Candidates who wish to include papers within their thesis, and who have determined that this
thesis format is appropriate to the research project, should also consider whether this thesis
format will be well received by the thesis examiners. The inclusion of papers may negatively
impact on the thesis upon assessment by the examiners where: the thesis format is not a
common or accepted practice within the candidates discipline area; where the inclusion of co-
authored papers makes it difficult for the examiner to establish the independence and
originality of the candidates work; where the thesis does not present to the examiner as an
integrated whole; or where there is too much repetition in the thesis which an examiner may
view as a weakness.
Theses that include papers are subject to the same examination criteria as theses submitted in
the traditional format. It should also be noted that the inclusion of published papers within the
thesis does not prevent an examiner from requesting amendments to that material.
Candidates should discuss the suitability of this thesis format for examination with their
supervisor(s).
Nomination of examiners
It is the responsibility of the principal supervisor to nominate thesis examiners, and the
process dictates that the principal supervisor must approach all nominees to determine their
willingness to examine. Where a candidate’s thesis is formatted to include papers, the
253
principal supervisor must also ensure that the examiners are familiar with and/or accepting of,
this thesis format.
Upon dispatch of a candidate’s thesis to an examiner, the examiner will be reminded that the
thesis has been formatted to include papers. The examiner will also be provided with the
relevant information and regulations regarding this thesis format.
254
Appendix B: Recruiting email to experts
GU ref no: 2016/840
INVITATION LETTER
Dear PBT Member,
We are writing to invite you to participate in a panel of experts to undertake a survey and to
participate in a workshop to assist a PhD study at the Cities Research Institute at Griffith
University.
The PhD project is entitled, “Developing and Testing a Pedestrian Level of Service (PLOS)
Audit Instrument for Footpaths in Commercial Streets". The key aim of this research is to
develop and test a more optimal PLOS audit tool. Pedestrian Level of Service (PLOS) is
commonly used for street assessment, and can be defined as “an overall measure of walking
condition on a route, path and facility”. We are seeking to develop this tool using three inputs:
i) previous research approaches; ii) expert viewpoints; and iii) pedestrian perceptions. We are
seeking your help for the second item.
We are adapting a slimmed-down version of the Delphi technique to work with experts, but in
a way that will reduce the burden on you. We are asking participants to help identify and
select the most important factors affecting PLOS on commercial streets. We will provide
participants with a link to an online survey that includes an extensive list of factors used in
previous studies. Participants will be asked to complete the survey providing their views on
the key factors, and on how we should approach measurement of PLOS.
The survey results will then inform a walking workshop (“walkshop”) as the second stage,
where a revised list of factors from the expert panel will be provided to participants, who will
then be asked to visit commercial streets to consider the revised list in the field. Upon
returning to the workshop venue, the group will then attempt to obtain a more meaningful
group consensus about what should be used to measure PLOS.
We would also like to inform you that as a thank you for participating there will be a prize
draw for an iPad or Samsung tablet at the end of the walkshop.
255
Based on the outcome of the Delphi study, a questionnaire will be developed. The
questionnaire will be used to gain pedestrian perceptions by intercept survey. One of the aims
of this research is to contribute to improving the walking environment for pedestrian. These
surveys will help us to determine whether the current walking environments meet the needs of
pedestrians in terms of quality of walking environment. Therefore, by participating in this
study, you will help other planners, urban designers, transport engineers and roadway
engineers to make informed decisions when designing or choosing an appropriate cross
section for any given commercial street, which meets the basic needs of pedestrians to
experience safety, comfort and attractiveness while walking, thereby achieving a high quality
walking environment. Your participation is also an important contribution in providing a
complete picture of the walking for all stakeholders and designing an inviting footpaths used
by pedestrians.
Note that the link below will take you to the online survey. Please, click it and complete
the survey and then click submit after completing.
https://prodsurvey.rcs.griffith.edu.au/prodls200/index.php/289371?lang=en
Yours sincerely,
Research team members
Nowar Abdalmajeed
(PhD Student)
Cities Research Centre / Griffith School of Environment
Contact Phone: +61 4 70656593

Contact Email: [email protected]
A/Prof Matthew Burke Dr Jenny Cui

(Principal Supervisor) (Associate Supervisor)
Cities Research Centre/ Griffith University Griffith School of Environment
Contact Phone: +61 7 373 57106 Contact Phone: +61 7 3735 7043
Contact Email: [email protected] Contact Email: [email protected]
256
Appendix C: Expert online survey questionnaire for selecting the
most important factors affecting PLOS
257
258
259
260
261
262
263
264
265
266
267
268
269
270
Appendix D: The poster of the advertising of a walking workshop
on the PedBikeTrans Facebook page
271
Appendix E: Information sheet of experts
GU ref no: 2016/840
Developing and Testing Pedestrian Level of Service (PLOS) Audit Instrument for Footpaths
in Commercial Streets
Information Sheet – Walking Workshop (Walkshop)
Research Team Members

Nowar Raad
(PhD Student)
Contact Phone:+61 4 70656593

Why is the research being conducted?

This research is a part of the requirements for obtaining the Ph.D. at Griffith University. This PhD
project is entitled “Developing and Testing a Pedestrian Level of Service (PLOS) Audit
Instrument for Footpaths in Commercial Streets". The key aim of this research is to develop and
test a more optimal PLOS audit tool. PLOS is commonly for street assessment and can be defined
as “an overall measure of walking condition on a route, path and facility”. We are seeking to
develop this tool using three inputs: i) previous research approaches, ii) expert viewpoints; and,
iii) pedestrian perceptions. We are seeking your help for the second item.
This research also seeks to achieve the following secondary aims:
 To identify the various factors affecting PLOS at footpaths in commercial areas.

 To select the appropriate variables which mainly influence PLOS in footpath conditions
using a systematic method.
 To rate the factors impacting walking conditions in terms of the degree of importance in
influencing PLOS based on the panel of experts and pedestrians' perceptions.
 To develop and test the inter-rater reliability of the new audit tool, which will be used to
measure PLOS.
 To assess the validity of a sub-set of a new audit tool items using intra-rater reliability.
 To define the limits of different PLOS (A-F) for footpaths in commercial areas under
affected factors.
 To develop appropriate methodology that leads to a new PLOS model, and to compare the
result of PLOS using this and other models, to determine if the new audit tool is an
adequate tool with which to assess walking conditions.
272
What you will be asked to do?
We are adapting a slimmed-down version of the Delphi technique to work with experts, but in a
way that will reduce the burden on participants of experts. We are asking participants to help
identify and select the most important factors affecting PLOS on commercial streets. We will
provide participants with a link to an online survey that includes an extensive list of factors used
in previous studies. Participants will be asked to complete the survey providing their views on the
key factors and how we should approach measurement of PLOS.
The survey results will then inform a walking workshop (a walkshop) where a revised list from
the expert panel will be provided, and then participants will be asked to go out onto commercial
streets and consider the revised list in the field. On return to the workshop venue the group will
then try to obtain a more meaningful group consensus about what should be used to measure
PLOS.
The walking workshop (walkshop) will take approximately 1-2 hours to obtain your views.
Participants of focus group workshop must keep the privacy and confidentiality for answers
gained from other participants. The walking workshop (walkshop) will be audio recorded.
Who are we seeking?

The target participants are individuals who have experience in pedestrian issues such as urban
planners, designers and transportation engineers in Brisbane. PedBikeTrans members are asked
for help to achieve this task.
The expected benefits of the research

This research will provide a new method for calculating PLOS for planning purposes. Also, this
research will seek to create software for PLOS calculation in order to get accurate results in a
speedy manner. The result of this research will be useful for urban planners, designers and
transportation engineers and will help them to know level of service on footpaths in commercial
areas. Then, they can re-design the streets based on pedestrian perceptions to meet the needs of
users. Once the PhD program will approach completion, significant attention will be given in the
final period in attempting to package this research, and the final product, for take up by key
agencies including consultants and local government agencies responsible for commercial streets.
Risks to you
There are no foreseeable risks from this research to you.
Prize Draw
As an expression of our thanks for your participation, there will be a prize draw for an iPad or
Samsung tablet at the end of the walkshop.
Your confidentiality
The records of this study will be kept private. All of your responses will be checked to remove
individual identification before analysis. Your responses will not be used for purposes other than
the researcher’s academic program including academic publications such as journal articles,
conference presentations, social media and PhD thesis. No information will be provided that could
identify you. As required by Griffith University, all audio recordings will be erased after
transcription. However, the research data (walking workshop transcripts and analysis) will be
retained in a locked cabinet and/or a password protected electronic file at Griffith University for a
period of five years before being destroyed.
273
Your participation is voluntary
Your participation is voluntary. We do not expect you to have in-depth knowledge of all questions
included thus you are not obliged to answer every question unless you wish to do so. There is no
penalty or loss of benefits for not participating and your decision to participate or not will in no
way impact upon your relationship with Griffith University or its staff. You are free to withdraw
from the study at any time without comment or penalty.
Questions / further information

If you have any additional queries, concerns or complaints about this research, please do not
hesitate to contact any of the research team members.
The ethical conduct of this research

