The Systems Approach

by C. West Churchman

The Systems Approach [Paperback]

Genres: Science, Business, Philosophy

Paperback

First published January 1, 1979

Reviews

[Edward · Rating 4 out of 5]

This book provides a good introduction and comprehensive overview of the input-output modeling approach to evaluating various problems related to organizational structure and social phenomena. This book is not an abstract technical description of the techniques for solving operations research (OR) problems using mathematical methods such as linear programming, etc. but its strength lies in elucidating concepts and ideas important to thinking about systems.

Churchman, an early practitioner in the field of OR, uses practical examples to provide a clear and simple explanation of different approaches to formulating the problem to be solved along with practical considerations and issues that should be taken into account when formulating a model based on the aforementioned methods.

Although I'm familiar with some of the technical concepts of OR, I found Churchman's explanations enlightening and useful because he addresses the fundamental issues associated with problem-solving. This issue is best illustrated in one of Churchman's papers that I read some time ago and dealt with what he termed "errors of the third kind" which is solving the wrong problem, an easy trap to fall into by making unwarranted assumptions and leaving out important considerations.

Churchman ventures into areas of behavior, values, and culture and discusses how they might impact the structure of the systems analysis model. These areas were fertile for research at the time and he really had to make a stretch to bring them to the fore for further exploration. The aspect of feedback between the various components within the model is also discussed and it can result in the failure of the model to accomplish the goals of the organization if not properly accounted for.

One thing that might have helped but was missing was a subject index at the back of the book so that the reader can refer back to a specific idea or concept.

[Yudhister · Rating 3 out of 5]

A core question this book studies is what it means to be a "management scientist": is it possible to study the management of a system without being an active party in it? Could you plan a system effectively even when considering yourself a part of it? Churchman doesn't quite provide a definitive answer to the question, but is certainly sympathetic to the affirmative.

Perhaps there are four archetypes of Systems Approach users:

The debaters are these:

(1) The advocates of efficiency; they claim that the best approach to a system is to identify the trouble spots, and especially the places where there is waste, e.g., unnecessarily high costs, and then proceed to remove the inefficiency.

(2) The advocates of the use of science in approaching a system; they claim that there is an objective way to look at a system and to build a “model” of the system that describes how it works. The science that is used is sometimes mathematics, sometimes economics, sometimes “behavioral” (e.g., psychology and sociology).

(3) The advocates of the use of human feelings, ie., the humanists; they claim that systems are people, and the fundamental approach to systems consists of first looking at the human values: freedom, dignity, privacy. Above all, they say, the systems approach should avoid imposing plans, i.e., intervention of any kind.

(4) The anti-planners, who believe that any attempt to lay out specific and “rational” plans is either foolish or dangerous or downright evil. The correct “approach” to systems is to live in them, to react in terms of one’s experience, and not to try to change them by means of some grandiose scheme or mathematical model. There are all kinds of anti-planners, but the most numerous are those who believe that experience and cleverness are the hallmarks of good management.

The author is clearly a Scientist, and believes in the potential of the scientific approach in both domains easily amenable to decomposition and quantitative modeling and those not.

The Scientist believes that efficiency-maximizers are old-fashioned and too simplistic. Obviously one must consider efficiency in the context of the broader system! Efficiency for efficiency's sake could have adverse second-order consequences!

The Scientist has some difficulty engaging with what some might call the "human perspective", and especially when dealing with value conflicts regarding the design of certain institutions. Operationalizing human values often takes the form of poorly lumping them in as economic considerations, or perhaps applying methods from behavioral science (haha.). As always, the Scientist's approach's effectiveness is directly downstream of the underlying ability of the system to be modeled, and we have not yet figured out how to nicely decompose sociological problems into an n-part moving system with edge relationships between the nodes.

Anti-planners aren't even trying.

With this in mind, we can outline five basic considerations that the scientist believes must be kept in mind when thinking about the meaning of a system:

1. the total system objectives and, more specifically, the performance measures of the whole system;

2. the system’s environment: the fixed constraints;

3. the resources of the system;

4. the components of the system, their activities, goals and measures of performance;

5. the management of the system.

Various dated case studies litter the rest of the book. One historical note is that "systems thinking" was developed in the military and especially in the aerospace industry: after World War Two, corporations looked at the effectiveness of e.g. fighter jet development and wanted some of it for themselves.

Two takeaways I found interesting were an inerrant focus on simplicity in modeling and a kind of self-referentiality paradox:

These considerations bring us to the last aspect of the system, its management. The management of a system has to deal with the generation of the plans for the system, i.e., consideration of all of the things we have discussed, the overall goals, the environment, the utilization of resources, and the components. The management sets the component goals, allocates the resources, and controls the system performance. This description of management, however, creates something of a paradox for the management scientist. After all, it is he who has been scheming and plotting with his models and analyses to determine the goals, environment, resources, and components.Is he, therefore, the manager; does he intend to “take over” with his computer army?

The truth of the matter is that he doesn’t want to. He is not a man of action, but a man of ideas. A man of action takes risks, and if he fails, not only does he get fired but his organization may be ruined; the man of action is willing to risk fortunes besides his own. The management scientist is typically a single risk-taker: if he fails, he doesn’t have to bear the responsibility of the whole organization’s failure.

Hence, we've found one chink in the scientist’s armor: he doesn’t really understand how he himself is a component of the system he observes. He likes to think that he can stand apart, like the elephant observer, and merely recommend, but not act. How naive this must appear to the politician is hard to say, but certainly the politician’s appreciation of the situation is the more sophisticated one. “Mere” recommendation is a fantasy; in the management scientist’s own terminology, it is doubtful whether the study of a system is a separable mission.

Recommended by Cosma Shalizi. A good read.