International Encyclopedia of Systems and Cybernetics

1. "The procedure of studying systems by identifying their limits and parts (subsystems), their aims, programs and effectiveness (O.R. YOUNG, 1974, p.299)

2. "Study and analysis of complex problems in terms of the relations between the subsystems that form its component parts with the objective of predicting the behavior of the whole from that of its parts" (UNE Glossary, 1983, p.26)

3. "The diagnosis, formulation, and solution of problems that arise out of the complex forms of interaction in systems of any kind" (Adapted from K. KRIPPENDORFF, 1986, p. 74).

4. "An analysis to suggest a course of action by systematically examining the costs, effectiveness and risks of alternative politics or strategies – and designing additional ones if those examined are found wanting" (E.S. QUADE, 1963, as quoted by M.C. JACKSON, 1992, p.75).

Systems analysis developed out of wartime military operations (1940-45) and was later extended to the study of complex systems in general, notably through the RAND Corporation (set up in 1947).

According to M.C. JACKSON: The methodology of systems analysis can be seen as consisting of seven major steps, as follows:

"- Formulating the problem

- Identifying, designing and screening alternative responses

- Building and using models for predicting the consequences of adopting particular responses

- Comparing and ranking alternative responses

- Evaluating the analysis

- Decision and implementation

- Evaluating the outcome" (p.76)

Systems analysis is basically oriented to technical systems planning, generally without participation of most stakeholders and scant consideration for the ways they are affected by the design.

KRIPPENDORFF states that this methodology is applicable "from hardware to corporations" whose aim is "to accomplish one or more specific objectives". Furthermore: "Systems analysis provides a variety of analytical tools, design methods and evaluative techniques to aid in decision making… " (Ibid).

I.G. BLOOR writes: "Systems Analysis cover such a computer-related technique which really bases its principles on Organization and Methods" (1987, p.55). This is of course a much reduced and sheer technical view of systems analysis.

According to J.de ROSNAY, in Systems Analysis one defines "the limits of the system to be modelized, identifies the most important elements and the interactions modes between these elements, determines the linkages that integrate them in an organized whole. Elements and linkages are classified and hierarchized. Next, one seeks and identifies the flow variables, the state variables, the positive and negative feedbacks, the delays, the sources and the sinks. Every feedback is segregated and its influence on the various subjects of the system is assessed." (1990, p.113). Accordingly to this view, systems analysis should be a preparatory stage for systems dynamics runs.

Systems Analysis defines the data to be used to built a model of the system. However, and unfortunately, in many cases, the data, when provided, are incomplete, insufficient or poorly assessed. G. KLIR 's "reconstructability analysis" aims to improve systems analysis as a method.

For L. TRONCALE, systems analysis is the "most reductionist of systems approaches; the collection, treatment and validation of concrete data on the multiple components of a specific real system;… often leads to simulation of the system for the purpose of quantitative prediction; results are context-dependent;… does not focus on isomorphies; (it) relies heavily on the use of mathematical formalisms and the use of the computer;… Ieads to usually restricted to the detailed study of one particular system (and) so it is less comparative across different cases, even within a conventional discipline, than systems theory; when it does involve multi-disciplinary comparisons, it focuses on one problem, or design goal, using one tool, as in the study of acid rain, or global climate" (1985, p.45).

In synthesis, the technical tool may well limit the vision of the modelizer and the scope of the enquiry.

Unfortunately systems analysis has produced in the fifties and the sixties "some exagerated and highly oversold claims… as a tool for making public policy" (G. KLIR, 1990, p.166). This situation led to strong criticism by I. HOOS (1972), D. BERLINSKI (1976) and R. LILIENFELD (1978).

Still more unfortunately, Systems Analysis was frequently confounded with the Systems Approach, and as a result, the critique of the former one, in not a few cases, was extended in a superficial way to the latter one. One of the main sources of these negative appreciations has been the somewhat hyperbolic and imprudent use of J. FORRESTER's Systems Dynamics in the Club of Rome studies. For a balanced appreciation of these critiques see G. KLIR (1991, p.166-8).