Advanced Information Systems Engineering: Third International Conference CAiSE '91 Trondheim, Norway, May 13–15, 1991 Proceedings

585 Pages·1991·12.089 MB·English

by  Rudolf Andersen, Janis A. Bubenko Jr., Arne Sølvberg (eds.)

Lecture Notes ni
Computer Science
detidE yb .G Goos dna .J sinamtraH
498
R. Andersen J.A. Bubenko .rj
A. Selvberg ).sdE(
decnavdA noitamrofnI
smetsyS gnireenignE
drihT lanoitanretnI ecnerefnoC CAiSE 19'
,miehdnorT ,yawroN yaM 13-15, 1991
sgnideecorP
[11
Springer-Verlag
nilreB grebledieH weN kroY nodnoL siraP
oykoT Hong gnoK anolecraB tsepaduB
Editorial Board
D. Barstow W. Brauer R Brinch Hansen D. Gries D. Luckham
C. Moler A. Pnueli G. Seegm~iller .I. Stoer N. Wirth
Volume Editors
RudolfA ndersen
Arne Solvberg
Information and Knowledge Systems Group
Department of Electrical Engineering and Computer Science
The Norwegian Institute of Technology
N-7034 Trondheim, Norway
JanAi.s Bubenko .rj
SISU-Swedish Institute for Systems Development
Box 1250, S-164 28 Kista, Sweden
Conference sponsor
NESREDNA
G N I T L U S N O C
NESREDNARUHTRA & ,.OC .C.S
CR Subject Classification (1991): D.2.1-2, D.2.9-10, D.2.m
ISBN 3-540-54059-8 Springer-Verlag Berlin Heidelberg New York
ISBN 0-387-54059-8 Springer-Verlag New York Berlin Heidelberg
subject is work This to copyright. All rights era ,devreser the of part or whole the whether lairetam
rights the specifically concerned, is of illustrations, of re-use reprinting, translation, ,noitaticer
,gnitsacdaorb on reproduction smliforcim or other in ,syaw storage and in data .sknab Duplication
thereof parts or publication this of si only dettimrep provisions the under of the namreG Copyright
Law paid. be always must fee copyright of a and version, current its in 9, 1965, September
snoitaloiV act prosecution the under fall of the namreG Copyright .waL
© galreV-regnirpS Heidelberg Berlin 1991
in Printed ynamreG
Printing dna Beltz, Druckhaus binding: .rtsgreB/hcabsmeH
012345-0413/5412 paper acid-free - on Printed
Preface
The CAiSEo91 Conference at the Norwegian Institute of Technology in
Trondheim is the 3rd in a series of Conferences on Advanced Information
Systems Engineering, and we are very proud to host this year's event.
CAiSE°91 was preceded by two conferences arranged in Stockholm in
1989 and 1990 by SISU, the Swedish Institute for Systems Development.
Both these conferences attracted international papers of high quality, and
indicated the need for an international conference on advanced informa-
tion systems engineering topics.
The call for papers for CAiSEo91 was given a wide international distribu-
tion. The programme committee was chosen from very well reputed re-
searchers in the international information systems engineering commu-
nity, as well as key professionals in European industrial and consultant
companies. The number and quality of the submitted papers is very satis-
factory and ensures the quality of CAiSEo91. The 26 accepted papers rep-
resent authors from 71 countries. This is very encouraging for the future
of the CAiSE conference series.
A major challenge for any research community is to provide a forum for
presentation and exchange of research results and practical experiences
within its field. It is our hope that this year's conference may serve as a
contribution to this end, and that the CAiSE conferences will prove their
worth by enhancing communication between the information systems re-
search community and information system professionals.
The CAiSE °91 conference programme would not have been possible with-
out the efforts of the authors and of the programme committee members.
They deserve our deepest thanks.
CAiSE is an annual event in the area of information systems engineering.
Next year's conference will be held in Manchester, United Kingdom, and
hosted by UMIST, University of Manchester Institute of Science and
Technology.
