Title: Backtracking immune algorithm for continuous multi-objective optimisation

Authors: Ahmed Tchvagha Zeine; Emmanuel Pagnacco; Rachid Ellaia

Addresses: LERMA Laboratory, Engineering for Smart and Sustainable Systems Research Centre, Mohammadia School of Engineers, Mohamed V University of Rabat, BP 765, Ibn Sina Avenue, Agdal, Rabat, Morocco ' LMN, INSA-Rouen, National Institute of Applied Sciences of Rouen-France, BP 08, University Avenue 76801, St. Etienne du Rouvray, France ' LERMA Laboratory, Engineering for Smart and Sustainable Systems Research Centre, Mohammadia School of Engineers, Mohamed V University of Rabat, BP 765, Ibn Sina Avenue, Agdal, Rabat, Morocco

Abstract: In this paper, a new multi-objective immune algorithm (MOIAs) called BIAMO is proposed for tackling continuous multi-objective optimisation problems. It uses the updated archive to sort the non-dominated solutions of the Pareto front as well as the mutation and crossover operators of the backtracking search algorithm (BSA). Experimental results are produced for various benchmark problems and for a variety of engineering design problems. They show that, compared to the recent multi-objective optimisation evolutionary algorithms, the proposed algorithm improves not only the convergence capacity but also preserves the diversity of the population. In this paper, thirteen benchmark and engineering design problems are presented and the obtained results were compared with other well-known optimisers. The obtained results demonstrate that the proposed algorithm requires less number of function evaluations and in most cases gives better results compared to others considered algorithms.

Keywords: multi-objective optimisation; evolutionary algorithms; backtracking search; hybrid recombination; hybrid mutation.

DOI: 10.1504/IJMMNO.2020.108619

International Journal of Mathematical Modelling and Numerical Optimisation, 2020 Vol.10 No.3, pp.287 - 306

Received: 19 Oct 2018
Accepted: 02 Aug 2019
Published online: 21 Jul 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article