An orthogonal-design hybrid particle swarm optimiser with application to capacitated facility location problem
Xianghua Chu, Ben Niu, Jing Liang, Qiang Lü · International Journal of Bio-Inspired Computation · 2016
To improve the performance of particle swarm optimiser (PSO) for global optimisation, a variant called orthogonal-design hybrid particle swarm optimiser (OHPSO) is presented in this paper. A permutation strategy based on orthogonal experimental design is developed as a metabolic mechanism to enhance population diversity. In addition, a hybrid learning strategy is proposed to exploit the particles' best experiences and direct the individuals more efficiently. OHPSO is tested on a set of 18 benchmark functions with various properties, and nine state-of-the-art PSO variants are adopted for comparison. Experimental results and statistical analyses indicate a significant improvement of the proposed algorithm. Furthermore, OHPSO is applied to a practical engineering problem, the capacitated facility location problem, to justify its real-world performance and applicability. The experiment results are highly competitive with existing bio-inspired algorithms in the location optimisation.