Griffith University conducts research in accordance with the National Statement on Ethical
Conduct in Human Research. If potential participants have any concerns or complaints about the
ethical conduct of the research project they should contact the Manager, Research Ethics, at
Griffith University Human Research Ethics Committee on +61 7 3735 4375 or research-
[email protected].
Feedback to you
All participants will have the opportunity to know the results of this research. The results and
findings of the research will be communicated through academic and other publications and
conference presentations. Participants in workshops will be notified about the publications related
to this research. In addition, in the final period, significant attention will be given to an attempt to
package this research and the final product for take up by key agencies, including consultants and
local governments agencies responsible for commercial streets.
Privacy Statement
The conduct of this research involves the collection, access and/or use of your identified personal
information. The information collected is confidential and will not be disclosed to third parties
without your consent, except to meet government, legal or other regulatory authority
requirements. A de-identified copy of this data may be used for other research purposes.
However, your anonymity will at all times be safeguarded. For further information, consult the
University’s Privacy Plan at
http://www.griffith.edu.au/about-griffith/plans-publications/griffith-university-privacy-plan
Or telephone (07) 3735 4375.
274
Appendix F: Consent forms for experts
GU ref no: 2016/840
Consent Form - Walking Workshop (Walkshop)
Research team Members

Nowar Raad
(PhD Student)

By signing below, I confirm that I have read and understood the information package and
in particular have noted that:
 I understand that my involvement in this research will include participation in focus

group workshop with other participants (approximately 20-30);
 I have had any questions answered to my satisfaction;
 I understand the risks involved;
 I understand that there will be no direct benefit to me from my participation in this
research;
 I understand that my participation in this research is voluntary;
 I understand that if I have any additional questions I can contact the research team;
 I understand that I am free to withdraw at any time, without explanation or penalty;
 I understand that I can contact the Manager, Research Ethics, at Griffith University
Human Research Ethics Committee on 3735 4375 (or research-
[email protected]) if I have any concerns about the ethical conduct of the
project; and
 I agree to participate in the project.
Name
Signature
Date
275
Appendix G: Walkshop PrizeDraw
GU ref no: 2016/840
Prize Draw - Walking Workshop (Walkshop)
Research team Members

Nowar Raad
(PhD Student)

PRIZE DRAW FOR THE RESEARCH PROJECT:
As an expression of our thanks for your participation, there will be a prize draw for an iPad or
Samsung tablet at the end of the walking workshop (walkshop).
Terms and Conditions of Entry:
1. When you enter the competition, you accept these terms and conditions of entry.
2. Employees of Griffith University ("the University") and their immediate families are
ineligible to enter.
3. Entry into the competition is by physically participating in a walking workshop
(walkshop).
4. The first randomly drawn entry will receive an iPad or Samsung tablet.
5. The decision of the University is final and no correspondence will be entered into.
6. The prize is not transferable and cannot be redeemed for cash. The prize is not
refundable.
276
7. The winner releases the University from any and all causes of action, losses, liability,
damage, expense (including legal expenses) cost or charge suffered, sustained or in
any way incurred by the winner as a result of any loss or damage to any physical
property of the winner, or any injury to or death of any person arising out of, or related
to or in any way connected with the University or the prize.
8. Any winner drawn for the prize who is unable to fulfil all of these terms and
conditions will forfeit the prize and another winner will be drawn.
9. The competition opens to entries at 26th September 2017, 4.00 pm and the
competition closes at 15th November 2017, 7.00 pm. The competition will be drawn at
the end of a walking workshop on 15th November 2017, 7.00 pm. Participants must be
present at the draw to win.
10. The winner will be notified in person directly after completing a walking workshop
and the prize will be delivered to winner immediately after the draw.
277
Appendix H: Walking course sheet
Walking Workshop
“Walkshop”
GU Ref No: 2016/840
Selecting the Most Important Factors Affecting Pedestrian Level of Service (PLOS)
This walking workshop is aimed at selecting the most important factors that affect pedestrian
LOS on footpaths in commercial areas. Pedestrian level of service (PLOS) is a common
measure for street assessment and is defined as “an overall measure of walking condition on a
route, path and facility”. The walking workshop forms part of a review of factors affecting
PLOS which will be used in my PhD project for developing and testing an audit instrument
for measuring the level of service for pedestrians on footpaths in commercial streets.
By participating in this study, you will help other planners, urban designers, transport
engineers and roadway engineers to make informed decisions when designing or choosing an
appropriate cross section for any given commercial street, which meets the basic needs of
pedestrians to experience safety, comfort and attractiveness while walking, thereby achieving
a high quality walking environment. Your participation is also an important contribution in
providing a complete picture of the environment for all stakeholders when designing inviting
footpaths used by pedestrians.
Our task today
We ask you to walk along footpaths within the Brisbane central business district. This
includes parts of Adelaide, George, Elizabeth, Albert, Charlotte, Edward, Queen and Creek
Streets.
At key locations we will ask you to make a personal judgement as to how YOU rate the PLOS
on that street for each segment. The street segments are provided on the attached map.
You are provided a 6-point (LOS “A” to LOS “F”) scale to evaluate how safe / comfortable
you feel it is for pedestrians as you travel each segment.
Level “A” is considered the most safe / comfortable (or least hazardous).
Level “F” is considered the least safe / comfortable (or most hazardous).
Note: please use the attached PLOS rating sheet to record your evaluation of the PLOS for
each segment identified on the map.
278
Whilst walking we ask you to use your imagination to consider which factors have more
effect and make more of a contribution to PLOS than others.
After completing the walk please return to the workshop room. You will be provided a sheet
listing the set of factors affecting PLOS that emerged from the literature review and our recent
survey of experts – that many of you contributed to. On this sheet are the group median scores
and Content Validity Ratios (CVRs) derived from the survey for each factor. You will be
asked to identify and make changes to each factor based on:
Your perceptions obtained during the walk; and
The group median score and CVR for each factor.
A note on privacy.
This survey is anonymous.
The record of your survey responses does not contain any identifying information
about you.
Itinerary details
Participant will travel in a counter-clockwise direction as follows:
Walk along Adelaide Street then:
Turn left at George Street
Turn left at Elizabeth Street
Turn right at Albert Street
Turn left at Charlotte Street
Turn left at Edward Street
Turn right at Queen Street
Turn left into Creek Street
Turn right into Adelaide Street
Return to 313 Adelaide Street Place
279
Itinerary Map
280
PLOS RATING SHEET FOR EACH SEGMENT
Please, record your rating on a 6-point scale from A to F (Best to Worst)
Footpath Segment No. 1
From 313 Adelaide St to 293 Adelaide St, Brisbane City QLD 4000, Australia (Walk 120 m, 1
min)
I rate the PLOS on this segment as …………..
min)
min)
From 91 Adelaide St to 239 George St, Brisbane City QLD 4000, Australia (Walk 230 m, 3 min)
From 239 George St to 36 Elizabeth St, Brisbane City QLD 4000, Australia (Walk 230 m, 3 min)
From 36 Elizabeth St to 111 Elizabeth St, Brisbane City QLD 4000, Australia (Walk 210 m, 3
min)
From 111 Elizabeth St to 121 Albert St, Brisbane City QLD 4000, Australia (Walk 120 m, 2 min)
From 121 Albert St to 166 Charlotte St, Brisbane City QLD 4000, Australia (Walk 230 m, 3 min)
From 166 Charlotte St to 201 Edward St, Brisbane City QLD 4000, Australia (Walk 230 m, 3
min)
281
From 201 Edward St to 345 Queen St, Brisbane City QLD 4000, Australia (Walk 230 m, 4 min)
From 345 Queen St via Creek St to 313 Adelaide St, Brisbane City QLD 4000, Australia (Walk
250 m, 4 min)
When finished, please remember to return to the workshop room to complete the activities.
If you leave early and do not plan to return to the workshop room please inform the research
team. Call or send an SMS with your name to Matthew Burke on 0402 346 464.
282
Appendix I: Workshop sheet
Walking Workshop
“Walkshop”
GU Ref No: 2016/840

The results of the survey
Important note:
Factors were rated from 1 to 5, where "1" indicates the factor is "not at all important" or
redundant i.e., “not be included" to "5" which indicates it is "extremely important".
*Content validity ratio (CVR): follows the equation "(n - N/2)/ (N/2)", where n is the number
of experts indicating “important” rating of a particular factor and N is the total number of
experts.
**Decision: According to the Lawshe measure of quantifying consensus and the validity of
judgments and because the panel of PedBikeTrans experts who responded to the 5-point scale
included 37 persons, therefore, the factors with a content validity ratio (CVR) less than 0.31,
a mean less than 3.5, and a median less than 3.5 should be eliminated.
*** New rating: This part is allocated if you wish to change your rating or give a new rating
on the same scale (from"1" which indicates the factor is "not at all important" or redundant
"i.e., not be included" to "5" which indicates it is "extremely important").
The revised factors sheet

