| Camera-ready
Conference Paper |
raffai@rs1.szif.hu |
| Camera-ready
Conference Paper |
raffai@rs1.szif.hu |
How to get
closer to the New Economy?
New aspects and paradigm in
Business Modeling and
in Application Development
Abstract
The last decade has brought a dynamic and
progressive growth both in the economy and informa-tion technology. It
is obvious that the continuous evolution forces the organizations to adapt
them-selves to the changing environments in order to survive. This paper
aims to point to the importance of using effective solutions, unified modeling
techniques and software tools for describing, analyz-ing and understanding
the business goals and processes. In the abstraction flow the analyzers
have to create different model-views of the business domain which makes
the high-level activities docu-mented and understandable. In my presentation
I emphasize the need and the importance of the right abstraction philosophy
and I show some unified graphical notation examples for the success-ful
co-operation and communication in the electronic economy. I point to the
necessity of a Unified Enterprise Modeling Language based on the UML standard.
Key Words: innovation, changing circumstances, role of ICT, way of adaptation, abstraction, mod-eling, enterprise modeling, effective forms of communication, visualizing, modeling language, meta-model, unifying, standards, UML, UEML
1. Introduction
Due to the building of the information society
–the rapidly growing telecommuting facilities, the continuous pressure
to adjust to the new conditions and the need to combat everyday threats–
the enterprises have great difficulties adapting themselves to the progressive
changing circumstances. In this situation we have to take into consideration
the components and the impact factors of the value chain [15]. The business
processes have to base on the most up-to-date information technol-ogy (IT),
and must strongly focus on the customers’ requirements. In the globalization
process we have to count with new forms of cooperation. As virtual enterprises
are coming into existence, the information technology plays key-role both
in business strategy and in business processes, better to say the IT is
built in the business domain and it becomes the most important driving
force for sur-viving (see Figure 1.). As the info-communication technologies
(ICT) create a “virtual business space”, it is easy to operate our information
systems (IS) regardless of physical-geographical loca-tion, and to communicate
on a synchronized concurrent way [6].
Figure 1. The Henderson-Venkatraman strategic
alignment model [10]
This evolution process forces the enterprises to use effective solutions and to fit themselves into the global e-economy. Both from theoretical and practical aspects, the organizations need to model the problem and the business domain and the corporate Information System, they must apply reusable components and computer-aided tools for planing and implementation in the development process. As it makes a high demand to find an appropriate philosophy for the abstraction of the real world, we have to review the IS’s definition and role having been used. The concept of the object-oriented paradigm (OO) and the Unified Modeling Language (UML) accepted in 1997 by Object Manage-ment Group (OMG) as a standard help the innovation projects to focus on fitting to the changing environments [17], [18]. Although the UML is capable to create and manage different model-views there is an urgent need to extend its capabilities
- for describing every aspects of the business domain,The UEML (Unified Enterprise Modeling Language) is now under developing, this is a prior sub-ject in several European projects with participation of the French, German, Spanish, Norwegian and Italian scientists. In my research work I deal with the extension of the UML, and I propose new elements emphasizing the business aspects to the existing model-views.
- for analyzing the existing system,
- for studying management’s efforts and
- for optimizing the business activities.
2. Towards the Information Society
The management responsible for the innovation
projects has to face to the new challenges, they have to fight against
both technological and economic power. To realize their business goals
in the new environment they have to change their philosophical outlook,
their way of thinking, they need to profess new paradigms, and they have
to find the best solutions. Information, by its influence on the society
and the economy, is the engine of the economic growth. The statistics on
creating, in-creasing, maintaining, querying and using databases show dynamic
progress, the on-line services of the E-economy suggest continuous evolution
and the whole economics is in rapidly changing in the stage of the knowledge
based information economy. The most progressive sector is that of info-communication
technology and ICT service.
The significance of information-jobs is permanently
growing, and the economy is ready to shift to the knowledge economy [9].