Rudolf Andersen Janis A. Bubenko jr. Arne S¢~lvberg
Programme Coordinator Programme Chair General Chair
CAiSE*91 Conference Organisation
General ChAir
Arne S¢lvberg, Norwegian Institute of Technology, Norway
Programme Chair
Janis A. Bubenko jr., SISU and The University of Stockholm, Sweden
VrogrRmme Coordinator
Rudolf Andersen, Norwegian Institute of Technology, Norway
Committee Proglmmme
Rudolf Andersen, Norway BjSrn Nilsson, Sweden
Alex Borgida, USA Markku Nokso-Koivisto, Finland
Janis A. Bubenko jr., Sweden Setsuo Ohsuga, Japan
Peter Dad(cid:127)m, Germany Ari Okkonen, Finland
Christer Daldgren, Sweden Antoni Olivd, Spain
Cockd e H. van Dorser, The Netherlands Gert Schmeltz Pedersen, Denmark
Jan Dietz, The Netherlands Barbara ,icinreP ylatI
GOran Edin, Sweden Colette Rollnnd, France
Anne Cecilie Fagerlie, Norway Ola Sannes, Norway
Eckhard Falkenberg, The Netherlands Gunther Schlageter, Germany
Pierre Falzon, France Amilcar Sernadas, Portugal
Antonio Ftwtado, Brazil Matti Sihto, Finland
Sol Greenspan, USA Svein Stavelin, Norway
Jacques Hagelstein, Belgium 03] Steluholtz, Sweden
Keith Jeffery, England Peter M. Stocker, England
Gregor Jonsson, Sweden Aiistar Sutcliffe, England
Hannu Kangassalo, Finland Costantino Thanos, Italy
Erik Knudsen, Sweden Berit Thaysen, Denmark
Kari Kltnsitl~, Finland Franz van Assche, Belgium
Paul Lindgreen, Denmark Reind van de Riet, The Netherlands
Peri Loucopoulos, England Kees van Hee, The Netherlands
Henrik Maegaard, Denmark Yannis Vassiliou, Crete, Greece
Oscar Nierstratz, Switzerland Peter Vendelbo, Denmark
Organising Chair
Jianhua Yang, Norwegian Institute of Technology, Norway
Conference Treasurer
Trude Kvalsvlk, Norwegian Institute of Technology, Norway
Table of Contents
The Infornmtlon Industry Scenario in Year 2000, Trends and Change of Focus, for
the 1T Professional and for the End-User
G. Wille (N) .................................................................................. 1
Representation and Utilization of Non-Functional Requirements for Information
SystemDes~
L. Chung (Can) .............................................................................. 5
Ai.I~.CSI: An Expert System for Requirements Engineering
C. Cauvet, C. Proix, C. Rol]and (F) ....................................................... 31
Reusing Requirements Through a Modeling and Composition Support Tool
M. Fugini , M. Guggin0, B. Pernici (I) .................................................. 50
The KIWIS Knowledge Base Management System
M. Ahlsen, A. D'Atri, P. J0hannesson, E. Laenens, N. Leone, P. Rullo,
P. Rossi, F. Staes, L. Tarantino, L. Van Beirendonck, F. van Caclsand,
W. Van Sant, J. Vanslembrouck, B. Verdonk, D. Vermeir (B, I, NL, S) ............ 97
A Functional Method of Data Processing Based on Relational Algebra
M.E. Orlowska, I~G. Jeffery (Aus, UK) .................................................