*Content ***
Factor Variable Mean Median Validity **Decision New rating
Ratio (optional)
Geometric
Characteristics
Buffer width 3.50 4.00 0.667
Crossing opportunities 4.25 4.00 0.944
Quality of footpath surface 4.03 4.00 0.944
Footpath slope & grade 3.61 4.00 0.722
Adequate footpath width 4.53 5.00 0.944
On-street parking 2.17 2.00 -0.389 Removed
Pedestrian-traffic
3.75 4.00 0.889
Separation
Ramp 3.64 4.00 0.722
Shoulder width 2.80 3.00 0.278 Removed
Height clearance 3.19 3.00 0.444
Obstructions and
Interruptions
283
Footpath connectivity 4.64 5.00 1
Footpath continuity 4.64 5.00 1
Frequency of lateral ways 3.30 3.00 0.778 Removed
Ease of movement 4.25 4.00 1
Permanent obstructions 4.19 4.00 0.944
Temporary obstructions 3.44 3.00 0.667 Removed
Obstruction width 3.69 4.00 0.778
*Content ***
Factor Variable Mean Median Validity **Decision New rating
Ratio (optional)
Aesthetics and
Amenities
Awnings 3.55 4.00 0.778
Benches 3.25 3.00 0.556 Removed
Bollards 3.00 3.00 0.389 Removed
Building articulation 3.03 3.00 0.556 Removed
Business presence 3.25 3.00 0.722 Removed
Crossing facility type 4.05 4.00 0.944
Activation of side facades 3.69 4.00 0.833
Facilities for the disabled 3.97 4.00 0.833
Drinking fountain 3.16 3.00 0.611 Removed
Enclosure 2.83 3.00 0.444 Removed
Fire hydrant 2.00 2.00 -0.444 Removed
Footpath cleanliness 3.64 4.00 0.667
Footpath maintenance 3.97 4.00 0.944
Footpath surveillance 3.50 4.00 0.5
Garbage facilities 3.03 3.00 0.5 Removed
Guidance signs 3.75 4.00 0.833
Lighting 4.08 4.00 0.889
Recreational facilities 2.78 3.00 0.222 Removed
Shade 4.33 5.00 0.889
Toilet nearby 3.00 3.00 0.389 Removed
3.92 4.00 0.778
Facilities
Trees and plants 4.08 4.00 0.889
Wayfinding guidance 4.19 4.00 0.944
Pedestrian
Traffic
Pedestrian free manoeuvre 3.91 4.00 0.778
Pedestrian conflict points 4.19 4.00 0.889
Pedestrian speed 3.50 4.00 0.722
Pedestrian traffic signals 3.66 4.00 0.722
Pedestrian volume 3.94 4.00 0.833
Adjacent Traffic
Bicycle speed 3.44 4.00 0.667
Bicycle volume 3.53 4.00 0.667
Number of traffic lanes 3.22 3.00 0.5 Removed
Vehicle conflict points 4.03 4.00 0.944
Vehicle lane width 3.30 3.00 0.611 Removed
Vehicle speed 4.16 4.50 0.889
Vehicle volume 4.05 4.00 0.889
284
Appendix J: Information sheet of pedestrian perceptions survey
GU ref no: 2016/840
Developing and Testing Pedestrian Level of Service (PLOS)

Audit Instrument for Footpaths in Commercial Streets
INFORMATION SHEET- Pedestrian Perceptions Survey
Research Team Members

Nowar Abdalmajeed
(PhD Student)
Contact Phone:+61 4 70656593

A/Prof Matthew Burke Dr Tooran Alizadeh

(Principal Supervisor) (Associate Supervisor )
Why is the research being conducted?
This research is a part of the requirements for obtaining the Ph.D. at Griffith University. The key aim
of this research is to create a new audit tool for use in identifying and measuring the factors affecting
pedestrian on footpaths along commercial streets and then to evaluate Pedestrian Level of Service
(PLOS) depending on pedestrians’ perceptions.
The final result of this project will help planners, urban designers, transportation engineers and
roadway engineers to determine whether the current walking environments meet the needs of
pedestrians in terms of quality of walking environment, to provide a complete picture of the walking
environment and to design “inviting” footpaths used by you. Therefore, the final product is useful for
stakeholders previously mentioned in making informed decisions when designing or choosing the
appropriate cross section for any given roadway – a cross section that meets pedestrians’ basic need to
feel safe and comfortable while walking, achieving a high quality walking environment and improving
the walking environment for pedestrians.
What you will be asked to do

In this part of the research your participation in completing a questionnaire is requested. The
questionnaire includes three different topics, namely, physical comfort regarding footpath
facilities, personal safety regarding traffic safety and mobility regarding ease of walking. The
questions are about your characteristics (age, education, occupation, trip purpose and
frequency on footpaths); your perceptions (include the set of questions to obtain your
285
perceptions about important factors affecting pedestrian level of service on footpaths and to
get their rating) and your opinions regarding overall footpath performance. Most of the
questions are multiple choice closed-end questions, and a few questions are open-ended
format.
286
Who are we seeking?
The target participants are pedestrians who use the footpaths for different purposes at different
times of day. Potential participants will be of different ages over 15 years across gender,
cultures and occupations in Brisbane and the Gold Coast.
The expected benefits of the research

This research will provide a new method for calculating PLOS for planning purposes. Also, this
research will seek to create specific software for PLOS calculation in order to get accurate results
in a speedy manner. The results of this research will be useful for urban planners, designers and
transportation engineers and will help them to know level of service on footpaths in commercial
areas. Then, they can re-design the streets based on pedestrians’ perceptions to meet the needs of
users. Once the PhD program approaches completion, significant attention will be given in the
final period in attempting to package this research and the final product for take up by key
agencies including consultants and local government agencies responsible for commercial streets.
Risks to you
There are no foreseeable risks from this research to you.
Your confidentiality
The record of this study will be kept private. All your responses will be checked to remove
individual identification before analysis. The survey will collect some personal information
such as your age, gender, education, occupation, original country and trip purpose and
frequency on footpaths. All data collected will be merged with data from other participants
and have personal identification removed. Your responses will only be used for the
researcher’s academic program including academic publications such as journal articles,
conference presentations, social media and PhD thesis. The data collected from you will be
treated confidentially. As required by Griffith University, all the research data (questionnaire
transcripts and analysis) will be retained in a locked cabinet and/or a password protected
electronic file at Griffith University for a period of five years before being destroyed.
Your participation is voluntary
Your participation is voluntary. We do not expect you to have in depth knowledge of all questions
included, thus you are not obliged to answer every question unless you wish to do so. There are no
penalties or loss of benefits for not participating and your decision to participate or not will in no
way impact upon your relationship with Griffith University or its staff. You are free to withdraw
from the study at any time without comment or penalty.
Mechanism for distribution and return

Pedestrian perceptions will be examined by using questionnaire tools. The questionnaires will
be distributed to pedestrians who walk on the footpaths between intersections. The
questionnaires will be distributed by the researcher with volunteers (if needed) and they will
explain how to fill the questionnaire using a tablet pc. Instructions and explanations of PLOS
will be provided in the first few pages of the questionnaire in order to clarify what will be
expected.
Questions / further information

If you have any additional queries, concerns or complaints about this research, please do not
hesitate to contact any of the research team members.
287
The ethical conduct of this research
Griffith University conducts research in accordance with the National Statement on Ethical
Conduct in Human Research. If potential participants have any concerns or complaints about
the ethical conduct of the research project they should contact the Manager, Research Ethics ,
at Griffith University Human Research Ethics Committee on +61 7 3735 4375 or research-
[email protected].
Feedback to you
You can request a plain language summary of results by sending an email to the researcher:
[email protected].
Privacy Statement
The conduct of this research involves the collection, access and/or use of your identified
personal information. The information collected is confidential and will not be disclosed to
third parties without your consent, except to meet government, legal or other regulatory
authority requirements. A de-identified copy of this data may be used for other research
purposes. However, your anonymity will at all times be safeguarded. For further information
consult the University’s Privacy Plan at http://www.griffith.edu.au/about-griffith/plans-
publications/griffith-university-privacy-plan Or telephone (07) 3735 4375.
Expressing consent – anonymous or de-identified information

Please retain this sheet for your later reference. Completing the survey/questionnaire will be
deemed as having consented to participate in this research.
288
Appendix K: Pedestrian Prize Draw
GU ref no: 2016/840
PRIZE DRAW - Pedestrian Perception Survey
Research team members

Nowar Abdalmajeed
(PhD Student)

