Operations research has been used to solve only a fairly limited number of managerial problems. Its limitations should not be overlooked.
In the first place, there is the sheer magnitude of the mathematical and computing aspects. The number of variables and interrelationships in many managerial problems, plus the complexities of human relationships and reactions, calls for a higher order of mathematics than nuclear physics does. The late mathematical genius John von Neumann found, in his development of the theory of games, that his mathematical abilities soon reached their limit in a relatively simple strategic problem. Managers are, however, a long way from fully using the mathematics now available.
In the second place, although probabilities and approximations are being substituted for unknown quantities and although scientific method can assign values to factors that could never be measured before, a major portion of important managerial decisions still involves qualitative factors. Until these can be measured, operations research will have limited usefulness in these areas, and decisions will continue to be based on non-quantitative judgments.
Related to the fact that many management decisions involve un-measurable factors is the lack of information needed to make operations research useful in practice. In conceptualizing a problem area and constructing a mathematical model to represent it, people discover variables about which they need information that is not now available. To improve this situation, persons interested in the practical applications of operations research should place far more emphasis on developing this required information.
Still another limitation is the gap between practicing managers and trained operation researchers. Managers in general lack a knowledge and appreciation of mathematics, just as mathematicians lack an understanding of managerial problems. This gap is being dealt with, to an increasing extent, by the business schools and, more often, by business firms that team up managers with operations research. But it is still the major reasons why firms are slow to use operations research.
A final drawback of operations research at least in its application to complex problems is that analyses and programming are expensive and many problems are not important enough to justify this cost. However, in practice this has not really been a major limitation.