An adaptive questioning procedure for eliciting PROMETHEE II's weight parameters

For most decision making problems, finding representative parameters of a decision maker's preferences remains a challenging task. We adapt to the PROMETHEE II outranking method an eliciting procedure named Q-Eval that has been developed in the context of multi-attribute utility theory (MAUT). It refines the estimation of the preference parameters by iteratively querying the DM through pairwise action comparisons. After formalising an extension of this method to two other types of queries: 1) selection of one action from a subset; 2) ranking of a subset of actions, we propose an adaptive query selection scheme that presents to the DM the most discriminating query type at each step of the process. Simulation results show that this approach improves the efficiency of the eliciting phase in terms of convergence speed.