PRIZE DRAW FOR THE RESEARCH PROJECT:
As an expression of our thanks for your participation, there will be a prize draw after
completing the Pedestrian Level of Service Perception Survey.
Terms and Conditions of Entry:
1. The prize draw is being run by Griffith University to encourage participation in the intercept
survey of PhD project "Developing and Testing Pedestrian Level of Service (PLOS) Audit
Instrument for Footpaths in Commercial Streets"
2. By electing to participate, you accept these terms and conditions as governing the prize
draw. Instructions on how to enter the prize draw and details advertising the survey form
part of the conditions. Any personal information you provide to us in the course of entering
the prize draw will be dealt with by us in accordance with our privacy policy (published at:
http://www.griffith.edu.au/aboutgriffith/governance/plans-publications/griffith-university-
privacy-plan).
3. One prize will be awarded in the prize draw, the prize being a Samsung tablet worth $180.
Should the advertised prize become unavailable as a result of circumstances beyond our
control, we are free (at our sole discretion) to substitute a cash prize equivalent to the value
of the prize advertised.
4. Entry is free. Entry is open for each person selected for conducting the survey.
5. You may not enter the prize draw if you are an employee of ours or an immediate family
member of an employee of ours or otherwise associated with the competition.
289
6. To enter the prize draw, you must:
(a) Complete and submit the survey on a site using an IPad tool.
(b) Provide a valid postal address/email/phone number.
7. You may only submit one entry in the prize draw.
8. All survey and other materials provided by you becomes our property. No responsibility is
taken for late, lost or misdirected surveys or entries.
9. Following the closing date, the prize winners will be selected randomly from valid entries
received. Each entry can only be drawn once.
10. Subject to system malfunction, the prize will be a Samsung tablet. If the systems
supporting the draw are not functioning as they should when the draw is due, the draw will
be held as soon as possible once the systems become functional again. Prize winners do not
need to be present at the time of the draw.
11. The prize winner’s names will not be published.
12. The relevant prize will be sent to each prize winner at the postal address captured within
the Consent form.
13. The majority of prizes will be mailed within two weeks of the draw.
14. If any prize winner cannot be contacted within three (3) months of the draw, then that
person’s right to the prize is forfeited and the prize will be treated as an unclaimed prize.
15. Only one redraw of unclaimed prizes will take place, and other existing prizes are not
affected. The redraw prize winner(s) will be randomly selected from the remaining valid
entries and notified within two (2) weeks of the redraw. If the redraw prize winner(s) cannot
be contacted within three (3) months of the redraw, then we may determine that the relevant
prize(s) will not be awarded.
16. Prizes cannot be substituted for another prize at the election of the prize-winner.
17. We are not liable for any loss, expense, damage or injury sustained by any entrant in
connection with this prize draw, the prize or redemption of the prize, except for any liability
which cannot be excluded by law (in which case, that liability is limited to the minimum
allowable by law).
18. We may suspend the promotion if we determine that the integrity or administration of the
promotion has been adversely affected due to circumstances beyond our control. We may
disqualify any individual who tampers with the entry process.
290
Appendix L: Pedestrian perception questionnaire
291
292
293
294
295
296
297
298
299
Appendix M: Pedestrian level of service audit tool (PLOSAT): Paper form
Auditor name -----------------------------, Auditor ID --------------, Date: -----------------------, Time of day: -----------------, Weather: ----------------
City: ----------------------, Suburb: -----------------------, Street name: ----------------------, Footpath side: ----------------, Segment ID: ---------------
Point System Method

Category Factor
0 1 2 3
Clear footpath width Less than 1.35m 1.35m- 2.25m 2.25m- 3.15 More than 3.15m
No crossings provided (hard to Insufficient crossings but in Insufficient crossings but in Sufficient crossings in logical
cross) illogical locations logical locations locations

Quality of footpath surface Poor quality Moderate quality Good quality Excellent quality
Footpath slope & grade Extremely steep slope/gradient Steep slope/gradient Moderate slope/gradient Flat or gentle slope/gradient
Buffer width Less than 1.35m 1.35m- 1.65m 1.65m- 2.15m More than 2.15m
Sufficient separation in
Existing somewhere in
Pedestrian-traffic No separation and the segment Existing in the reasonable reasonable locations at
unreasonable locations in the
separation is needed locations but not sufficient adequate frequency or the path
segment
not needed
Insufficient kerb ramps and Sufficient kerb ramps but does
Sufficient kerb ramps, meets all
Kerb ramp No kerb ramps provided does not meet all required not meet all required
Clearance over footpath Less than 2.0m Equal to 2.0m 2.0m- 2- 2.5m Greater than 2.5m
1 or fewer connections (no 2 - 4 connections (low 5-7 connections (moderate More than 7 connections (high
Footpath connectivity
Obstructions and
connectivity) connectivity) connectivity) connectivity)

interruptions
Footpath continuity Less than 50% completed 50%- 75% completed 75- 90% completed More than 90 % completed
Pedestrian crowding > Pedestrian crowding > Pedestrian crowding > 8ppmm
Pedestrian crowding ≤ 8ppmm,
26ppmm, movement is 100% 17ppmm≤ 26ppmm, movement ≤ 17ppmm, movement < 18% ≥
Easy of movement movement < 9% restricted so
restricted so very restricted ≤ 26% ≥ 18% restricted so 9% restricted so normal
free movement
movement restricted movement movement
Many obstructions (more than Some obstruction (5 - 11 per Few obstructions (1 - 4 per
Permanent obstructions No obstructions (0 per m)
11 per segment) segment) segment)
300
Point System Method
Category Factor
0 1 2 3
Few awnings (less than 50% is Sufficient awnings (50-75% is Fully sufficient awnings (more
Awnings No awnings
covered) covered) than 75% is covered)
Sufficient trees/ plants along Fully sufficient trees/ plants
Few trees/ plants along the
the audited segment (more than along the audited segment
No or very few trees/ plants audited segment (25-75% is
Trees and plants 75% of the segment is covered) (more than 75% of the segment
(less than 25% is covered) covered) but does not meet all
but does not meet all required is covered) and meets all
specifications required specifications
Insufficient pedestrian scale Minimum sufficient pedestrian Full sufficient pedestrian scale
Shade No shade
shade scale shade shade
Type of crossing facility Poor Moderate Good Excellent
Activation of side facades Non-active frontage Poor active frontage Good active frontage Excellent active frontage
Sufficient TGSI and meets all

Insufficient TGSI and does not Sufficient TGSI but does not
Facilities for the disabled No TGSI provided required specifications or it is
meet all required specifications meet all required specifications
not needed
Footpath maintenance Poor Moderate Good Excellent
Footpath cleanliness Poor Fair Good Excellent
Fully sufficient surveillance
Footpath surveillance No surveillance means Few surveillance means Sufficient surveillance means
means
Existing somewhere but does Existing sufficient frequency Full sufficient frequency and
Garbage facilities No rubbish bin not meet all required but does not meet all required meets all required
Existing somewhere but Sufficient frequency but does Full sufficient frequency and
Pedestrian signs No signage insufficient and does not meet not meet all required meets all required
all required specifications specifications specifications
Insufficient pedestrian scale Sufficient pedestrian scale Full sufficient road oriented
Lighting No artificial lighting
lighting lighting lighting
Insufficient bus stops and does Sufficient bus stops but does
Public transportation Sufficient bus stops and meets
No bus stops provided not meet all required not meet all required
facilities all required specifications
Existing somewhere but does Sufficient frequency but does Full sufficient frequency and
Wayfinding guidance No wayfinding facility not meet all required not meet all required meets all required
301
Point System Method
Category Factor
0 1 2 3
Pedestrian evasive
Massive evasive manoeuvres Moderate evasive manoeuvres Normal evasive manoeuvres No evasive manoeuvres
Pedestrian
manoeuvres
traffic
Pedestrian conflict points Poor Moderate Reasonable Perfect

Pedestrian speed Very restricted walking speed Restricted walking speed Normal walking speed Free walking speed
Pedestrian volume Very crowded path Moderately crowded path Uncrowded path Very uncrowded path

Adjacent
Vehicle conflict points points more than 16 per km points 5- 1 per km as

traffic
points 16- 6 as moderate points = 0 as perfect

(8/500m) as poor reasonable
Vehicle speed More than 60 km/h 60 km/h- 40 km/h 40 km/h- 30 km/h Less than 30 km/h
Vehicle volume High traffic volume Moderate traffic volume Normal traffic volume Low traffic volume
302
Appendix N: Pedestrian level of service audit tool (PLOSAT):
Google form
303
304
305
306
307
308
309
310
Appendix O: The weight assessment form (WAF)
Realistic Factor Weighted
Categoris
Group Category Factor weight assessment assessment
ed PLOS
(RW) score (FAS) score (WAS)
Footpath width 3.2
Crossing opportunities 3.2
Characteristics
Quality of footpath surface 3.2

Geometric
Footpath slope & grade 3.1

24%
Buffer area and width 3.2

Pedestrian-traffic separation 3.1
Ramp 3.1
Factors related to the pedestrian environment
Clearance over footpath 1.9

Footpath connectivity 3.1
interruptions
Obstructions
Footpath continuity 3.2

12.6%
and
Easy of movement 3.2

Permanent obstructions 3.2
Awnings 2.9
Trees and plants 3.2
Shade 3.2
Type of crossing facility 3.0
Activation of side facades 2.5

Facilities for the disabled 3.2
Footpath maintenance
42.5%
3.2
Footpath cleanliness 3.2
Footpath surveillance 2.3
Garbage facilities 3.2
Pedestrian signs 3.2
Lighting 3.2
3.2
facilities
Wayfinding guidance 3.1
Evasive movements 2.7
Factors related to the
Pedestrian
traffic system and
11.6%
traffic
Pedestrian conflict points 3.2

