A GENERALIZATION OF THE OUTRANKING APPROACH BY INCORPORATING UNCERTAINTY AS INTERVAL NUMBERS

Resumen

The outranking approach is an important method in multi-criteria decision aid. Nevertheless, it requires information about the preferences of the decision maker that sometimes is difficult to determine. When it is ignored, often this difficulty causes vagueness or imprecision generating a poor performance in the overall model. The vagueness and imprecision generated this way can be considered in the form of interval numbers containing all possible values for the actual parameters. Evidently, the consideration of interval numbers instead of real numbers in the outranking approach demands restructuring the original method. In this study, we present a generalization of the outranking method that incorporates vagueness and imprecision in its parameters through interval numbers. Additionally, we propose a novel function to establish the possibility that an interval number is greater than or equal to another. The results of the experiments indicate that the method proposed is able to characterize uncertainty satisfactorily. Finally, we conclude that the classic outranking approach is a specification of the proposed method and that several intuitive requirements are accomplished.

KEYWORDS: decision aiding, outranking approach, handling uncertainty, preference modelling, interval numbers theory.

MSC: 90B50