Architectures for Enterprise Integration

Transcription

1 Architectures for Enterprise Integration

2 Architectures for Enterprise Integration Edited by Peter Bernus Griffith University Queensland Australia Laszlo Nemes CS/RO Victoria Australia and Theodore.J. Williams Purdue University Indiana USA Published by Chapman & Hall on behalf of IFIP and IFAC c-ifac 0 m SPRINGER-SCIENCE+BUSINESS MEDIA, B.V.

3 First edition Peter Barnes, Laszlo Nemes and Theodore J. Williams Originally published by Chapman & Hall in 1996 Softcover reprint of the hardcover 1st edition 1996 ISBN DOI / ISBN (ebook) Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the UK Copyright, Designs and Patents Act, 1988, this publication may not be reproduced, stored or transmitted, in any form or by any means, without the prior permission in writing of the publishers, or in the case of reprographic reproduction only in accordance with the terms of the licences issued by the Copyright Licensing Agency in the UK, or in accordance with the terms of licenses issued by the appropriate Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publishers at the London address printed on this page. The publisher makes no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility or liability for any errors or omissions that may be made. A catalogue record for this book is available from the British Library Printed on permanent acid-free text paper, manufactured in accordance with ANSI/NISO Z and ANSIINISO Z (Permanence of Paper).

4 Contents List of contributors xiii Preface XIV Acknowledgements xvi 1 Introduction Enterprise Integration: the next generation of manufacturing System building principles Methodologies and the enterprise as a product Major projects References 8 2 The concept of enterprise integration The purpose assigned to the Task Force The status of enterprise integration The relationship of the Task Force's work to that of others Further definition of the relationship between architectures Main features of architectures Type 2 architectures in relation to Type Organization of this book References Further reading 20 3 The needs of the field of integration The needs of the field of integration The needs of EI beyond those of an architecture Why do we need an enterprise reference architecture? References 30

5 vi Contents 4 State-of-the-art on models, architectures and methodologies Basic concepts and definitions Early works on models of advanced manufacturing systems Methods for designing advanced manufacturing systems Recent works on models /architectures within ESPRIT Recent CIM concepts and developments in the USA Models developed by IBM and Digital Architecture Works, Asia-Pacific area Simulation models Object-oriented approaches for CIM Recent methodology development in the GRAI Laboratory Summary Bibliography and References Further reading 59 5 Some important articles about enterprise integration NIST CIMOSA ISO/TC-184/SC5/WG 1 Reference Model GRAI Laboratory, University of Bordeaux Purdue Case-Western Reserve University CAM-I Nippon Steel Corporation Society of Manufacturing Engineers (SME) International Business Machines Corporation (IBM) American Society for Testing and Materials (ASTM) Carnegie-Mellon University Digital Equipment Corporation (DEC) Honeywell, Inc. 66

6 Contents vii 5.15 Industrial Technology Institute (ITI) 5.16 The PROCOS-A/S system 5.17 SI Groups- Loughborough University, UK 5.18 SOFfECH, Inc. (ICAM) 5.19 University oftwente, The Netherlands 5.20 DoD materials and other architectures 5.21 Additional models recommended for study 5.22 Publications of general importance 6 A summary of the CIMOSA reference architecture 6.1 CIMOSA and its associated constructs 6.2 Enterprise integration - the solution 6.3 Management of change 6.4 Scope of enterprise integration 6.5 Enterprise engineering and operation 6.6 Coexistence with heritage/legacy systems 6. 7 Business benefits 6.8 Enterprise modelling and its requirements 6.9 AMICE project 6.10 CIMOSA approach 6.11 System life-cycle 6.12 Enterprise integration 6.13 CIMOSA overview 6.14 CIMOSA modelling framework 6.15 CIMOSA system life-cycle and environments 6.16 CIMOSA integrating infrastructure 6.17 Applying CIMOSA - a summary 6.18 Summary of architecture and concepts of CIMOSA 6.19 References 6.20 Further reading

7 viii Contents 7 The GRAI-GIM reference model, architecture and methodology 7.1 The global model The decision center model Ill 7.3 Modelling formalisms in the GRAI-GIM modelling framework Principle of the GRAI -GIM structured approach The modelling framework 7.6 Modelling formalisms 7. 7 Validation between models 7.8 GRAI-GIM structured approach 7.9 Initialization phase 7.10 Analysis phase 7.11 Design phase 7.12 Implementation phase 7.13 A summary of how to use GRAI-GIM 7.14 Conclusion 7.15 References 8 An overview of PERA A structure for PERA Description of the concept and definition layers A system of nomenclature The place of the human and organizational architecture Development of the implementation of the CIM system Summary of overview as related to manufacturing Enterprises and the architecture Expansion of PERA to cover any enterprise CIM system implementation - PERA and the Purdue Methodology 8.10 Using the implementation procedures manual 8.11 References 8.12 Further reading

8 Contents ix 9 Introduction to the evaluation methods used Method 1 - Questionnarires Method 2 - Mappings, one on one Method 3 - Mapping against a set of requirements Evaluation questionnaires Architecture '-Modelling Methodology Evaluation questionnaires for CIMOSA Architecture Modelling Methodology Evaluation questionnaire for GRAI-GIM Architecture Modelling Methodology Evaluation questionnaire for PERA Architecture Modelling Methodology Analysis of questionnaire results Architecture Short form evaluation questionnaire Answers to short form questionnaire for CIMOSA Answers to short form questionnaire for GRAI-GIM Answers to short form questionnaire for Purdue 239