Pedestrian speed 2.7
user
Pedestrian volume 3.1

Vehicle conflict points 3.2
Adjacen
t traffic
9.3%
Vehicle speed 3.1

Vehicle volume 3.1
Overall weighted assessment score (OWAS)
Results Overall PLOS
PLOS grade
311
Appendix P: Final report form of PLOS
1. Background
Auditor name: _________________ Auditor ID: _________________ City:
____________________ Suburb: _______________________ Street name:
_______________Footpath side: ____________ Audit purpose: _________________
Weather conditions _____________ Audit date: __________
2. Results
Total PLOS for entire targeted footpath can be calculated by summation of the results of
PLOS values of all audited segments for the targeted footpath divided on a number these
segments as follows:
Where; overall PLOS value is the final result of audited footpath, PLOS value is for each
audited segment (i) on targeted footpath, n is the total number of audited segments
The overall rating of the Pedestrian Level of Service for the targeted footpath is________
This rating of PLOS value is based on the PLOS grades from A to F that were developed for a
range of the values based on a point system method as shown in Table A and B below.
3. Recommendations and remedial actions

You can list a summary of all the required information related to the Remedial Actions Plan
and Recommendations in Appendix 4
Table A Definition of PLOS grade and score, conditions, acceptability, priority of remedial
action and remedial action of factors affecting PLOS
PLOS
PLOS Score Conditions Acceptability Remedial action
Grade
A ≥ 250 Excellent Very acceptable Very limited
B 200-249 Good Acceptable Limited
C 150-199 Regular Fairly acceptable Medium
D 100-149 Basic Unacceptable High
E 50-99 Poor Highly unacceptable Unlimited
F < 50 Appalling Totally invalid Rebuilding
Table B PLOS grade scale of each category of the factors affecting PLOS
Categorical PLOS grade
PLOS factors’ category
A B C D E F
Geometric characteristics > 60 49-60 37-48 25-36 13-24 ≤ 12
Obstructions and interruptions > 32 27-32 21-26 15-20 9-14 ≤8
Aesthetics and amenities > 107 87-107 66-86 45-65 24-44 ≤ 23
Pedestrian traffic > 29 24-29 18-23 12-17 6-11 ≤5
Adjacent traffic > 23 19-23 14-18 9-13 4-8 ≤3
312
Appendix Q: Summary of current situation, recommendations and remedial action plan required
for audited footpath
Priority for
Condition Acceptabili Remedial
Factor remedial Audit Recommendations
s ty action
action
Increase the clear footpath width to be 3.15 or more
Footpath width (ONLY which will be sufficient for the current pedestrian
Good Acceptable Low Limited
example) volume and appropriate for various types of users such
as wheelchair users.
Quality of footpath surface
Footpath slope & grade
Buffer area and width
Pedestrian-traffic
separation
Ramp
Clearance over footpath
Footpath connectivity
Footpath continuity
Easy of movement
Permanent obstructions
Awnings
Trees and plants
Shade
313
Type of crossing facility
Activation of side facades
Facilities for the disabled
Footpath maintenance
Footpath cleanliness
Footpath surveillance
Garbage facilities
Guidance signs
Lighting
facilities
Wayfinding guidance
Evasive movements
Pedestrian conflict points
Pedestrian speed
Pedestrian volume
Vehicle conflict points
Vehicle speed
Vehicle volume
314
Appendix R: Quick guidelines for using pedestrian level of service audit tool for footpaths in
commercial streets in the field work
Measurement Assessment
Factors Help Questions Specification or steps Selection Criteria method
method
Measure Steps:
 Measure a clear footpath width for each 20m length along the audited
segments.
 Calculate the average width for the audited segment (using a tape measure).  If the total clear footpath width+
 Consider the pedestrian volume for each audited area, using dependable frontage width is less than 1.35m,
statistics if available or a field survey if needed.
 What is the average clear footpath width select 0.
 Compare the obtained results of the average width with the standard width  If the total clear footpath width+  Measured from
along the audited footpath segment?
provided by the Australian road guidelines based on the location, purpose, plans or maps.
Should be measured in the field frontage width is 1.35-2.25, select 1.
and expected demand or pedestrian volume of the audited area.  Measured using
Footpath width  Is the present footpath width clear for  If the total clear footpath width + Objective
Notice: GIS or
the pedestrian volumes and type of frontage width is 2.25-3.15, select 2.
 The minimum and desired width for the pedestrian through-route zones is  Measured during
users? Should be checked with the
2.4m (or higher based on demand). In commercial or shopping environments  If the total clear footpath width+ site visit
standard frontage width is more than 3.15,
the maximum pedestrian flow is 80 pedestrians/ minute.
 In busy alfresco dining areas such as the central city area, a minimum select 3.
through-route of 3m - 4m should be provided, reduced to 2.5m in areas with
less pedestrian traffic.
 The minimum and desired width for a property frontage zone is 0.75m which
is being added to the average of clear footpath width.
 If there are no crossings provided
 Are sufficient crossing opportunities
(hard to cross) in the segment, select
provided along the audited footpath?
0.
Should be yes.
 If there are insufficient crossings but
 Are crossing opportunities located at
Crossing in illogical locations in the segment, Measured during
opportunities
logical locations along the audited - select 1. site visit
Subjective
footpath? Should be yes.
 If there are insufficient crossings but
 Do pedestrians correctly use the crossing
in logical locations, select 2.
opportunity provided along the audited
 If there are sufficient crossings in
logical locations, select 3.
Specification:
 Design transition shall be 0 mm. Construction tolerances shall be as follows:
 Is the walking surface of the footpath 1. 0-3 mm vertical  If the segment condition seems of
appropriately designed? Should be yes. 2. 0-5 mm, provided the edges have a bevelled or rounded edge to poor quality, select 0.
 Are the pedestrian crossovers smooth? reduce the likelihood of tripping. Any gaps, lips, joins or changes in  If the segment condition seems of
Quality of Should be yes. height cannot exceed 5mm as these can create trip hazards. moderate quality, select 1. Measured during
Subjective
footpath surface  Is the texture of the surface traversable by  A poor path is one with a massive roughness and non-slip resistance which  If the segment condition seems of site visit
people who use a wheelchair and those contains a lot of uneven levels, many tripping hazards, major uplifts and lips good quality, select 2.
with an ambulant or sensory disability? with paving using poor quality material.  If the segment condition seems of
Should be yes.  A moderate path is one with a little smoothness and low slip resistance excellent quality, select 3.
which contains few uneven levels, few tripping hazards, few uplifts and lips
with paving with moderate quality materials.
315
 A good path is one with a smooth and slip resistant surface which contains
very few uneven levels, very few tripping hazards, very few uplifts and lips
with paving with good quality materials.
 An excellent path is one which is perfectly smooth and which has high slip
resistance. It does not contain uneven levels, tripping hazards, uplifts and lips
with paving with excellent quality materials.
 If the audited segment has an
extremely steep slope/gradient, select
0 (an extremely steep slope is one
Measure Steps:
with an extremely sharp or extremely
 Check the slope condition of the footpath segment paralleled to the roadway.
rapid incline).
 Check the gradient condition of the footpath segment perpendicular to the
 If the audited segment has a steep
roadway. Often this becomes a problem at driveways, especially for
slope/gradient, select 1 (a steep slope
pedestrians in wheelchairs who must maintain a straight path.
is one with a sharp or rapid incline). Subjective
Footpath slope &  Assess the slope and gradient based on the conditions of the majority of the Measured during
-  If the audited segment has a moderate and
grade segment. For example, if the beginning of a segment has a moderate slope site visit
slope/gradient, select 2 (a moderate objective
but the majority is flat, score it as flat.
condition is one with a medium
 Assess the slope and gradient based on the prevalent conditions if the
incline).
segment is evenly divided between two gradients. For example, if the
 If the audited segment has a flat or
segment has both a slight and a moderate slope, then score it as moderate.
gentle slope/gradient, select 3 (a flat
condition is one with a gradual
incline).
Measure Steps:
 Measure a buffer width for each 20m length along the audited segments.
 If the total buffer width+ kerb zone
 Calculate the average width for the audited segment.
width is less than 1.35 or there is no  Measure from
 Compare the obtained results of the average width with the standard width
buffer zone, select 0. plans or maps.
provided by the Australian road guidelines based on the location, purpose,
Buffer area and and expected demand or pedestrian volume for the audited area.  Measure using
width - Notice:
width is 1.35- 1.65, select 1.
GIS or
Objective
 If the total buffer width + kerb zone
 The desired width for the buffer zones is 2m for commercial and shopping  Measure during
width is 1.65- 2.15, select 2.
environments or city streets. site visit
 The desired width for the kerb zone is 0.15m which is being added to the width is more than 2.15m, select 3.
average of the buffer width.
 Is separation urgently needed at the  If there is no separation and it is

