A Revised Particle Swarm Optimization Approach for Multi-objective and Multi-constraint Optimization

Abstract. Many real world design or decision-making problems involve si-multaneous optimization of multiple objectives, while satisfying multiple con-straints. In this paper, some novel adaptations were given to the recent bio-inspired optimization approach, Particle Swarm Optimization (PSO), to form a suitable algorithm for these multi-objective and multi-constraint optimization problems. Divided Range Multiobjective Particle Swarm Optimization (DRMPSO) was presented, extending PSO for distributed computing. Inspired by the biological phenomenon of symbiosis, a problem-independent constraint handling technique was created, by introducing symbiosis mechanism to PSO, to deal with the multiple constraints. The proposed algorithm was tested on three benchmark problems, comparing with two other approaches in an effi-cient comparison form.