Quasi-Hierarchical Approach to Discrete Multiobjective Stochastic Dynamic Programming
Quasi-Hierarchical Approach to Discrete Multiobjective Stochastic Dynamic Programming

Summary/Abstract: In this paper we consider a multi-stage, multi-criteria discrete decision process under risk. We use a discrete, stochastic dynamic programming approach based on Bellman’s principle of optimality. We assume that the decision maker determines a quasi-hierarchy of the criteria considered; in other words, he or she is able to determine to what extent the optimal expected value of a higher-priority criterion can be made worse to improve the expected value of a lower-priority criterion. The process of obtaining the final solution can be interactive. Based on the observations of the consecutive solutions, the decision maker can modify the aspiration levels with respect to the criteria under consideration, finally achieving a solution which satisfies him/her best. The method is illustrated on an example based on fictitious data.

• Details
• Contents