audited segment? needed in a segment, select 0.
 Is any kind of separation provided  If there is separation somewhere in
between motorists and pedestrians? unreasonable locations in the
 Does the pathway provide enough segment, select 1.
Pedestrian- separation from high-speed, high-volume  If there is separation in reasonable
traffic separation traffic? - locations but they are insufficient,
 Is there adequate separation between select 2.
vehicular traffic and pedestrians at the  If there is a sufficient separation in
expected conflict points between reasonable locations at adequate
pedestrians, cyclists, and motorists? frequency, or a footpath is not
needed, select 3.
316
 Are kerb ramps provided for each
crossing area? Should be yes. Specification:
 Are ramps aligned with the crossing area,  There should not be there any level differences between a kerb ramp on a
so that pedestrians (especially those with footpath and an adjacent roadway.
low vision) are correctly oriented when  A smooth change in the level between the footpath and road pavement
entering the crossing area from the ramp? should be provided.  If no kerb ramps are provided in the
Should be yes.  A minimum footpath width of 1.33m beyond the top of the ramp should be segment, select 0.
 Is a level area of sufficient size at the top provided.  If there are insufficient kerb ramps
of the ramp to allow wheelchair users (or  A gradient of 1:8 should not be exceeded, however, a more desirable and they do not meet all required
people using walking aids) to wait or gradient is 1:10 or less. specifications, select 1.
Subjective
manoeuvre without the risk of rolling into  Total length of 1.5m should not be exceeded.  If there are sufficient kerb ramps but Measure during site
Kerb ramp and
the roadway? Should be yes.  The kerb ramp should be flush with the adjacent footpath, street surfaces, they do not meet all required visit.
objective
 Are kerb ramps flush with the street and gutter (i.e., no lip or gap). specifications, select 2.
surfaces (i.e., no lip or gap)? Should be  The ramp should be located perpendicular to the direction of travel.  If there are sufficient kerb ramps
yes.  The ramp should have a warning style tactile ground surface indicator which meet all required
 Are gaps, cracks and discontinuities (TGSI) at each crossing point. specifications, select 3.
between kerb ramps and street surfaces  The ramp should be at least 0.9m wide.
avoided at each crossing? Should be yes.  The ramp should have a firm and stable surface.
 Is water pooling on pedestrian ramps,  The ramp should be located within the corner area of the footpath, aligned
making walking uncomfortable and with the street wall, and set at a straight angle to the street alignment.
unsafe, especially if pools of water stand
for prolonged periods? Should be no.
 If the height clearance is less than
2.0m, select 0.
 If the height clearance is = 2.0m,
Specification: select 1.
Clearance over Measure during site
- An absolute minimum height clearance of 2.0 m above the accessible path of  If the height clearance is 2.0m- 2.5m, Objective
footpath visit.
travel is required and the ideal height clearance equal to 2.5m is preferred select 2.
 If the height clearance is greater than
2.5m, select 3.

Measure Steps:
 If the total connectivity score is 1 or
 Stop at the beginning of the targeted segment and look backward, right and
fewer connection, select 0 as no
 Is the footpath connected to other left. Then check and identify how many footpaths or crosswalks are
connectivity.  Measure from
pedestrian facilities to form a pedestrian connected.
 If the total connectivity score is 2-4
network? Should be yes  Stop at the middle of the targeted segment and look right and left. Then plans or maps.
connection, select 1 as low
Footpath  Is the footpath provided connected to key check and identify how many footpaths or crosswalks are connected.  Measure using
connectivity. Objective
connectivity destinations along the route (like public  Walk to the end of the targeted segment and look forward, right and left. GIS or
 If the total connectivity score is 5- 7
transport stops, schools, shops, parks, Then check and identify how many footpaths or crosswalks are connected.
connections, select 2 as moderate  Measure during
community centres, offices)? Should be  Stop signs at the end of the segment can be treated as crosswalks. site visit
connectivity.
yes  Count the number of connections to other footpaths and crosswalks and
 If the total connectivity score is more
calculate the total connectivity score by summing the number of connections
than 7, select 3 as high connectivity.
of all parts.
 Does a footpath have any breaks within  If segment condition seems ≤ 50%
Footpath the segment? Should be no continuous/complete (e.g., Measure during site
continuity  Does a footpath have any gap or missing - incomplete), select 1. visit.
Subjective
pavement pieces within the segment?  If segment condition seems > 50% ≤
317
Should be yes 75% continuous/complete, select 2.
 Is the footpath continuous; does it have  If segment condition seems > 75 ≤
any missing sections? Should be no 99% continuous/complete, select 3.
 If segment condition seems 100%
continuous/complete, select 4.
 If the pedestrian crowding is >
26ppmm, movement is 100%
 Is there ample space for pedestrians to
restricted, select 0 as very restricted
walk on their chosen route? Should be
movement.
yes.
Measure Steps:  If the pedestrian crowding is >
 Does the current path provide enough
 Examine the results of pedestrian volume factor in each segment. 17ppmm≤ 26ppmm, movement is ≤
space for some choice in routes taken?
 Use subjective assessment to determine the restriction indicator in personage 26% ≥ 18% restricted, select 1 as
Ease of Should be yes. Measure during site
for the audited segment. restricted movement. Objective
movement  Do pedestrians have very little personal visit.
 Make a decision taking into consideration previous conditions, based on the  If the pedestrian crowding is >
space and restricted movement? Should
next criteria. 8ppmm ≤ 17ppmm, movement is<
be no.
18% ≥ 9% restricted, select 2 as
 Do pedestrians experience extreme
normal movement.
difficulty bypassing slower pedestrians or
 If the pedestrian crowding is ≤
moving in reverse flows? Should be no.
8ppmm, movement is < 9%
restricted, select 3 as free movement.
Measure Steps:
 Identify the existence of items that are considered to be permanent
 If the total permanent obstructions
obstructions on the audited footpath segment.
score is more than 11 per segment,
 At the location of each obstruction, measure the minimum effective width of
select 0.
the path.
 Evaluate whether the effective width is still sufficient for the pedestrian
Permanent score is 5-11 per segment, select 1. Measure during site
obstructions - volumes and types of users on the path, according to the recommended
 If the total permanent obstructions visit.
Objective
footpath width identified in the footpath width factor.
score is 1-4 per segment, select 2.
 If the effective width is insufficient, count this item as being a permanent
obstruction; otherwise do not consider this item to be an obstruction.
score is = 0 (no obstructions), select
 Add the number of items that are considered as permanent obstructions on
3.
the audited footpath segment and save the score to calculate the total score of
permanent obstructions for the audited segment.
 If there are no awnings along the
audited segment, select 0.
 If there are few awnings along the
audited segment (less than 50% is
covered), select 1.
 If there are sufficient awnings along Measure during site
Awnings - - the audited segment (50-75% is visit.
Objective
covered), select 2.
 If there are fully sufficient awnings
along the audited segment (more than
75% is covered), select 3.
Subjective
Specification:  If there are no/very few trees/plants Measure during site
Trees and plants -  Trees and plants do not intrude into the clear continuous path of travel. along the audited segment (less than visit
and
objective
318
Ideally, they should be placed in the street furniture zone within the footpath 25% is covered), select 0.
cross-section in order to keep the pedestrian through route free (preferably  If there are few trees/plants along the
1.5m or wider) from obstruction. audited segment (between 25% and
 Trees and plants should be placed in a location that does not obscure 75% of the segment is covered) but
pedestrian or driver visibility when installed or when mature. they not meet all required
 Trees should be five metres tall at installation. specifications, select 1.
 Trees and plants should be placed within four metres of the edge of the  If there are sufficient trees/ plants
nearest traffic lane. along the audited segment (more than
75% of the segment is covered) but
they not meet all required
 If there are fully sufficient trees/
plants along the audited segment
(more than 75% of the segment is
covered) and they meet all required
 If there is no shade in the area, select
0.
 If there is insufficient pedestrian scale
 Do trees or built structures provide shade
shade (less than 25% of the area is
along the footpath? Should be yes
shaded), select 1.
 Do trees or built structures provide shade Measure during site
Shade -  If there is minimum sufficient Subjective
at resting places and places with street visit.
pedestrian scale shade (25% to75% of
furniture? Should be yes
the area is shaded), select 2.
 If there is full sufficient pedestrian
scale shade (more than 75% of the
area is shaded), select 3.
 If the crossing facilities are unmarked
crosswalks without crossing aids,
Specification:
select 0 as poor.
 Sufficient for the type of road, size of road, volume of traffic, pedestrian
 If the crossing facilities type are
volumes and type of pedestrians.
unmarked crosswalks with
 Note whether ramps are clearly visible to pedestrians, motorists, and other
insufficient crossing aids, select 1 as
Type of crossing road users. Measure during site
moderate. Subjective
facility  Kerb ramps should be provided with the required crossing aids, including visit.
 If the crossing facilities are marked
pedestrian refuge islands, curb extensions, pedestrian crossing warning
crosswalks with insufficient crossing
signs, pedestrian signals, pedestrian countdown signals and ‘yield to
aids, select 2 as good.
pedestrian’ signs.
 If the crossing facilities are marked
crosswalks with sufficient crossing
aids, select 3 as excellent.
 Are most buildings situated close to the Specification:  If the frontage of buildings along the
street with entrances accessible from the audited segment is non active, select
 Several doors and windows with few blank walls.
footpath (without having to cross a 0.
 The frontage of buildings gives a vertical rhythm to the street scene. Subjective
Activation of side parking lot)? Should be yes.  If the frontage of buildings along the Measure during site
 Facades with projections such as bays and porches provide a welcoming and
facades  Do buildings have multiple windows that audited segment is poorly active, visit.
feeling. objective
allow people to see into and out of the select 1.
 Lively internal uses are visible from the outside, or spill onto the street.
building? Should be yes.  If the frontage of buildings along the
 Do buildings with several windows close audited segment is fairly active, select
319
to the street create a safer and more 2.
interesting environment for pedestrians?  If the frontage of buildings along the
Should be yes. audited segment is excellently active,
select 3.
Specification:
 Should only be used where necessary. Check whether they are needed on the
audited segment; then check the next specifications and record the score for
this factor. Otherwise, leave this factor and record it as 4.
 Should be colour contrasted against the surrounding pavement.
 Should be used across a path of travel to identify the point at which a  If there are no TGSI provided in the
segment, select 0.
pedestrian needs to turn to access a particular target such as a mid-block
pedestrian crossing, bus stop or point of entry to a significant public facility.  If there are insufficient TGSI which
In these situations, TGSI should: do not meet all required
1. Be set perpendicular to the flow of pedestrian traffic. specifications, select 1.
Facilities for the Measure during site
- 2. Extend across the entire width of the walkway.  If there are sufficient TGSI but they Subjective
disabled visit.
do not meet all required
3. Be aligned to the left hand side of the adjoining pathway.
 Should be used as a dedicated guidance strip to identify a safe path of travel
 If there are sufficient TGSI which
through an otherwise complicated site, with few or no alternative cues, from
meet all required specifications or if
point A to point B. In these situations, TGSI should:
they are not needed, select 3.
1. Be 0.3m to 0.4m wide.
2. Be laid along the midline of the safe path of travel.
3. Allow for minimum clearance of at least 0.6m on both sides.
 If the segment condition seems less

