Title: A competitive and bi-space sparsity-based artificial bee colony algorithm for large-scale multi-objective optimisation

Authors: Jiayao Qian; Jinyu Xu; Shitao Liao; Hui Wang

Addresses: School of Information Engineering, Jiangxi University of Water Resources and Electric Power, Nanchang 330099, China ' School of Information Engineering, Jiangxi University of Water Resources and Electric Power, Nanchang 330099, China ' School of Information Engineering, Jiangxi University of Water Resources and Electric Power, Nanchang 330099, China ' School of Information Engineering, Jiangxi University of Water Resources and Electric Power, Nanchang 330099, China

Abstract: This paper proposes a novel artificial bee colony (ABC) algorithm, called LMOABC, to enhance the global search capability of ABC in solving large-scale multi-objective optimisation problems (LSMOPs). By introducing a competitive mechanism, the population is segregated into two distinct groups: 'losers' and 'winners'. To achieve a trade-off between exploration and exploitation, employed bees perform the global search on the 'losers', while onlooker bees conduct the local search on the 'winners'. Moreover, a bi-space sparsity method is designed in the scout bee stage, in which stagnated solutions are re-initialised to sparse regions. To verify the performance of LMOABC, nine well-known LSMOP benchmark problems with dimensions of 100, 500 and 1,000 are tested. Experimental results show that it performs competitively compared to several recent LSMOEAs.

Keywords: large-scale multi-objective optimisation; artificial bee colony algorithm; multi-objective optimisation; large-scale optimisation; evolutionary algorithms.

DOI: 10.1504/IJBIC.2025.149625

International Journal of Bio-Inspired Computation, 2025 Vol.26 No.3, pp.157 - 168

Received: 01 Apr 2025
Accepted: 02 Aug 2025
Published online: 07 Nov 2025 *

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