Taguchi-based disturbance with tournament selection to improve on MOPSO

In this paper, a Taguchi-based disturbance mechanism is deployed in our preliminary study on MOPSO. The proposed scenario includes tournament selection for global best solutions, jump-improved operation to expand the searching space, cluster operation to improve the diversity, and Taguchi based disturbance can enhance the searching ability and reduce the possibility of falling into local optima of particles. Experiments are conducted on seven two-objective benchmarks. The results show that the proposed method operates better than other algorithms in three performance metrics.