An Attractor-Based Multiobjective Particle Swarm Optimization

This paper presents a new particle swarm algorithm for multiobjective optimization. The approach integrates particle displacement based on local attractors. This adapts evolutionary techniques such as mutation and crossover operators and elitism strategies based on crowding distance and random global guides selection. The proposed approach is implemented on a set of well-known multiobjective optimization test-functions and compared with other multiobjective optimization tools in the literature. The obtained results demonstrate its ability to achieve the main optimization solution and a well-distributed Pareto front. © 2016, Springer India Pvt. Ltd.