Uncertainty and Multiple Objectives in Storage Control Problems (A Min-Max Approach)

Orlovsky, S., Rinaldi, S., & Soncini-Sessa, R. (1982). Uncertainty and Multiple Objectives in Storage Control Problems (A Min-Max Approach). IIASA Collaborative Paper. IIASA, Laxenburg, Austria: CP-82-017

Preview

Text CP-82-017.pdf Download (604kB) | Preview

Abstract

Analysis concerned with problems of the rational use of natural resources almost invariably deals with uncertainties with regard to the future behavior of the system in question and with multiple objectives reflecting conflicting goals of the users of the resources. Uncertainty means that the information available is not sufficient to unambiguously predict the future of the system, and the multiplicity of the objectives, on the other hand, calls for establishing rational trade-offs among them. The rationality of the trade-offs is quite often of subjective nature and can not be formally incorporated into mathematical models supporting the analysis, and the information with regard to the future may vary with time. Then the challenge to the analyst is to elaborate a mathematical and computer implemented system that can be used to perform the analysis recognizing both the above aspects of real world problems.

These were the issues addressed during the summer study "Real- Time Forecast versus Real-Time Management of Hydrosystems", organized by the Resources and Environment Area of IIASA in 1981. The general line of research was the elaboration of new approaches to analyzing reservoir regulation problems and to estimating the value of the information reducing the uncertainties. Computationally, the research was based on the hydrosystem of Lake Como, Northern Italy. This paper starts a short series of IIASA publications based upon the results obtained during the study. It describes the theoretical background for the new max-min approach to analyzing multiobjective problems with uncertainties and outlines briefly some applications of the approach to water reservoir regulation problems.

Actions (login required)

View Item