The European Union (EU) has a great responsibility to get Europe afloat
on the right way in development. EU needs to have an own strategy to offset
the backward-ness compared to USA and Japan in order to join to the global
new economy circulation. From this point of view, it is necessary to find
the appropriate reaction to the great challenge.
As the traditional application-development concept
causes deformed models of business domain and as it can not handle the
project-risks
- because of thinking in terms of functions rather than processes,the system-analysts and designers give preferences to the OO technology (OT). The primary reason is the dramatically higher productivity derived from the main features of OT such as polymorphism, encapsulation, the reuse capability of different model-elements such as classes, objects and compo-nents, and afterwards the shortening of the project life cycle!
- because of the definitive, structured steps in the development process and
- because of the separated data and functional system-features etc.,
3. Modeling business processes
To describe and understand the business processes
it needs to abstract the main features, to define and analyze the components,
the relation among them, and the architecture of the organization. Modeling
is a central part of all activities that lead up to the deployment of good
systems. We build models to communicate the desired structure and behavior
of our system. We build models to visu-alize and control the system’s architecture.
We build models to better understand the system we are building, often
exposing opportunities for simplification and reuse. We build models to
reduce risk.
A model is a simplification of the realty or
we can say it provides the blueprint of a system. A good model includes
those elements that have broad effect, and omits those minor components
that are not relevant to the modeling objectives and/or to the given level
of abstraction. Every system may be specified from different aspects using
different models or model-views, and each model is therefore a semantically
closed abstraction of the system.
3.1. The Importance and the Principles of Modeling
Modeling is a proven and well-accepted engineering
technique for describing and understanding the real world. We build architectural
models of houses to help the owners visualize the final product, mathematical
models in order to analyze different effects, or models in the field of
economics, and business management to validate our theories, or try out
new ones with minimal risk and cost. Through modeling we achieve four different
aims:
1. The choice of what models to create has a profound influence on how a problem is attacked and how a solution is shaped.Depending on the nature of the system, some models may be more important than others. For ex-ample: In a data-intensive business domain the static entity views dominate. In the real-time systems both the static and dynamic use case views are very important. In a knowledge-based environment the knowledge-representation models, or in a distributed system such as the Web-intensive E-business solutions the implementation and deployment models play key-role [14].
2. Every model may be expressed at different levels of precision, at different aspects of ab-straction, and the models have to be understandable.
3. The best models are in close connection to the reality, they use a unique terminology, and there is an easy way to control and update them.
4. No single model is sufficient, every nontrivial system is best approached through a small set of nearly independent models.
3.2. The OO as an Effective Modeling Paradigm
Concerning creating good models we have to answer
many questions as the followings: How to interpret the real-world components?
How to model the real world agents, roles, business units, communications,
system-functionality, interaction, behavior and activity of the system?
Why is it necessary to shift towards a new modeling concept?
Well-suited to the demands of the new e-world,
the modeling technology has to be designed to fulfil requirements of distributed,
concurrent, and connected features of systems [19]. This need to based
on real world agents, which distinct themselves from all others and maintain
their own state. Objects are concurrent each one can potentially execute
in parallel with all others. These business element can be called objects,
which have their own identity, state, and behavior. The objects are connected,
each one can send messages to others in different ways. Simply put, the
object is a thing, both a static and a dynamic representation generally
drawn from the problem-space or the solution-space. For example, consider
a simple architecture for a billing system, involving user interface, middleware,
and a database. In the database we can find concrete objects, such entities
from the business domain, including customers, products, and orders. In
another layer of abstraction there are objects such as transactions and
business rules, as well as higher level views of problem entities, such
as customers, products, and orders.
We can come closer to the terminology and the
essence of object-orientation from different ap-proaches [20]:
- Object-orientation is a technology for producing models that reflect a certain domain.The object-oriented approach to development is decidedly a part of the mainstream simply be-cause it has proven to be of value in building systems in all sort of problem-domains and encom-passing all degrees of size and complexity. Therefore the object-oriented UML is to bridge net-works of different systems, and it was designed with extensibility in mind, so it can adapt to new issues as they arise [19].
