Systems Thinking’

PETER CHECKLAND

Introduction

It may appear obvious to an optimist that ‘management information systems’ (MIS) are a particular kind of ‘system’ and that systems theory should be there in the background to underpin the creation of such systems. This chapter cannot but dispel that naive optimism. There is no simple link between systems theory and work on MIS, no spick-and-span body of ready-made theory which can be used in a straightforward way to help in the design of such systems. But all is not lost. The process of systems thinking that is to say: consciously organised thinking using systems ideas, is very relevant to the problems of work in the MIS field. And some of the problems which have been faced and resolved in the development of systems thinking itself are very relevant to the MIS field, and can usefully illuminate its problems. That is the kind of underpinning which this chapter will seek to provide for the book as a whole. At the start it may be useful to illustrate the kind of difficulties which have to be faced in this task.

In the early 1970s, my group at Lancaster University were invited by a director of what was then the British Aircraft Corporation to take a ‘systems engineering’ approach to the Anglo-French Concorde project. Our aim was to give advice on how we could improve the management of the project. This was at a time when the two pre-production aircraft, one in Bristol, one in Toulouse, were nearing completion, but had not yet flown. The Concorde project was, at the time, the subject of much public debate in the UK, since it was by then very apparent that it was going to take years longer to develop than originally thought and would cost many millions more than originally estimated. As we started the work, we took it as completely obvious, not to be questioned, that this was clearly an engineering project, one which aimed to create the world’s first supersonic passenger aircraft; and it seemed obvious too that this project was ‘a system to create a Supersonic passenger aircraft according to a defined technical specification, within a certain time, at a certain cost, and under the constraints that it must gain the certificate from the Civil Aviation Authority which will enable the public to fly in it, and that it must not unacceptably damage the environment’. We had the idea that we could model this, and work out from the models the implied information requirements and the necessary management activity. Then we could, in the light of the models, examine the project’s real-world information and activity, and make suggestions for improvement.

In the event we were amazed when our models seemed to bear no relation at all to real-world structures and activities and, most importantly, engendered little interest in the engineers and managers working on ‘the Concorde project’. Our eventual hard-won learning was that there was in fact no ‘project’ in the accepted sense of the way the word is used in the management literature. Although the phrase ‘the Concorde project’ was on everyone’s lips, used many times each day, project management was not at all in evidence; that is not how Concorde was created.

            Realising our mistake provided valuable learning for us. We were beginning to

learn that there are great difficulties in an ill-formed and conceptually confused field like management (of which study of ‘MIS’ is a part) which stem from the fact that there is no language available for serious discussion which is separate from everyday language. Physical chemists know exactly what they mean by ‘entropy’, or ‘the ­branch of an infra-red spectrum’. Would-be scholars in the management field, on the other hand, have no shared precise meaning for many of their relevant concepts, for example ‘role’, ‘norm’, ‘culture’, or ‘information system’; all these terms are fuzzy as a result of their unreflective use in everyday chat. Serious work in management and in MIS needs always to be aware of that.

TO BE CONTINUED........