Handbook on Architectures of Information Systems

This chapter is an introduction into the scope of the Handbook on Architectures of Information Systems. We will point out that this volume gives a comprehensive survey of the most important aspects in this area giving not only a list of available alternatives but providing also a guidance amidst the many proposals.

Information modeling is concerned with the construction of symbolic structures which capture the meaning of information and organize it in ways that make it understandable and useful to people. Given that information is becoming a ubiquitous, abundant and precious resource, information modeling is serving as a core technology for information systems engineering. We present a brief history of information modeling techniques in Computer Science and survey such techniques developed within Knowledge Representation (Artificial Intelligence), Data Modeling (Databases), and Requirements Analysis (Software Engineering and Information Systems). The presentation then offers a comparative framework for information modeling proposals which classifies them according to their ontologies, i.e., the type of application for which they are intended, the set of abstraction mechanisms (or, structuring principles) they support, as well as the tools they provide for building, analyzing, and managing application models. Examples of ontologies include static worlds consisting of entities and relationships, or dynamic ones consisting of processes. Generalization, aggregation, and classification are three of the best known abstraction mechanisms, adopted by many information models and used widely in information modeling practice. The final component of the paper uses the comparative framework proposed earlier to assess well known information modeling techniques, both from a user and a designer perspective.

The ISO standard 10303-11, also known as EXPRESS, is a formal modelling language for the specification of static aspects of an information model. To this intent EXPRESS provides object oriented constructs, such as specialisation, aggregation and modularization. For the specification of dynamic, behavioural and other non-static aspects of a model, various dialects of EXPRESS have been developed which will be unified in the future ISO standard, EXPRESS edition 2. EXPRESS or one of its dialects is being used in a number of research and industrial projects related to the area of product data technology for the specification of even large scale models such as, for instance, the application protocol development in STEP or the modelling of environmentally sound products.

Object-Role Modeling (ORM) is method for modeling and querying an information system at the conceptual level, and mapping between conceptual and logical (e.g. relational) levels. ORM comes in various flavors, including NIAM (Natural language Information Analysis Method). This contribution provides an overview of ORM, and notes its advantages over Entity Relationship and traditional Object-Oriented modeling.

SQL, a data sublanguage used to access relational databases, is sometimes described as “English-like” because many of its statements read a bit like English. It is a non-procedural language since complex data operations are formulated by specifying their intended result rather than the method used to obtain that result. Both ANSI and ISO have published three generations of the de jure SQL standards. The syntax and semantics of SQL is examined and the conformance requirements are stated; a few components of the language are considered in greater detail and the future of the language is outlined.

The objective of this contribution is to provide the basics of Petri net theory in order to model and evaluate Discrete Event Systems (DES). The first part of the contribution is devoted to the common definitions and properties of Petri nets. Qualitative properties are then introduced. These properties are those who are of importance when manufacturing systems are concerned. Finally, a short introduction of event graphs is proposed; these graphs are of utmost importance to study cyclic DES.

State transition diagrams are a graphic notation that has long been used to represent computing systems. Two basic models of state transition diagrams were introduced simultaneously by G.H. Mealy and E.F. Moore in the mid fifties, and have played a major role in hardware design for a long time. These basic models have been expanded significantly in the recent past to include such features as the ability to represent hierarchy, timing and communication, and have been used to model complex software systems. In this contribution, we discuss the original models of state transition diagrams, their semantic definition and their extensions; then we discuss current application domains and support tools.

This document describes the rationales and the specification of the Process Interchange Format (PIF). PIF is an interchange format designed to help automatically exchange process descriptions among a wide variety of process tools such as process modelers, workflow software, flow charting tools, planners, process simulation systems, and process repositories. These tools interoperate by translating between their native format and PIF. Then any system will be able to automatically exchange process descriptions with any other system without having to write translators for each pair of such systems. This document specifies the PIF-CORE 1.2, i.e. the core set of object types (such as activities, agents, and prerequisite relations) that can be used to describe the basic elements of any process. The document also describes a framework for extending the core set of object types to include additional information needed in specific applications. These extended descriptions are exchanged in such a way that the common elements are interpretable by any PIF translator and the additional elements are interpretable by any translator that knows about the extensions.

Business process management is the task to accomplish work in organisations such that processes are carried out in some form of “optimal” way. Two important tasks of business process management are planning and scheduling. Planning is concerned with structuring the processes i.e. determining what needs to be carried out and in what sequence to achieve the objective of the process. Scheduling is concerned with assigning limited resources over time to competing activities of business processes. A modelling language is presented for the purposes of planning and scheduling in support of business process management.