- Object-oriented models, when constructed correctly, are easy to communicate, change, ex-pand, validate, and verify.
- When correctly done, systems built using object technology (OT) are flexible to change, have well-defined architecture, and provide the opportunity to create and implement reus-able component.
- Object-orientation is not just a theory, but a well-proven technology used in a large number of projects and for building many different types of systems.
- Whatever the object orientation paradigm requires a method that integrates a development process and a modeling language with suitable construction technique and tools.
3.3. Enterprise Modeling Technique
How does the modeling concept in the case of
Enterprise Modeling look like? The Enterprise Mod-eling (EM) is an abstraction
process, which results a representation of a part or a set of enterprise
components, structure, activities, processes at a global and a detailed
level in order to understand its essence. This definition takes into consideration
different aspects, such as technological, economic, social and human ones.
The Enterprise Modeling Technique (EMT) allow to describe and under-stand
the organization in terms of strategy, objectives, architecture, functionality,
relationship (in-ternal and external). As the EMT gives a full description
of the enterprise it is an effective tool for a global optimization. If
we count with different aspects of the enterprise models we can say, that
the EMT is an n-dimensional model which we can term an enterprise meta-modell.
The Task Force of IFAC/IFIP defined the
Enterprise Modeling Technique on GERAM concept as a frame for the
enterprise modeling activity [11]. As it is known there are three main
enterprise modeling techniques meet all the demands of the requirements
of Task Force IFAC/IFIP: the GRAI, the CIMOSA and the PERA solutions [1],[21].
As the aim of this presentation is to emphasize the importance of a unified
enterprise modeling language I do not intend to give detailed information
about the EMTs, the reader can inform themselves from excellent publications
such as [3], [1].
4. The Unified Enterprise Modeling
Language
How to describe and document the different real
world components, their features and behavior, the activities and processes?
For the first sight there is a simple answer to this question, we have
our own language, and we can explain or write the situation with words
and sentences, or we can even sketch out ideas on a paper. However, there
are several problems with this, especially if You are a participant of
an enterprise in the E-economy. First, communicating the conceptual models
to oth-ers is error-prone unless everyone involved speaks the same language
I mean in terminology. Typi-cally, projects and organizations develop their
own technical language, and it is difficult to under-stand what is going
on if you are an outsider, or new to the group. Second, there are some
things concerning a software application system you can not understand
unless you build models that tran-scend the textual programming language.
Some things are best modeled textually, others
are best modeled graphically. But a bunch of graphical symbols alone can
not fulfill the requirements against a modeling language. It can be completed
only if there is a well-defined semantic behind each notation. In this
manner the analyzer and the developer can write the models in this modeling
language, and the another specialist, or even another tool can interpret
that model unambiguously.
4.1. The UML Standard
The Unified Modeling Language (UML) is a process-independent
standard language for modeling systems ranging from enterprise information
systems to distributed Web-based applications and even to hard real time
embedded systems. It has a very expressive formalism addressing all the
views needed to develop and then deploy the business processes. Although
the UML is developed mainly for visualize, specify, construct and document
the artifacts of a software-intensive system, it is qualified to describe,
analyze and understand the elements, relations and architecture of the
or-ganization and the business processes.
We have to state, that a modeling language differs
from a methodology, as it is for better under-standing and for the unambiguous
communication. But it is even more than a bunch of graphical symbols; rather
it has a conceptual architecture and a well-defined semantics. The syntax
describes the formalisms of the language, how the symbols should look,
and combine. The semantic rules tell us what each symbol means, how it
should be interpreted by itself, and they are planned to create models
of different abstraction levels, which are capable to clear the business
entities from the ir-relevant components and parts. The pragmatic rules
define the intentions of the symbols through which purpose of a model is
achieved and becomes understandable for others .
