Robustness using multi-objective evolutionary algorithms

Abstract

In this work a method to take into account the robustness of the solutions during multi-objective optimization using a Multi-Objective Evolutionary Algorithm (MOEA) was presented. The proposed methodology was applied to several benchmark single and multi-objective optimization problems. A combination of robustness measures and the use of the Reduced Pareto Set Genetic Algorithm with Elitism (RPSGAe), that is an algorithm that distributes the solutions uniformly along the Pareto frontier, provided good results and are expected to be adequate for "real" optimization problems.