9 X Contents 16 Techniques to map the architectures directly against one another Some initial observations A comparison of views (semantic problems) Mapping against other architectures Mapping of GRAI-GIM against CIMOSA and Purdue Introduction to the GRAI-GIM mapping Definitions Mapping of GRAI-GIM onto CIMOSA Mapping of GRAI-GIM onto Purdue References Mappings against a matrix The matrix of requirements or needs Rationale of the matrix presentation Eliminate unwanted combinations Development of the matrix form A way to situate a generic architecture in the matrix A set of matrix form evaluations General observations in comparing candidate architectures Comments on the modelling framework ofcimosa Comments on the IMPACS modelling framework Generic, partial, and particular building blocks References Conclusions and summary of findings General conclusions Specific conclusions regarding each candidate architecture Recommendations for future work General Recommendations Further development of each architecture I methodology 326

10 Contents xi 20.3 A combination of existing architectures Funding and cooperation Future roadmap for the promotion of enterprise integration A proposal for future work for the Task Force A proposed scope for any desired continuation of the Task Force References 332 Appendix A The background of the Task Force Background IFACIIFIP Task Force Some observations made at the earliest workshop meeting Scope and work program for the Task Force Suggested products of the Task Force Immediate work program Requirements for a Reference Architecture to be used as a set of evaluation criteria 336 Appendix B Members of the Task Force Index

11 List of contributors Peter Bernus (Griffith University, Australia) David Chen (Bordeaux University I, France) Guy Doumeingts (Bordeaux University, France) Hong Li (Purdue University, USA) Laszlo Nemes (CSIRO, Australia) John P Shewchuk (Virginia Polytechnic, USA) Jakob Vlietstra (AT&T Networks International, the Netherlands) Theodore J. Williams (Purdue University, USA) Dick Zoetekouw (AT&T Networks International, the Netherlands)

12 Preface This book is the result of the joint efforts of many people who have worked together over the past three years ( ). A group of manufacturing engineers, computer scientists and information technology managers formed a Task Force in 1990 to study, compare and evaluate the different available architectures for enterprise integration which were in the open literature. This joint working group was formed under the auspices and sponsorship of the two major international federations on automation and on information processing. These federations were the International Federation of Automatic Control (IFAC) for automation and the International Federation for Information Processing (IFIP) for the latter field. The basic task assigned to the Task Force was to study the available architectures and make recommendations for the future identifying one existing architecture as best. Failing to do this, the second best option seemed to be to build on an existing system by supplying the apparently missing capabilities. It was recognized that as the last resort it might be necessary to specify a whole new architecture from the characteristics of several of the others. The Task Force reported its findings in the predecessor report to this book. A summary of the report of the Task Force was presented at three separate major conferences of IFAC and IFIP during the second half of From the response generated there it was felt that the contents of the detailed study report would be of interest to many people, hence this book. Industry could find this book useful when determining the best methods to be used to help integrate their manufacturing operations. Researchers both in industry and academia should find it a valuable source of reference when investigating new areas of research and when comparing their present or proposed activities with the current state-of-the-art. Although is not a textbook as such it does, hopefully, contain enough information for the reader to understand the main principles and methodologies of the integration architectures described. Thus, the book should also find its way to the desks of university staff and students. Integration is, and will continue to be, a key factor in building and maintaining comprehensive enterprises. What we provide here is the state-of-theart of the methodologies required for use by everybody who is interested in expanding his/her knowledge in that field.

13 We would like to encourage all readers to report back any omissions or errors in the book. 1 We also welcome suggestions on how to improve further editions of this book when we have occasion to add our new results to the existing material. A complete list of the membership of the Task Force is given in Appendix B. XV The Editors 1. See on the World Wide Web.

14 Acknowledgements Special contributions to the work reported herein have been made by the following persons: Professor Ted Williams, chairman of the Task Force, and of the Purdue Industry Consortium that worked out the PERA Reference Architecture, as well as the associated Implementation Procedures Manual, developed many of the ideas including leading the one-on-one mapping of reference architectures, with the help of John Shewchuk and Hong Li, and Dick Zoetekouw (who initially proposed the one-on-one mapping). Professor Guy Doumeingts, together with David Chen prepared the state of the art survey on models, architectures and methods for integrated systems design as included in. this book; special mention should be made of Bruno Vallespir for his valuable insight in discussions. Dr Peter Bemus, vice-chair of the Task Force, initially proposed and developed the detailed questionnaire for the evaluation of reference architectures and analysed the results. He also developed the matrix presentation of reference architectures and prepared with Ted Williams, as well as the help of Francois Vemadat and David Chen, the detailed mapping of the three major architectures to the evaluation matrix. Dr Laszlo Nemes, together with Peter Bemus worked on the development of the detailed questionnaire evaluation form and the matrix presentation form of reference architectures. He also chaired the Canadian section (Toronto) meeting of the Task Force and has given his valuable management experience to the Task Force's strategic direction. He was also chairman of the IFAC Technical Committee on Manufacturing Technology (MANTEC) from where the Task Force originated. Ing Jakob Vlietstra, in the period of the first triennium of the Task Force, was information technology director of AT&T Networks International, the Netherlands (formerly AT&T Phillips), chairman and initiator of the AMICE consortium which developed CIMOSA. He has contributed to this book by his detailed explanations of the technical as well as strategic aims of the CIMOSA work. He also proposed the short form evaluation questionnaire. Jim Nevins was the initiator, in the IFAC Technical Committee on Manufacturing Technology (MANTEe), of the Task Force, and had a major influence on the establishment of the Task Force's scope and working plan at the initial Bordeaux meeting. The Task Force Members express their sincere

15 xvii gratitude to Mrs. Judy Ruzich, Mrs. Barbara Beaver and Ms. Paula K. Slyder of Purdue University, for their speedy and accurate compilation of this text and its earlier report and the numerous figures involved from their original massive, hand-written draft form. We also want to thank Dr Mingwei Zhou of CSIRO for his help in the proofreading process.