An improved version of a reference-based multi-objective evolutionary algorithm based on IGD +

Abstract

In recent years, the design of new selection mechanisms has become a popular trend in the development of Multi-Objective Evolutionary Algorithms (MOEAs). This trend has been motivated by the aim of maintaining a good balance between convergence and diversity of the solutions. Reference-based selection is, with no doubt, one of the most promising schemes in this area. However, reference-based MOEAs are known to have difficulties for solving multi-objective problems with complicated Pareto fronts, mainly because they rely on the consistency between the Pareto front shape and the distribution of the reference weight vectors. In this paper, we propose a reference-based MOEA, which uses the Inverted Generational Distance plus (IGD+) indicator. The proposed approach adopts a novel method for approximating the reference set, based on an hypercube-based method. Our results indicate that our proposed approach is able to obtain solutions of a similar quality to those obtained by RVEA, MOEA/DD, NSGA-III and MOMBI-II in several test problems traditionally adopted in the specialized literature, and is able to outperform them in problems with complicated Pareto fronts.