Decision Making under Uncertainty with Ordinal Linguistic Data

Abstract

Decision Science has long provided prescriptive methods for choice decisions (in the tradition of Von Neumann and Morgenstern) based on interval scaled (dis)utility consequence measures and ratio scaled subjective probability measures. Over the past three decades, soft computing has provided alternative approaches with less stringent scale assumptions. After a brief introduction, this paper presents a description of five ordinal methods for decision making under uncertainty in the context of linguistic data: Possibilistic Decisionmaking, Revised Possibilistic Decisionmaking, Commensurate L-Fuzzy Risk Minimization, Fuzzy Relational Ordinal Risk Minimization, and Quadratic Ordinal Psychophysical Optimization. Finally, these five techniques are illustrated using a single example.