The purpose of this contribution is to serve as a clear introduction to the modeling languages of the three most widely used IDEF methods: IDEFO, IDEF1X, and IDEF3. Each language is presented in turn, beginning with a discussion of the underlying “ontology” the language purports to describe, followed by presentations of the syntax of the language — particularly the notion of a model for the language — and the semantical rules that determine how models are to be interpreted. The level of detail should be sufficient to enable the reader both to understand the intended areas of application of the languages and to read and construct simple models of each of the three types.

CIMOSA is an open system architecture for Enterprise Integration (EI), and especially for integration in manufacturing. The architecture comprises an Enterprise Modelling Framework, an Integrating Infrastructure and a System Life Cycle. This contribution presents the modelling languages used in the Enterprise Modelling Framework. The CIMOSA languages are based on an event-driven process-based model and cover functional, information, resource and organisational aspects of an enterprise (including a workflow language for specifying enterprise behaviour). They can be used at various modelling levels along the system life cycle, including requirements definition, design specification and implementation description. Principles of these languages have influenced standardisation work in the field (CEN and ISO) as well as the development of commercial tools for business process modelling and analysis.

ConceptBase is a meta data management system intended to support the cooperative development and evolution of information systems with multiple interacting formalisms. It supports a simple logic-based core language, O-Telos, which integrates deductive and object-oriented features in order to support the syntactical, graphical, and semantic customization of modeling languages as well as analysis in multi-language modeling environments.

Conceptual graphs (CGs) are a system of logic based on the existential graphs of Charles Sanders Peirce and the semantic networks of artificial intelligence. Their purpose is to express meaning in a form that is logically precise, humanly readable, and computationally tractable. With their direct mapping to language, conceptual graphs can serve as an intermediate language for translating computer-oriented formalisms to and from natural languages. With their graphic representation, they can serve as a readable, but formal design and specification language. CGs have been implemented in a variety of projects for information retrieval, database design, expert systems, and natural language processing. A draft ANSI standard for CGs has been developed by the NCITS T2 committee, the liaison to the ISO Conceptual Schema Modelling Facility (CSMF) project under ISO/IEC JTC1/SC21/WG3.

Among formalisms used to model complex systems and organisations, the GRAI Grid has a special status because it focuses on the decisional aspects of the management of systems. The GRAI grid defines the points where decisions are made (decision centres) and the information relationships among these. Models built using the grid allow the analysis and design of how decisions are co-ordinated and synchronised in the enterprise.

SOM is an object-oriented methodology for comprehensive and integrated modeling of business systems. It is based on a framework consisting of the layers business plan, business process model, and business application system as well as views on these layers focusing on specific aspects. This contribution presents the SOM language for business process modeling and shows how business application systems can be linked to business process models. The language uses concepts of systems theory and is based on the notions of business object, business transaction, task, event, and service. Business objects are coordinated by feedback control or by negotiation. Decomposition rules allow a stepwise refinement of a business process model. The contribution includes a detailed example to illustrate the methodology.

We survey the requirements, concepts, and usage patterns of workflow languages which are used in today’s commercial or prototypical workflow management systems. After briefly reviewing workflow application development processes, basic notions of workflow modeling and execution and their relevant properties are introduced. A coarse classification of workflow languages is presented, and the main features of common workflow languages are described in the context of a sample application process.

This contribution attempts to take a systematic look at the methods and tools for the design and construction of Software/Information System (ISs) of today and the future. Development of an ISs is a set of many complex and inter-related activities. These activities are shaped not only by the information technology, but also by the business need and trends. The paper provides a (conceptual) framework that the author found very useful while thinking about different aspects and activities of IS development, methods prescribing these activities and tools that support them. The paper also tries to “predict the future” of the methods and tools by taking into account the new developments in the areas of distributed computing and the Internet.

This chapter discusses the history and evolution of Information Engineering, with emphasis on the business-driven IE variant. It describes the methods used at each phase in the systems development life cycle: strategic business planning; strate-gic, tactical and operational data modelling; process modelling; systems design; and systems implementation. It describes the application and use of IE for Forward Engineering, Reverse Engineering and Business Re-Engineering, and illustrates business-driven IE principles with a Business Re-Engineering example. The chapter concludes with a summary of Internet and Intranet technologies; discussing development of Client/Server systems and Data Warehouses, and their deployment via the Internet and corporate Intranets.

Object-oriented software engineering is coming of age. The focus in the first two generations of object-oriented (OO) methods (around 1990 and 1994 respectively) was on techniques and modelling. In the current third generation approaches, exemplified here by OPEN, a software engineering process is the key element which supplies the necessary underpinning to link together the second generation techniques into a viable approach to software development in a commercial/business environment.

