Mutation strategies toward Pareto front for multi-objective differential evolution algorithm

Mutation strategies toward Pareto front for multi-objective differential evolution algorithm

0.00 Avg rating0 Votes
Article ID: iaor2014418
Volume: 19
Issue: 3
Start Page Number: 315
End Page Number: 337
Publication Date: Feb 2014
Journal: International Journal of Operational Research
Authors: ,
Keywords: evolutionary algorithms, Pareto front
Abstract:

This paper presents a multi‐objective differential evolution algorithm, called MODE, to search for a set of non‐dominated solutions on the Pareto front. During the iterative search process, the non‐dominated solutions found are stored as the 'Elite group' of solutions. The study focuses on utilising the solutions in the Elite group to guide the movement of the search. Several potential mutation strategies in MODE framework are proposed as the movement guidance in order to obtain the high‐quality front. Each mutation strategy possesses distinct search behaviour which directs a vector in the DE population in different ways with the purpose of reaching the Pareto optimal front. The performance of the proposed algorithm is evaluated on a set of well‐known benchmark problems and compared with results from other existing approaches. The experimental results demonstrate that the proposed MODE algorithm is a highly competitive approach for solving multi‐objective optimisation problems.

Reviews

Required fields are marked *. Your email address will not be published.