An improved chaotic-based African buffalo optimisation algorithm

Abstract

Optimisation remains inevitable in any organisation as the need to maximise the limited resources persists. It justifies the seemingly endless research in this area. This study explores the effectiveness of chaos to mitigate false or premature convergence problem in African buffalo optimisation (ABO) algorithm. Chaos employs the ergodic and stochastic properties to handle this limitation. Three resourceful chaotic functions in the literature are evaluated to find the best strategy for ABO improvement. The same strategy is applied across the algorithms under study to provide an unbiased judgment. The study validates the proposed system’s performance with a range of nonlinear test functions. The proposed system’s result is compared with standard ABO, Particle swarm optimisation (PSO), and chaotic particle swarm optimisation (CPSO) algorithms. Although chaotic ABO (CABO) gave 92% performance in comparison with standard ABO, chaotic PSO, and standard PSO; it requires further investigation. To be more explicit, the reason for no significant difference between chaotic-ABO and standard ABO in some functions calls for further research attention. The present study also highlights the research future scope. In all, the study gives insight to researchers on the appropriate algorithm for a real-world problem.