Abstract
It is widely recognised by decision analysts that real decision-makers always make estimates in an interval form. An overview of techniques to find an optimal alternative among such with imprecise and interval probabilities is presented. Scalarisation methods are outlined as most appropriate. A proper continuation of such techniques is fuzzy rational (FR) decision analysis. A detailed representation of the elicitation process influenced by fuzzy rationality is given. The interval character of probabilities leads to the introduction of ribbon functions, whose general form and special cases are compared with the p-boxes. As demonstrated, approximation of utilities in FR decision analysis does not depend on the probabilities, but the approximation of probabilities is dependent on preferences.
Notes
1. Based on ‘Eliciting Parameters in Decision Analysis and Fuzzy Rationality’, by N.D. Nikolova, D. Dimitrakiev and K. Tenekedjiev, which appeared in the Proceedings of the 4th International IEEE Conference on Intelligent Systems IS’08, pp. 15-18–15-23.