Dealing with Granularity of Time in Temporal Databases
G. Wiederhold, S. Jajodia, W. Litwin (USA, F) ...................................... 421
Information as a Strategic Resour~: A Question of Communication
J. Olaisen (N) ............................................................................. 141
MetaEdit A - Flexible Graphical Environment for Methodology Modelling
I~ Smolander, I~ Lyytinen, V.-P. Tahvanainen, P. Marttiin (SF) ............... 861
PPP - An Integrated CASE Environment
J.A. Gulla, O.I. Lindland, G. Willumsen (N) ....................................... 491
SPADE - Towards CASE Tools That can Guide Design
V. Seppiinen, M. Heikkinen, R. Lintulampi (SF) .................................... 222
DBE: An Expert Too1 for Database Design
D. Bitton, J.C. Millman, S. Torgersen (USA) ........................................ 042
Ecrln~ll~sign A - Graphical Editor for Semantic Structures
F. Adreit, M. B0nj0ur (CH, F) ........................................................... 462
A Case Study Using the IMSE Extmrinmntation Tool
J. Hillston, A.L. Opdahl, R. Pooley (UK, N) .......................................... 482
A Rule Language to Capttwe and Model Business Policy Specifications
P. McBrien, M. Nidzette, D. Pantazis, A.H. Seltveit, U. Sundin,
B. Theodoulidis, G. Tziallas, R. Wohed (UK, B, GR, N, S) ......................... 703
Infornmtion Systems Development Using a Combination of Process and Rule
Basod~es
J. Krogstie, P. McBrien, R. Owens, A.H. Seltveit (N, UK) .......................... 913
Towards Automating Software Maintenance
S. Jarzabek, K. Tham (Singapore) ..................................................... 633
Software Process Planning and Execution: Coupling vs. Integration
C. Liu (N) .................................................................................. 653
Cooperating Transactions and Workspaces in EPOS: Design and Vre|iminary
Implementation
R. Conradi, C.C. Malm (N) ............................................................. 573
IV
Implementation and Critique of an Algorithm which Maps a RelationaDla tabase
to a Conceptual Model
K. Kalman (S) ............................................................................ 393
Attnlaute Abstraction
B. Wangler (S) ............................................................................ 614
The Essential System Model
J.L.G. Dietz (NL) ......................................................................... 144
Federated CASE Environment
S. Papahristos, W.A. Gray (UK) ....................................................... 164
An Organizational Information Systems Enhancement Advice Unit (OISEAU)
,~ Kieback, J. Niemeier (D) ............................................................ 974
Structuring ModelliKnngo wledge for CASE Shells
T.F. Verhoef, A.H.M. ter Hofstede, G.M. Wijers (NL) ............................. 205
AnO bject-Oriented Approach to Model MAnALsement
V. Wuwongse, J. Ma (Thailand) ....................................................... 525
Representing a System Specification with a Temporal Dimension in an Object-
Oriented I~nguage
A. Oelmann (S) ........................................................................... 045
The OO.Binary Relationship Model: A Truly Object Oriented Conceptual Model
O. De Troyer (NL) ........................................................................ 165
Organisation's IT Needsi n the Next Decade - The Swedish Vision
B. Nilsson (S) ............................................................................. 975
THE INFORMATION INDUSTRY SCENARIO IN YEAR 2000,
TRENDS AND CHANGE OF FOCUS, FOR THE IT PROFESSIONAL AND FOR
THE END-USER
GUNNAR E. WILLE, DR.ING. WILLE CONSULTING A.S.
PROFESSOR II, NORWEGIAN INSTITUTE OF TECHNOLOGY
"Le fin du siecle" of the last centuries, have had elements of
chaos, turmoil and change.
As far as the IT industry is concerned, this will be the case
as we move into the next century, hopefully not backing in
with "a bag full of worms".
The overall purpose of IT applications, is to improve
infrastructure, leading to competitive advantages and better
decision support in industry.
For government organisations likewise. IT must add value to
the performance of the day to day activities, and to the
longer term strategic development.
This close link implies that there must be consistency between
the Business Architecture and the IT Architecture, as
described by the "John Henderson Strategic Alignment Model".
When the Master Business Platform is changing, the Business
Architecture must be adjusted. This must have an impact on the
IT Architecture.
With the IT architectural crisis many companies do have
to-day, and at the same time, difficulties in meeting business
requirements, there is a need for fundamental change in the IT
Platform.
We are facing a paradigme shift.
The Business Platform is changing. The IT Platform is
changing. The effect is that the rate of change, is changing.
We will be entering a period of turmoil.
We are entering a "Fin de siecle" where expectations to
business in terms of profitability, employment, mode of
operations and employee flexibility will be changing.
Peter Drucker estimates that the number of management layers
will be halved, and number of middle managemnt will be reduced
with two thirds over the next 20 years. Management structure
and decision support systems must be adjusted.
IBM and the major players in the IT industry, including the
Open Systems Foundation and standardisation bodies, have
announced shift in platforms.
Beacause of company re-engineering, restructuring and
competitive repositioning, a non-linear new level of IT
functionality will be required.
Gartner Group Inc. suggests that the driving forces behind the
new functionality, will be:
* Ubiquity: Systems must run accross more than 90 % of all
functions, locations and people.