Figure 2. The conceptual model of the UML
To develop a UML-based UEML we have to analyze the conceptual architecture which consists of four major elements: (1) building blocks, (2) rules, (3) common mechanisms, and (4) architecture, as it is summarized in Figure 2.
1. The UML encompasses three kinds of building blocks:- The things are the abstractions of the real world components, such as entities, humans, sys-tem-elements and all kinds of other agents. The structural things are the nouns of the language, they are mostly static part of the model, representing elements that are either con-ceptual or physical. Structural things are classes, interfaces, collaborations, use cases, com-ponents and nodes. The behavioral things are the dynamic part of the UML, which are the verbs of a model, representing behavior over time and space, such as interactions and the state machine. The grouping things compose the organizational parts, this is an effective possibility for the modeler to organize elements in groups. These UML components are the packages. The annotational things are the explanatory parts, a chance to give comments and remarks about any element in a model.2. As the building blocks can not simply be thrown together in a random fashion it is needed to describe rules. The UML rules dictate how the building blocks may put together and specify what a well-formed model should look like.
- The relationship specification express the fact and the type of the connection between model-elements, it shows how these things are tied together. The dependency is a semantic relationship between two things in which a change to one thing may effect the semantic of the other. The association is a structural relationship, that describes a set of links, which is a connection between objects. The generalization is a relationship in which objects of the specialized element (child) are substitutable for objects of the generalized element (parent). The realization is a semantic relation between classifiers, wherein one classifier specifies a con-tract that another classifier guarantees to carry out.
- The diagram is a graphical presentation of a set of element, most often rendered as a connected graph of vertices and arcs. The diagram aims to visualize the system from different perspectives, so a diagram is a projection into a system. There are nine different kind of dia-grams defined in the UML, five reflect the static view of the system: class-, object-, compo-nent-, deployment- and the use case diagrams, and four of them are give a dynamic view: the sequence-, the collaboration-, the state- and the activity diagrams.
3. The common mechanisms apply throughout the UML, and help to refine the specification. The UML propose to use consistent solution:4. The modeling of the system architecture can be the most important artifact of the whole activ-ity, because from this point of view we can manage the different viewpoints and control the it-erative and incremental development of a system throughout its lifecycle. The architecture is a set of significant decisions about the organization of the software system, the selection of the elements, the behavior of the elements and the subsystems, and the architectural-style of the or-ganization, such as static and dynamic element, interfaces, collaborations, compositions etc. The UML handles different views of models such as: use case, process, logical (design), component (implementation) and deployment views (see Figure 3).
- - The UML is not only a graphical representation tool, rather behind every part of its graphi-cal notation there is a specification that provides a textual statement of the syntax and se-mantics of that building block. The UML’s graphical notation is for visualizing the real world, and the UML’s specification is to state the system’s details.
- - There is the adornments possibility in the UML to order to the elements a unique and direct graphical notation providing a visual representation of the most important features of the element.
- - The common divisions concept means, that the world often gets divided in at least a couple of ways.
- - The UML’s extensibility mechanisms gives an opportunity to express all the possible nu-ances of all models across all domains across all the time in order to be sufficient with ex-tending the standard rules of the modeling language. For this reason, the UML is open-ended, making it the modeler possible to extend the language in controlled way. There are three different possibilities in the extensibility mechanisms: (1) The stereotype extends the vocabulary of the UML, allowing to create new kinds of building blocks. (2)The tagged value extends the properties os a UML building block, allowing to add new information in the element’s specification. (3) A constraint extends the semantics for allowing to complete with new rules or to modify the existing ones.
