A metaheuristic optimization-based indirect elicitation of preference parameters for solving many-objective problems

A priori incorporation of the decision maker's preferences is a crucial issue in many-objective evolutionary optimization. Some approaches characterize the best compromise solution of this problem through fuzzy outranking relations; however, they require the elicitation of a large number of parameters (weights and different thresholds). This paper proposes a novel metaheuristic-based optimization method to infer the model's parameters of a fuzzy relational system of preferences, based on a small number of judgments given by the decision maker. The results show a satisfactory rate of error when predicting new outcomes with the parameter values obtained by using small size reference sets.