Use of Business Modeling in Requirements

Definition Phase

V Naritoshi Tsugane V Takeo Asakura

(Manuscript received April 25, 2006)

For many enterprises, how to anticipate various changes in the business environment
and promptly respond to those changes are important issues. However, in many
cases, the IT systems that support the basis of business are extremely complicated
and difficult to respond to changes. To address these problems, Fujitsu recently
reviewed its SDAS system development methodology from the requirements defini-
tion phase by reconsidering its original concept. First, enterprise businesses and
related IT systems can be correctly visualized and commonly understood by everyone
by improving a few key modeling techniques for the requirements definition phase
(e.g., business process modeling and business data modeling) based on our previous
practical experience. Secondly, the total optimization of systems by taking new
approaches such as Service-Oriented Architecture (SOA) becomes possible by
providing the means to redesign complex IT systems into flexible, mutually separated
ones. This paper describes the concept of modeling and then examines business
process modeling and business data modeling. Lastly, this paper describes how these
business modeling techniques are used to realize SOA systems.

1. Introduction It is now clear, as suggested in the paper,

Is your company free from these problems? “Toward Realization of Service-Oriented Architec-
• Although a business flow was created for a ture (SOA),”1) elsewhere in this special edition,
previous development project, it cannot be that one of the most important issues for a
used because it has not been appropriately company is to quickly identify and respond to
maintained. changes in the business environment and that it
• Specifications cannot be defined because the is very important for IT systems to be able to flex-
people who are familiar with the business are ibly respond to changes by using SOA concepts.
too busy. As discussed in the SOA paper, the introduction
• Only one person knows about the system of a service bus alone is insufficient to make a
(i.e., the system is personalized). company capable of flexibly responding to chang-
• Furthermore, the people who know a lot es. After all, the key points are to understand
about the system are nearing retirement (the what is required to flexibly respond to changes
“baby-boomer generation” problem). and to design a system that can flexibly respond
• Although your company has decided how to to changes. Changes occur not in a system, but in
modify its business, little is known about a business. To become capable of quickly respond-
where and how the system must be modified ing to changes, it is necessary to visualize the
and what influences the modifications will fundamental structure and behavior of the busi-
have. ness in which the changes occur, quickly identify

316 FUJITSU Sci. Tech. J., 42,3,p.316-322(July 2006)

 N. Tsugane et al.: Use of Business Modeling in Requirements Definition Phase

the changes, and quickly adapt the system to the are used to realize SOA systems.
changes. These objectives cannot be achieved
without a business model that accurately maps 2. Concept of business modeling
the reality of the business. In addition, the mod- A business consists of two elements: proce-
el must be implemented in a system using a dures (processes) and information (data). To
language the system understands. In other words, reflect this in a system, business modeling re-
the system must closely represent the business quires the two perspectives of business process
world, and the relationship between the business modeling and business data modeling.
world and the computer world must be carefully Figure 2 outlines the concept of business
maintained. The SOA development project gave modeling. The business process model in the
Fujitsu an opportunity to recognize again the im- figure shows that order acceptance starts when
portance of upper processes, and we therefore an order arrives and that shipment starts after
started reviewing them. We believe the following order processing ends. An order has two sections
three points (Figure 1) are especially important (header and details), there are two types of order-
in the requirements definition phase: ing, and multiple products can be ordered together
• Reach a consensus among the stakeholders. (details section).
• Visualize and model the business, and run The figure shows that the essence of busi-
plan-do-check-act (PDCA) cycles to improve ness modeling is to clarify the business flow
the business (business process modeling). (procedures) and the structure of the information
• Construct the system so it closely reflects the to be handled.
real world (business data modeling) in an ar-
chitecture that is responsive to changes 3. Business process modeling
(service architecture modeling). Business process modeling is a technique
Although these three points are all impor- that visualizes business in the real world, mainly
tant in the requirements definition phase, this using a business flow chart. Sharing existing busi-
paper focuses on the last two business modeling ness knowledge and information about future
points. It outlines the concept of business model- changes in a business helps to improve business.
ing and then discusses business process modeling
and business data modeling. Lastly, this paper 3.1 Problems with conventional business
describes how these business modeling techniques process modeling
Although business process modeling depends
[Key activity] mainly on a business flow chart, notations are not
Management/business
standardized. In fact, when we collected business
Planning Consulting
flow chart examples, we found that their infor-
Business mation was inconsistent. In addition, because the
Business analysis items to be investigated and the steps to
improvement Business process modeling
modeling be taken were not defined, the modeling results
IT system tended to be personalized. This sometimes led
IT improvement Operation to failure in reaching a consensus with the
modeling Business data modeling
customer.
Service architecture modeling
Construction
Design Another issue is that a business problem that
is already known can be addressed at the time of
modeling, whereas an unknown problem cannot
Figure 1
Three points in upper processes. be addressed if the problem handling method is

FUJITSU Sci. Tech. J., 42,3,(July 2006) 317

N. Tsugane et al.: Use of Business Modeling in Requirements Definition Phase

Order reception: Order

When an order is received Customer: A
from a customer via facsimile (Ota-ku, Tokyo)
or e-mail, the order data is 1. Choc-bean: 10 What’s
registered in the system 2. Gum-ball: 5 that?
via a terminal. 3. Candy-ball: 20

Modeling can simplify things and eliminate ambiguity.