Specification:
than 25% well maintained (a lot of
 Poor maintenance is with the existence of a lot of bumps, cracks, holes and
bumps, cracks, holes and weeds),
weeds growing in the surface or between the cracks of footpaths, abrupt
select 0 as poor maintenance.
changes in surface level, drainage problems, uplifts or lips in the pavement
 If the segment condition seems
surface.
between 25-50% well maintained
 Moderate maintenance is with the existence of some bumps, cracks, holes
(some bumps, cracks, holes and
and weeds growing in the surface or between the cracks of footpaths, abrupt
weeds), select 1 as moderate
Footpath changes in surface level, drainage problems, uplifts or lips in the pavement Measure during site
maintenance - surface.
maintenance.
visit.
Subjective
 If the segment condition seems
 Good maintenance is with the existence of very few bumps, cracks, holes and
between 50-90% well maintained
weeds growing in the surface or between the cracks of footpaths, abrupt
(very few bumps, cracks, holes and
changes in surface level, drainage problems, uplifts or lips in the pavement
weeds), select 2 as good maintenance.
surface.
 If the segment condition seems more
 Excellent maintenance is with the absence of bumps, cracks, holes and weeds
than 90% well maintained (no bumps,
growing in the surface or between the cracks of footpaths, abrupt changes in
cracks, holes and weeds), select 3 as
surface level, drainage problems, uplifts or lips in the pavement surface.
excellent maintenance.
 If there is a large amount (a very
noticeable amount) of litter, rubbish,
Footpath on the segment, select 0 as poor. Measure during site
cleanliness - -  If there is small amount of litter, visit.
Subjective
rubbish, etc., in the segment, select 1
as fair.
320
 If there is very little litter, rubbish,
etc., in the segment, select 2 as good.
 If there is no little litter, rubbish, etc.,
in the segment, select 3 as excellent.
 Are many pedestrians walking along the
 If there are no surveillance means
 Do people feel safe walking on this route
along the audited area, select 0.
section? Should be yes.
 If there are few surveillance means
 Is the path visible? Is there an opportunity
along the audited area, select 1.
for surveillance from adjacent land uses
Footpath  If there are sufficient surveillance Measure during site
surveillance
and buildings? Should be yes. - means along the audited area, select visit.
Subjective
 Are there enough people around to make
2.
people feel safe? Should be yes.
 If there are fully sufficient
 Are there enough police patrols or
surveillance means along the audited
security cameras along the footpath?
area, select 3.
Should be yes.
Specification:  If there are no rubbish bins in the

 Are they in areas where litter may be
 Rubbish bins should not intrude into the clear continuous path of travel. audited segment, select 0.
generated, such as bus stops, transport
Ideally they should be placed in the street furniture zone within the footpath  If there are bins somewhere but they
interchanges and fast food outlets? cross-section so that the pedestrian through route is kept free (preferably 1.5 do not meet all required
Should be yes. m or wider) from obstruction. specifications, select 1. Subjective
Garbage  Are they provided along the route  Bins should be approximately 0.4m-1m high.
Measure during site
 If there is sufficient frequency of bins and
facilities segment? Should be yes. visit.
 Bins should be placed within four metres of the edge of the nearest traffic but they not meet all required objective
 Are they in good condition, i.e., are they
lane. specifications, select 2.
well maintained (no damage or graffiti)?
 Bins should be accessible to all users.  If there are sufficient rubbish bins
Should be yes.
 Bins should be colour contrasted from background surfaces to ensure that meeting all required specifications,
they are visible to the vision impaired. select 3.
 Is guidance signage sufficient on the
path? Should be yes.
 Is important information clearly written
on signs? Should be yes.  If there is no signage in the area,
 Are they visible to pedestrians? Should select 0.
be yes. Specification:  If there are some signs but they are
 Are there sufficient warning signs on the  The signs do not intrude into the clear continuous path of travel (ideal insufficient for the purpose and do
path? Should be yes. locations). not meet all required specifications,
 Are marked and signed pedestrian  The signs are appropriately oriented.
select 1. Subjective
Measure during site
Pedestrian signs crossings clearly visible to pedestrians?  If there is sufficient frequency of and
 The signs are visible, legible and understandable. visit.
Should be yes. signs but they not meet all required objective
 The signs are placed above a pathway at a height of 2.5m.
 Is the type of pedestrian infrastructure specifications, select 2.
 The signs are regularly maintained and in good condition.
clearly marked? Should be yes.  If there is full sufficient frequency of

