Papers

Author(s)
Thomas Saaty
Joseph M. Katz Graduate School of Business
University of Pittsburgh
United States

Publication date: Feb, 1998

Journal: Operations Research
Vol.: 4- Issue: 6- Pages: 9-16

Abstract: This paper is an outgrowth of a talk given at a plenary session of the national meeting of ORSA/TIMS (now INFORMS) in May of 1996. Rather than speculate on what might be the possible domain of OR/MS within a systems framework, the paper gives a sketch of what basic ideas there are today that can help us get to a unified mathematical theory of the subject. The concept of influence occurs in all fields of knowledge—from physics, with its gravitational and electromagnetic influences, to sociology with its societal, political, economic, and technological influences. We argue that influence, a sensed, perceived or inferred stimulus, is the single most central concept for analyzing causal relations in OR/MS problems. Problem solving is contextual and focuses on the distribution of influence in allocation, queuing, inventory, and similar problems by manipulating measurable quantities. Because most influences are abstract and intangible, emphasis on creating structures to represent and measure the flow of influence of intangibles and their propagation is critical for the development of a general scientific theory for OR/MS, more critical than in any other field because our problem domain is very general and interdisciplinary. The paper advocates the need for a systemic integration of the diverse approaches used in OR/MS within a single framework for all areas, including dependencies and feedback among influences to maintain the full integrity of the problems we solve. Examples and illustrations are given together with observations about the use of creativity and intelligence to move the process of creating a theory beyond the traditional process of problem solving.

Keywords: Philosophy of modeling, Creativity, Mathematical modeling, Systems solution

URL: https://doi.org/10.1287/opre.46.1.9