Acquiring an information system to meet new business needs is not a trivial task. It includes deriving the acquisition goal, developing a strategy for its implementation, contracting for parts of the acquisition goal, integrating the parts into the complete information system and into the business processes of the acquiring organisation. Properly addressing these issues during acquisition significantly increases the likelihood of a successful outcome. Effective acquisition of an information system and related services requires clear descriptions of the desired final state and the current situation. It is important that the customer and supplier have the same understanding of the current situation and the information system and related services to be achieved. Euromethod has been designed to help organisations with the acquisition of effective information systems and related services in a variety of situations. It encourages customers and suppliers to control costs and timescales, to manage risks, to improve mutual understanding and to reach a fair contract. Through the achievement of these objectives, the European Commission aims to encourage the opening of the information system (IS) market, to improve the mobility of people internationally, to ease the organisation of international projects by a flexible contract management.

Increasingly complex systems have stimulated the development of sophisticated methods and tools for enterprise design and analysis. Advances in information technology as well as significant progress in analytical and computational techniques have facilitated the use of such methods in industry. However, enterprise modeling and analysis methods are yet to make a significant impact in the decision-making process of most companies and organizations. In this contribution, we provide a detailed analysis of some of the major roadblocks to a broader use of enterprise modeling methods in industry. We then describe an approach that addresses each of those roadblocks. Finally, we provide an overview of a commercial software environment that implements the approach, explaining the primary motivations for developing the software and describing its main features and characteristics.

Workflow management is concerned with the execution of business processes, supporting division of labor between participants and partial automation of individual tasks. Currently, workflow management systems are primarily employed in the service business, e.g. banks, insurance companies, and public authorities. However, the objectives of introducing workflow applications — improved productivity, faster and more reliable processing of workflows, immediate availability of information in response to customer inquiries — are equally important in industrial business processes. Workflow management systems can serve as a platform for process design in areas like purchasing, sales, order processing, human resources management, and quality management for industrial processes. Workflows are well-defined work processes that are repeatedly carried out according to predetermined rules. Labor may be divided between multiple participants. Rules determine which work steps are carried out, in which sequence, and who is responsible. Workflow management systems like WorkParty from Siemens Nixdorf comprise tools for the definition of business case templates and workflows and support their execution in the corresponding runtime environment. This contribution describes the concepts and tools of the scaleable workflow product family WorkParty. It concludes by presenting the idea that the architecture of future application systems will rely on a workflow management infrastructure.

With PROPLAN any business process can be analyzed, depicted, documented and described by a standardized language. Hence it is an instrument to provide a high transparency of business processes and activities. Furthermore it supports strategic management tasks by improving the overall department information flow from sales to delivery. The examined process is depicted as a sequence of symbols following the applied modeling language. Therefore, weak points are revealed for optimizing the process. The implemented middleware concept PRAGMA enables PROPLAN’s mobile computing ability used in combined locations. Combined with Inter-/Intranet browsers in WWW formats PROPLAN can easily be integrated in an existing LAN or WAN environment.

In this article a general business process architecture is presented, which is based on the Architecture of Integrated Information Systems (ARIS) and which is composed of the four levels of process optimization, process management, workflow and application. The ARIS-House of Business Engineering encompasses the whole life-cycle range: from business process design to information technology deployment, leading to a completely new process-oriented software concept. At the same time, the architecture bridges the gap between business process modeling and workflowdriven applications, from Business Process Reengineering to Continuous Process Improvement.

Bonapart is a general use modeling, simulation and dynamic analysis tool designed to support both experts and non-experts in organizational decision making and strategic planning. Bonapart’s OO (object-oriented) design supports most key organizational and IS (information system) methodologies so it can be effectively used in organizational design, restructuring (e.g. BPR), communication, organizational memory management, IT design and implementation, cost control (e.g. activity-based costing) and documentation (e.g. ISO 9000 standards). Tailored interfaces to other critical work-related and IS applications maximize the usefulness of Bonapart data when implementing workflow, CASE technologies, data warehouse and enterprise-wide (e.g. SAP) applications. Since both process and actor-oriented views are modeled, information processing is considered from both the technological and the organizational perspectives allowing information within an organization to be more visible and controllable.

The planning of information systems requires discussions between different project groups, within the respective project group, and between experts and managers in the enterprise and the project members. Therefore, the modeling of business processes and the related information systems is an essential step in the process of reorganizing enterprises. The software tool MO2GO (method of object oriented business process optimization) supports the modeling process based on the IEM method. Different analysis of a given model are available using the MO2GO tool like the planning of informations systems.