Figure 3. The different views of UML [13]
4.2. The concept of the UEML
Setting out from the basic purpose of Enterprise
Engineering (EE) we can distinguish three main fields:
- The Business Process Engineering (BPE) is concerned with the analysis and design/redesign of business processes and activities.For engineering activity there are different modeling methods available, which are different in their nature and in emphasizing the system components, but they have similar features too. To describe our system and their features we can use different techniques and tools for modeling in the field of BPM, MSM or ISM . It is necessary to emphasize that the idea of UEML is not to develop a new modeling language but to use the standard of such kind of languages, and to develop and introduce new capabilities for enterprise modeling. In this sense the objective of the UEML is to collect the modeling techniques used in the different enterprise formalisms solutions and to create a common enterprise modeling language on the basis of UML in order to make bridges between the various model approaches and the different applications.
- The Manufacturing Systems Engineering (MSE) analyzes and designs the manufacturing ac-tivities, aims to improve and optimize the manufacturing performances.
- The purpose of the Information Systems Engineering (ISE) process is to analyze and design the Information Systems and the information processing flow, to integrate the software appli-cations and the IT infrastructure with the business domain.
4.2.1. Business Modeling Solutions
An enterprise domain model is an abstraction
of the whole organization, and it captures the most important features
such as the objects, activities, processes, and environment in the context
of the related system. The purpose of the domain modeling is to understand
and describe all the compo-nents and parameters which are relevant in the
given problem-solving procedure. With the help of such a model the users,
the customers, the developers, and the stakeholders have a common termi-nology
in order to share knowledge with others.
Business modeling is a technique for understanding
the business processes of an organization. The most important tool of business
modeling is the use case. A use case models the functionality of the system
on different level of abstraction and represents not only a function, but
with the help of the diagram technique it can express the system’s agents
who are responsible to carry out these tasks. On the level of business
modeling the agents are usually persons, who are playing different roles
in the business processes. We call these persons as actors. We can differentiate
actors as workers, entities, internal and business workers. With the help
of class- and object diagrams we can model the agents of a system. The
frame of a conference presentation can not allow to show all as-pect of
the given problem so I can focus only on the most interesting questions.
Let us analyze a sales procedure a little bit more detailed. The pattern
is concerning to the business model of a computer-assembling enterprise.
On the Figure 5. we can see how is it possible to model the system architecture. As it is known for an assembly process (procAss) there are several different tasks to carry out. In this sense this process can be taken as a system of some subsystems as follows:
- it is needed to purchase the necessary assemblies listed in the Bill of Material from a supplier: assSupply process and IdelivSheet list with the delivered assemblies,
- in the possess of all assemblies the hardware specialist (hwSpecial) assembles the required computer-type: assembly, and finally
- the administrator informs the customer (customer) about the installation date: custInform task and IcustNotice interface.
Figure 5. Embedded packages with external workers
and interfaces
The Figure 6. shows how is it possible to symbolize the business characteristics, and from the Figure 7 we can see, that there are different forms and ways to represent the same model for better understanding.
Figure 6. The extended graphical notations to
the business domain model
In Figure 6. we see, that we can use different symbols for the actors. A customer is an external per-son related to the system, the administrator is an employee working for the enterprise, and the comp-Type represents different types of computers, which are the entities of the system. If we would like to see the characteristics of these agents then we can draw the same model in another representation form also signing the main features (see Figure 7).
Figure 7. The next step of the abstraction related
to the Figure 6
4.2.2. Computer Aided Modeling and Engineering
Tools
Because it is quite difficult to manage development
and modeling tasks without any help of some kind of tools, so developer
are working continuously on different computer aided software product.
These CASE tools are very useful to analyze the business domain, to design
and visualize the dif-ferent model-views, to store and maintain the development
information in the repository and it is responsible for the change management
and version control. I prepared the patterns with Rational Rose’2000, but
there are also other powerful CASE tools as well, such as System Architect,
Power Designer etc.
5. Conclusions
As we can see from the pattern examined a business
problem from some different aspects the mod-eling techniques and the graphical
tools for formalisms gives us new possibilities particularly in development
and/or updating the business applications. No doubt, the organizations
are forced to fit to the aggressive changing environments and conditions,
and in this process, they have to use the most up-to-date information technology.