 Are pedestrian routes and crossings signs meeting all required
clearly visible to motorists, with specifications, select 3.
pedestrian warning signs and pavement 
markings? Should be yes.
 Is the condition of signage and markings
adequate (for example, well painted, non-
321
slippery material, visible during day and
night, not damaged or worn)? Should be
yes.
Measure Steps:
 At locations of potential conflict and high pedestrian usage, such as
intersections of paths and railway stations, the point horizontal and vertical  If there is no artificial lighting in the
illumination is a minimum of 20 lux (although this is not always practical to area, select 0.
achieve).  If there is insufficient pedestrian scale
Subjective
 Ideally the location of lamp posts should be located in the street furniture lighting, select 1. Measure during site
Lighting - zone within the footpath cross-section, in order to keep the pedestrian  If there is sufficient pedestrian scale visit.
and
objective
through route free from obstruction. This width can vary between 0.9m - lighting, select 2.
1.2m. When street furniture is located on the footpath then a clear path of  If there is fully sufficient road
travel of at least 1.2m (preferably 1.5m or wider) must be maintained along oriented lighting, select3.
the building or property line. This will allow a navigational path for those
with low vision, and allow a wheelchair to pass along the path.
 If there are no bus stops provided in
 Are public transport bus stops needed at Specification: the segment, select 0.
the audited segment? If yes, go to the
 Bus stops should have both directional and warning tactile ground surface  If there are insufficient bus stops
next questions. If no, score this factor 4
indicators (TGSI). which do not meet all required
points.
Public  Bus stops should have standard shelters if there is a clear need. specifications, select 1. Subjective
 Are sufficient public transport bus stops Measure during site
transportation  Bus stops should maintain a continuous path with a desired width of 1.54m  If there are sufficient bus stops but and
provided along the audited segment? visit
facilities (absolute minimum width required is 1.2m) between the bus stop and the they do not meet all required objective
Should be yes.
kerb. specifications, select 2.
 Are public transport bus stops located at
 Bus stops should be placed at footpath kerb extensions zones if there is  If there are sufficient bus stops and
logical locations along the audited
insufficient width. they meet all required specifications,
segment? Should be yes.
select 3.
Specification:
 If there is no wayfinding facility in
 Wayfinding facilities should not intrude into the clear continuous path of
the area, select 0.
 Is signage provided to guide and direct travel. Ideally, these facilities should be placed in the street furniture zone
within the footpath cross-section so as to keep the pedestrian through route  If there is a facility somewhere but it
pedestrians to the key destinations in the
free from obstruction. does not meet all required
area? Should be yes
 Is information, such as street names,  Wayfinding facilities should be appropriately oriented (should face the Subjective
Wayfinding
oncoming pedestrian flow).  If there is sufficient frequency of Measure during site
clearly written on signage and visible to and
guidance wayfinding facilities, but they do not visit.
pedestrians? Should be yes  Wayfinding facilities should be visible, legible and understandable (clear objective
meet all required specifications, select
visual and physical).
2.
 Wayfinding facilities should use clear font and simple language (use both
 If there is full sufficient frequency of
plain English and symbols that are easy to read).
wayfinding facilities and they meet
 Wayfinding facilities should be usable for all users, and be of appropriate
all required specifications, select 3.
height - in the range of 1.2m to 1.6m.
 Are many evasive manoeuvres  If there are many evasive manoeuvres
undertaken by pedestrians on the undertaken by pedestrians in the
footpath? Should be no segment, select 0 as massive evasive
Pedestrian  Are there a massive number of evasive manoeuvres.
Measure during site
evasive movements on the audited segment? -  If there are a moderate amount of Subjective
visit.
manoeuvres Should be no evasive manoeuvres undertaken by
 Are there many attempts by pedestrians to pedestrians on the segment, select 1
avoid other pedestrians coming from the as moderate evasive manoeuvres.
opposite direction, passing other  If there are few evasive manoeuvres
322
pedestrians in the same direction, or undertaken by pedestrians in the
walking abreast? Should be no. segment, select 2 as normal evasive
manoeuvres.
 If there are no evasive manoeuvres
undertaken by pedestrians in the
segment, select 3 as no evasive
manoeuvres.
 If there are many pedestrian conflict
 Are there frequent conflict points and, in Measure Steps:
points, select 0 as poor.
retail areas, do people begin to avoid  Estimate the potential points that are considered as pedestrian to pedestrian
 If there are a moderate number of
these areas? Should be no. conflict points or pedestrian to bicycle conflict points on the audited footpath
pedestrian conflict points, select 1 as
Pedestrian  Do the majority of people experience a segment. Measure during site
moderate. Subjective
conflict points conflict or closeness with other  Focus on pedestrians in retail areas and the areas of closeness with other visit.
 If there are few pedestrian conflict
pedestrians and does bi-directional pedestrians and bi-directional movement during observations of pedestrian
points, select 2 as reasonable.
movement become difficult? Should be conflict points.
 If there are no pedestrian conflict
no.
points, select 3 as perfect.
 If the path provides very little
personal space for pedestrians to walk
at the speed chosen, select 0 as very
 Is there plenty of space for pedestrians to
restricted walking speed.
walk at the speed chosen? Should be yes.
 If the path provides reasonable
 Does the current path provide enough
space for the walking speed taken?
at the speed chosen, select 1 as
Should be yes.
restricted walking speed. Measure during site
Pedestrian speed  Do pedestrians have very little personal -  If the path provides sufficient visit.
Subjective
space, and is their speed very restricted?
Do they experience extreme difficulty in
at the speed chosen, select 2 as
passing slower pedestrians? Should be
normal walking speed.
no.
 If the path provides plenty of
at the speed chosen, select 3 as free
walking speed.
Measure Steps:
26ppmm, select 0 as very crowded
 Stand in a location that does not disrupt normal activity on the audited
path.
footpaths.
 Draw an imaginary line perpendicular to the footpaths that are targeted for a
17ppmm≤ 26ppmm, select 1 as
pedestrian volume count.
Pedestrian moderately crowded path. Measure during site
-  Count the number of pedestrians who cross the imaginary line for a specific Objective
volume  If the pedestrian crowding is > visit.
period (15 minutes at each segment) using a manual method. It is advisable
8ppmm ≤ 17ppmm, select 2as an
to use tally counters to record this information, particularly in busy sites.
uncrowded path.
 Pedestrian crowding is measured in pedestrians per metre of clear footway
 If the pedestrian crowding is ≤
width per minute (ppmm) and is calculated using the following formula:
8ppmm, select 3 as a very uncrowded
Pedestrians per 15 minutes ÷ 15 ÷ clear footpath width in metres.
path.
 Are there many traffic-pedestrian conflict Measure Steps:  If the total traffic-pedestrian conflict
Subjective
Vehicle conflict points along the audited segment? Should  Identify the existence of points that are considered as traffic-pedestrian points are more than 8 per segment, Measure during site
and
points be no. conflict points in the audited footpath segment. This includes points in the select 0 as poor. visit.
objective
 How many traffic-pedestrian conflict following areas: signalized or unsignalized intersections, driveways, left or  If the total traffic-pedestrian conflict
323
points are there? Should be identified. right-turning traffic points and mid-block crossings. points are 4-8 per segment, select 1 as
 Do you feel safe when you walk in these  Add the number of points that are considered to be traffic-pedestrian conflict moderate.
conflict points? Should be yes to be points in audited footpath segments and save the score to calculate the total  If the total traffic-pedestrian conflict
excluded from count as a conflict point. score of traffic-pedestrian conflict points for the audited segment. points are 1-3 per segment, select 2 as
reasonable.
 If the total traffic-pedestrian conflict
points = 0 (no potential conflict
points), select 3 as perfect.
 If the traffic speed is more than 60
km/h, select 0.
Measure Steps:
 If the traffic speed is 41-60 km/h,
 Check the current allowed traffic speed in the audited area.
select 1.
 Consider this speed as the speed of the traffic flow, regardless of offending Measure during site
Vehicle speed -  If the traffic speed is 31-40 km/h, Objective
drivers. visit.
select 2.
 Compare this speed with the criteria determined in the next column and
 If the traffic speed is 30 km/h or
identify selection.
lesser, select 3.

 If there is a high traffic volume,
select 0 as poor conditions.
 If there is moderate traffic volume,
select 1 as moderate conditions. Measure during site
Vehicle volume - -  If there is normal traffic volume, visit.
Subjective
select 2 as good conditions.
 If there is low traffic volume, select 3
as excellent conditions.
324
325

Abdalmajeed, Nowar Raad - Final Thesis - Redacted

Uploaded by

Copyright:

Available Formats

Abdalmajeed, Nowar Raad - Final Thesis - Redacted

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Abdalmajeed, Nowar Raad - Final Thesis - Redacted

Uploaded by

Copyright:

Available Formats

Pedestrian Level of Service (PLOS) Measurement for

Footpaths in Commercial Streets

Griffith Research Online

Bachelor of Civil Engineering (University of Babylon)

The School of Environment and Science, Cities Research Institute

Submitted in fulfilment of the requirements of the degree of

Assoc Matthew Burke

Dr. Jenny Cui

I would also like to express my sincere acknowledgement to Dr Robert J. Schneider and Dr

Brisbane, August 2019

All papers included are co-authored

To be named as an author, a researcher must have made a substantial scholarly contribution to

 Conception and design of the research project.

 Analysis and interpretation of research data.

 Drafting or making significant parts of the creative or scholarly work or critically

Section 9.3 of the Griffith University Code (“Responsibilities of Researchers”), in accordance

Researchers are expected to:

 Accept or decline offers of authorship promptly in writing.

(Signed) _________________________________ (Date) 26 August, 2018

Name of Student and Corresponding Author: Nowar Raad

(Countersigned) ___________________________ (Date) 26 August, 2018

Supervisor: A/Prof Matthew Burke

(Signed) _________________________________ (Date) 26 August, 2018

Name of Student and Corresponding Author: Nowar Raad

(Countersigned) ___________________________ (Date) 26 August, 2018

Supervisor: A/Prof Matthew Burke

(Signed) _________________________________ (Date) 26 August, 2018

Name of Student and Corresponding Author: Nowar Raad

(Countersigned) ___________________________ (Date) 26 August, 2018

Supervisor: A/Prof Matthew Burke

1.2.1 Primary question 1

1.2.2 Primary question 2

1.2.3 Primary question 3

1.2.4 Primary question 4

1.3 Structure of thesis

2.2 The study of pedestrian behaviour

Figure 3 Hierarchy of walking needs model (source: Alfonzo, 2005)

2.2.1 Urban environment characteristics and physical activity

2.2.2 Impact of characteristics of physical environment on pedestrians' behaviour

2.2.3 Impact of urban design qualities on pedestrians’ behaviour

Table 1 Perceptual qualities (source: Ewing & Handy, 2009)

Mehta (2009) studied three neighbourhood commercial streets in Cambridge, Massachusetts,

2.3 Complete streets

Figure 4 Complete streets

2.5.1 Pedestrian level of service (PLOS) evaluation

2.5.2 Capacity based studies

2.5.3 Roadway characteristics based studies

2.5.4 Capacity roadway characteristics based studies (combination approaches)

Figure 6 The analysis methods of PLOS (Source: Asadi-Shekari et al. 2013)

Muraleetharan, Stepwise Geometric,

2.6.1 Previous attempts to develop and test audit tools

Figure 7 Conceptual framework of study by Ewing et al. (2006)

2.6.2 Conceptually framing previous audit tool studies

2.8 Research objective

3.2.1 Description of the study area

3.3 Research design

Task 1: A study of literature and context

Task 2: Data collection and analysis

Task 3: Development of PLOS models