* Fertile environement: Provision of architecture and tools to
lower dramatically the full life-cycle cost of functions.
* Va~ue-added networks: Administrative and communications
functions must be a natural part of the network.
* On-Line-Transaction-Processing: To run in real time, there
is a need to run massive transaction loads.
* Cross functional information management: Availability of
data independent of physical location.
* Data granularity: Access to relevant data at different
levels in the organisation.
* Consistency: Automatic completion and confirmation of all
data transactions.
Gartner Group derives the following relationship and paradigme
shift from the 1970-1987 Platform, to the new Master Platform
for year 2000:
1970-1987 Platform New Master Platform
System 370 Software architectures
Hierarchial Peer to peer
User managed resource System managed resource
System performance User performance
Mainframe at core Workstation at core
MIPS/$ constrained Functionality constrained
Distributed processing Cooperative processing
Database management Repository management
Wait on software Wait on software, middleware
IS gated Buyer gates
Data only Multi media
Project Process, prototype
The 1990-ies will be a time of deep and profound change in
information systems.
The movement of the PC-age, and the consquential substantial
use of workstations, will continue.
The price performance gap, measured as $/MIPS, was in 1980
15:1 between mainframe and PC. To day it is 100:1, and in year
2000, it should be 700:1. This will be an important factor as
a driver for new applications. It will also impact who will be
in power to make decisions, the IT coordinator or the enduser.
Application development and decision making as to solutions
and selection of hardware, middleware and frontware will to an
increasing degree be taken by the end-user.
The most important driver for new applications, will not be
the "Function per dollar" syndrome, but the low entry price
per workstation per se. This includes software and training.
Both the technological and the organisational forces, will
lead to a broader use of IT. The present architectual chaos,
in terms of lack of compatibility between the PC, the PC/LAN
and the mainframe network, must be solved.
Connectivity is a key issue, that must be solved first, to
make a move towards a new architectural framework, client
server.
Client Server technology will be the solution. Cooperative
processing, where the mainframe is the node in the network,
and where departemental processors can off-load the PCs for
major tasks. All hooked together through a telecommunication
network, and with a systems managed storage concept. This is
what we are looking for. Access to data and to applications,
including Office Systems, will be via a client server
structure.
Unfortunately, we will have to wait till the turn of the
century, before it is available. However, we will have to
start planning our infrastructure such that we are ready.
Along the road, open systems will emerge. It is estimated that
by 1995, about 50 % of all systems, will be open systems
(UNIX, PS/2 and DOS).
As a parenthesis, the really important decision is not whether
to decide for UNIX or not. The important decisions are
selection of middleware (what is between the operating system
and the applications), and the API (the Application Program
Interface).
The future compatibility issues will be, application exchange
and data exchange through telecommunication interfaces. These,
are the issues to focus on, as we wander through the "fin du
siecle" trauma.
Hardware products will be more commodity like, as they can be
used on open platforms. Margins will be reduced, and there is
going on a restructuring of the industry.
IBMs market share will probably continue to decline through
1995, and then regain some momentum. Fujitsu/ICL will probably
strengthen their position as a strong number two in the
Information technology industry.
For the IT department this change towards commodity products,
open standards, end-user decision making and fundamental
change in the Business Platform, will imply substantial
challenges.
The role of the IS function will change.
The formal IS budget in the US, according to Gartner Group
Inc. is 2.4% of revenue in 1990, and will grow to 2.7% in
1995.
End-user IT spending is estimated to 2.4% in 1990, split 50/50
between budget and unseen expenses. This item is assumed to
increase to 5.0% in 1995, bringing the total number to 7.7%.
This total amount must be managed, and the rules must be set
by the IS manager. Following the rule of "Least resistance",
will lead to crisis and complete loss of control.
IT resources must be managed. An Enterprise Architecture must
be established and adhered to. Standards must be established,
and partnerships between IT professionals and end-users
formed.
Success requires joint efforts and good teamwork. IS
professionals must take their own medicine. CASE tools and
modern project management will ensure a smooth development and
preparation for the transition that we will have to go
through. The end-user and the IT professional must work
closely together, in this "fin du siecle" trauma we have ahead
of us.