It is also proved that in this adaptation process to the New Economy all
participants have to change their way of thinking, have to solve their
problems by the most effective methods, techniques, and technologies. One
of these possibilities is to use a Unified Enterprise Modeling Languages
which helps all the customers, analyzers, developers, users and stakeholders
to “speak the same language” for better understanding each other. It is
especially ur-gent, because if all of them use the same graphical notations
for the same thing in the same context and with the same rules then the
adaptation life cycle can be dramatically shortened.
References
[1] AMICE: CIMOSA: Open System Architecture for
CIM – Springer Verlag, 1993.
[2] BOOCH, G.– JACOBSON, I.– RUMBAUGH, J.: The
Unified Modeling Language – User Guide – Addison-Wesley, 1999
[3] CHEN, D.- DOMEINGTS, G.– VALLESPIR, B.: GRAI
Grid Decision Modeling – in Handbook on Architecture of Information Systems,
Springer Verlag, 1998.
[4] CHROUST, G.: ISO15504: The New Standard for
Assessment and its Evaluation Tool –KSA VSE Õraha 1998.
[5] DOBAY, P.: Organization Development and Information
Technology – Conference at the University of Economics, Budapest, 1998.
[6] DOBAY, P.: Tele-cooperation, Tele-working,
Tele-teaching ... – IDIMT’2001, Keynote paper,
[7] DOMEINGTS, G.– DUCQ, Y. – KLEINHAUS, S..:
Enterprise Modeling in Year 2000 – 16th WCC ITBM Conference, 2000.
[8] DOUCEK, P.: Software Process Improvement
in Region of Europe – Czech Republic – 16th World Computer Congress, ITBM
Conference’2000 Beijing, China
[9] DOUCEK, P.: The new “E-Economy” – Challenge
for the Present and Future ... – Keynote paper,
http://www.sea.uni-linz.ac.at/systtechnik/conferences/idimt2001,
IDIMT’2001
[10] HENDERSON, J.C.–VENKATRAMAN, N.: Strategic
Alignment: Leveraging Information Tech-nology for Transforming Organization
– IBM Systems Journal, 1993. 32(1)
[11] IFAC/IFIP TASK FORCE: GERAM: Generalized
Enterprise Reference Architecture and Meth-odology – Version 1.4, ISOTC184/SC5/WG1,
N398, 1997.
[12] KNAPMAN, J.: Business-Oriented Constraint
Language – 3rd International Conference on the Unified Modeling Language,
University of York, 2000.
[13] ERIKSSON, H.E.– PENKER, M.: UML Toolkit
– John Wiley and Sons, Inc., 1998.
[14] KOBRYN, C.: UML 2001: a Standardization
Odyssey – Revision the UML 1.3 – Communica-tion of the Ach, Vol 42. No.
10., 1999 October
[15] RAFFAI, M.: Application
Development with UML standard by RUP methodology – UML user guide –
Novadat Publisher Inc., 2001.
[16] RAFFAI, M.: Information
Engineering – Novadat Publisher Inc., 1999.
[17] RAFFAI, M.: Objects
in Business Modeling – The Paradigm and Methodologies of the Object-oriented
Application Development – Novadat Publisher Inc., 2001.
[18] RAFFAI, M.: RTEBP
Re-engineering methodology – An adapted BPR Methodology for the Central
and Eastern European Business Cultures – Novadat Publisher Co., 1999.
[19] RUMBAUGH, J.: Trends in UML and E-Development
– http://www.therationaledge.com/dec_00/f_uml.html
[20] TAYLOR, D.: Object-oriented Technology:
A Manager’s Guide – Reading, MA: Addison-Wesley, 1991.
[21] WILLIAMS, T.J.: The Purdue Enterprise Reference
Architecture – Computer in Industry, 1994 Vol 24.