Business = information + procedure A business is defined by clarifying 1) the information

used for the business and 2) the procedures used to handle the information.
Information modeling = business data modeling Procedure modeling = business process modeling

Facsimile Order
Order There are two
ordering methods: Order Facsimile or
1 E-mail
facsimile and e-mail. acceptance e-mail
Each order
contains at
Order Shipment
least one line
details
for details.

Figure 2
Concept of business modeling.

limited to modeling and does not include contin- be processed.

ued monitoring using PDCA cycles. Unknown 4) Its modeling procedures are standardized.
problems are detected, for example, when there 5) It clarifies the points of modeling
is a sudden workload increase. (e.g., business activities are hierarchically
represented).
3.2 Features of SDAS-provided business 6) It provides a modeling tool (in-house use
process modeling only).
To handle these issues, we developed a new These features allow us to construct a busi-
business process modeling technique called ness process model that contains the information
Valuevision Modeling Methodology (VMM). required to reach a shared recognition and mini-
The features of this technique include the mize the level of personalization.
following: Figure 3 shows a model that was construct-
1) It uses Unified Modeling Language (UML) ed using this business process modeling technique.
so that notations can be standardized. This model was constructed by observing the
2) It uses UML profile extensions to improve predefined rules (e.g., the business must be
the readability of chart symbols, the appear- represented in a hierarchical structure and
ances of which are shown using extensions the sections in charge of the business must be
so business and information types can be in- clarified).
tuitively distinguished. We also note the importance of continued
3) It predefines the basic analysis items, for efforts to find potential business problems to
example, the time required to complete a pro- promote non-stop improvement. For this purpose,
cess and the number of backlogged slips to we provide a middleware package that combines

318 FUJITSU Sci. Tech. J., 42,3,(July 2006)

 N. Tsugane et al.: Use of Business Modeling in Requirements Definition Phase

Sales department Shipping department (warehouse) Supplier

Business flow
<<System-assisted>> Order information
Order form entry input screen

Purchase
<<System-assisted>> order form
Product stock inquiry Stock information Business data
confirmation screen
Relationship
<<System>> with
Work instructions business
Purchase order
input screen process
Stock insufficient

Business process <<System-assisted>>

Purchase order form entry

Stock sufficient <<Human action>>

Acceptance & Send-out
<<System-assisted>>
Business process flow Arrival confirmation

Shipment information Details (outline,

input screen activation conditions,
Shipment required time, cost,
processing amount,
etc.) are confirmed.
Terminal Delivery form Shipment form
entry 1

Figure 3
Image of business process modeling.

monitoring technology with modeling techniques.

This package can measure rating factors such as
the workload and time required for business
activities and visualize them on a business flow
chart (Figure 4). This function can visualize busi-
ness problems in a more specific representation,
allowing us to identify changes in the business
environment in a timely manner and respond to
them quickly. This package, therefore, provides
important monitoring technologies for ensuring Data collected
Running system through monitoring
PDCA cycles work as required for business A0010-01 Request Sept. 1
A0010-01 Reply Sept. 2
improvement. The area to which this monitoring Quotation Quotation
... A0010-02 Request Sept. 1
request reply
service service ....
technology belongs is called Business Process
Management (BPM).2) This area is attracting
much attention in the industry because it is Figure 4
Visualization with monitor tool.
required for enterprise internal control.

4. Business data modeling essential structure of a business, we can

Business data modeling visualizes entities in construct a change-responsive IT system in which
the real world, mainly by using a class diagram. essential and non-essential elements are clearly
By focusing on these entities to identify the distinguished.