The following contribution describes the IBM VisualAge product family for object oriented application development. The different offerings constitute an application development environment for object oriented languages like C++, Smalltalk and Java. VisualAge for Smalltalk will be used as an example to describe some of the features of VisualAge in more detail.

The IBM Insurance Application Architecture (IAA) provides an architectural framework for the development of application solutions for the insurance industry. It is based on a general Insurance Business Architecture developed to provide common structures capable of representing the various, different business requirements occurring in the worldwide insurance companies. We will focus on the concepts, the contents and the positioning of the models representing this IAA Insurance Business Architecture. That way the motivation for a Business Architecture as a prerequisite for insurance specific business software components should become evident.

Modeling in the field of application “production and logistics systems” is still based on ad hoc procedures. This contribution describes the development of reference models as well as of supporting methods and guidelines for this field along with design criteria that are applied for the systematic decomposition of this complex area. Modularization and interoperability are the main terms in this discussion. Furthermore, this contribution indicates steps necessary for the development of a simulation integration platform within the Fraunhofer Society. This will allow application-centered working with a set of appropriate model libraries and simulators. An application example demonstrates usage and benefits of the described approach by means of utilizing the reference model “manufacturing systems”.

In the past few years business process modeling has become established practice in many enterprises. One area where it is used is in implementing standard business application systems. In such projects, reference models provide valuable support to enterprises when they are creating the business process models that describe their enterprise. Reference business process models give an overview of the business processes that are supported by the application system, and in doing so they help select the processes to be applied in an enterprise. However, it has been much more difficult to make use of business process models when you were setting parameters that change the behaviour of a standard business application system accordingly. It is described how the architecture of reference business process models can be extended to support the setting of parameters in a standard business application system. We then provide a practical illustration of the configuration of such a reference model and the setting of parameters in the system concerned.

The German Savings Banks Organization has established a large enterprise-wide data model as a standard for heterogeneous IT organizations. The basic elements, the architecture of the data model and practical experiences are described which show significant benefits for the organization on several levels.

The Reference Model of Open Distributed Processing (RM-ODP) was a joint effort by the international standards bodies ISO and ITU-T to develop a generic architecture for the standardisation of open distributed processing (ODP). The model describes a framework within which support of distribution, interworking, interoperability and portability can be integrated. The Object Management Architecture (OMA) was developed by the Object Management Group (OMG) as a specific architecture based on the generic principles and structures of RM-ODP. The OMA provides a framework for an integrated suite of standards for object-oriented distributed computing.

Information architectures only become relevant to most enterprises when they instance in commercially supported products. The mapping of technical appropriateness to commercial appropriateness is not straightforward; it involves a number factors from a larger perspective that act as architectural constraints, often resulting in unintuitive decisions. This contribution reviews those factors. It extends results from a large joint U. S./European precompetitive activity among major information infrastructure suppliers and a recent re-examination. Any such review which relies on specific products as examples is likely to become dated, so in the interest of making these insights more longer-lived, we’ll be more general than specific with regard to products.

Requirements of general purpose integration infrastructures are analysed in the context of realising more agile manufacturing systems. The analysis provides a framework for characterising generic capabilities of existing integration infrastructures. The framework is used to highlight the role of the CIM-BIOSYS integration infrastructure and its associated software tools. Also classified are necessary future developments before integration infrastructures can underpin evolutionary behaviour in distributed systems.

DCE and CORBA are two distributed processing technologies that provide remote procedure calls in a location-transparent manner between heterogeneous platforms. Java is not a distributed processing technology, but a programming language that can be executed remotely using Web browsers. There are advantages and disadvantages to the use of each of these technologies, and there are some benefits in combining them.

Agents are software components that support the construction of distributed information systems as collections of autonomous entities that interact according to complex and dynamic patterns of behavior. A major problem of multi-agent structured information systems is the coordination of these interactions and behaviors to achieve the goals of the participants and coherence of the system as a whole. This paper articulates a precise conceptual model of coordination based on a representation of coordination knowledge as plans described in a special planning language enhanced with communicative actions. The execution of these plans by agents results in multiple structured ‘conversations’ taking place among agents. The model is extended to a complete language design that provides objects and control structures that substantiate its concepts and allow the construction of real multi-agent systems in industrial domains. To support incremental, in context acquisition and debugging of coordination knowledge we provide an extension of the basic representation and a visual tool allowing users to capture coordination knowledge as it dynamically emerges from the actual interactions. The plan-action organization exhibited by the coordination language departs in several ways from the standard object orientation of computational languages and is, we argue, more appropriate to modeling coordination. The language has been fully implemented and successfully used in several industrial applications, the most important being the integration of multi-agent supply chains for manufacturing enterprises. This application is used throughout the paper to illustrate the introduced concepts and language constructs.