FUJITSU Sci. Tech. J., 42,3,(July 2006) 319

N. Tsugane et al.: Use of Business Modeling in Requirements Definition Phase

4.1 Problems involved in conventional Because Kaname Analysis classifies all enti-
data modeling ties and all events as being essential or
Data modeling lies at the heart of the Data non-essential, the framework of an IT system can
Oriented Approach (DOA), which assumes that be well defined. This allows us to design a change-
information is the essence of a company. Compa- responsive IT system without spending much time
nies typically handle a huge amount of complex or human resources.
information. With DOA, a single attempt is made Figure 5 indicates the differences between
to faithfully map this complex world onto a mod- conventional DOA and Kaname Analysis.
el. Therefore, the IT systems constructed from Figure 6 shows the flow of Kaname Analysis.
these models tend to be as complex as the models. In this new business data modeling tech-
Another problem with this conventional tech- nique, we first represent the structure of the
nique is that it is time-consuming and requires a information to be handled for the business as a
large amount of human resources because it data model using a UML class diagram or similar
attempts to include all information (data) handled means. During the modeling, we focus only on
by the company in the model. the kaname entities at the core of the business.
Next, we analyze the relationships among the
4.2 Features of SDAS-provided business entities to confirm they have been correctly
data modeling extracted. Then, we represent the state changes
To overcome the problems mentioned in the that could occur for each entity as an event
previous section, we developed a new business sequence model. These two models represent the
data modeling technique that is based on the con- essence of the business in the simplest way.
cepts of kaname entities and events.3) After completing the above steps, we can find
One of the features of this technique is that the points where the relationships between
data modeling is divided into two stages: the busi- kaname events and entities are the loosest. This
ness analysis phase and the system design phase. enables us to identify the segmented units that
The business analysis phase focuses on Kaname compose an IT system and then make the system
Analysis to grasp the essence of the business. In capable of flexibly responding to changes. The
the system design phase, the business is functions provided by the segmented system are
modeled based on the results of Kaname Analysis called services in the SOA world (described
using conventional DOA with an emphasis on later).
comprehensiveness.
Kaname Analysis provides a simple model 5. Relationship between
that enables the essence of real-world situations business modeling and SOA
to be easily grasped. It does this by focusing on Fujitsu announced the establishment of its
real-world kaname entities and events, which are system of SOA technologies in July 2005 under
defined below. the slogan of “Build a change-resistant IT sys-
1) Kaname entities tem.” In Japan, the system provides an SOA
Kaname entities are things of interest (e.g., development technique called SDAS/Service
people, goods, and money) that are connected with Modeling. This technique combines business pro-
a company activity (business) and whose state can cess modeling and business data modeling
change. (described in other papers in this special issue)
2) Kaname events with the service modeling described in the SOA
Kaname events are activities that trigger a paper. Fujitsu provides industry-specific SOA
state transition of a kaname entity. development solutions in Japan that are based on

320 FUJITSU Sci. Tech. J., 42,3,(July 2006)

 N. Tsugane et al.: Use of Business Modeling in Requirements Definition Phase

Conventional technique Kaname Analysis

Real world Kaname Analysis Real world

(complex) (kaname entities/events)

DOA-handled data
Mapping the essence
Faithful mapping represented by kaname
using ER model, etc. entities/events
Kaname
world

Stock Purchase Order

Purchase order and confirmation order confirmation
stock confirmation
Payment
Service bus

Quote/order
Logistics Quote Logistics Payment

Loosely arranged IT system centered around kaname entity/event

Chaotic IT system with
unclear boundaries

Figure 5
Difference between conventional DOA and Kaname Analysis.

Trigger of entity
state transition

Order Product Customer

Kaname event

Order Product
Order
1 acceptance Service X Service Y Service M

....

1 Service bus
1
Shipment
Customer

.... Order Shipment Service N

Kaname entity Kaname event Service

Figure 6
Image of Kaname Analysis.

FUJITSU Sci. Tech. J., 42,3,(July 2006) 321

N. Tsugane et al.: Use of Business Modeling in Requirements Definition Phase

this SOA development technique and include 6. Conclusion

industry-specific know-how. More specifically, This paper described business modeling,
Fujitsu’s SOA development solutions include which is an important technique used in phases
Valuevision, a solution for industy and distribu- where requirements are defined through discus-
tion focusing on business and process sions with the customer.
improvement; FUTURITY, a solution for social The need for business modeling is increas-
infrastructures based on reviewed images of how ing year by year in many areas, for example, SOA
systems should be; and EVOLUO, a solution for and EA (total optimization planning). We will
financial institutions focusing on business tie-ups make further efforts to improve our techniques,
that satisfy needs related to industry reorganiza- educate and train modeling staff, and accumulate
tions and increasingly diversified distribution modeling results and make them available as
channels. reference information.
The purpose of SOA is to construct an IT
system architecture that allows IT systems to flex- References
ibly and quickly respond to changes in business. 1) I. Morita: Toward Realization of Service-Orient-
ed Architecture (SOA). Fujitsu Sci. Tech. J., 42,
Before a flexible IT system can be constructed, 3, p.306-315 (2006).
business modeling is required. To conduct busi- 2) Excellences of “BPM Infrastructure.” (in Japa-
nese), Fujitsu Journal, June 2004.
ness modeling appropriately, it is necessary to http://jp.fujitsu.com/about/journal/272/
recognize again the importance of upper process- newtechnology/02.html
3) Z. Nakamura: A Planning Method of Successful
es. SOA truly provides an opportunity for this Simple Work by Kaname Entity/Event Analysis.
recognition. (in Japanese), The Nikkan Kogyo Shimbun, Ltd.,
2003.

Naritoshi Tsugane, Fujitsu Ltd. Takeo Asakura, Fujitsu Ltd.

Mr. Tsugane graduated from Aichi Mr. Asakura received the B.S. and M.S.
Commercial High School, Japan in degrees in Informatics from Shizuoka
1983. He joined Fujitsu Ltd., Tokyo, University, Japan in 2002 and 2004,
Japan in 1983, where he has been respectively. He joined Fujitsu Ltd.,
engaged in development of technology Tokyo, Japan in 2004, where he has
for business requirement analysis and been engaged in development of
system requirement analysis. technology for business requirement
analysis and system requirement
analysis.

322 FUJITSU Sci. Tech. J., 42,3,(July 2006)