Crowding-distance-based multi-objective particle swarm optimization

Abstract

Multi-objective optimization methods are essential to resolve real-world problems as most involve several types of objects. For this solution several multi-objective genetic algorithms are proposed. This paper presents a Crowding-distance-based Multi-objective Particle Swarm Optimization (CMPSO) algorithm. According to the size of archive members' crowding-distance, the algorithm selects the global optimal position in the archive for each particle on the basis of Roulette Gambling and maintains external archives based on crowding distance. Finally, three tests are conducted to evaluate this algorithm. The experiment results show that CMPSO has better ability to continuously optimize the performance, shorter running time, better convergence and robustness, compared with strength Pareto EA (SPEA2) and some other common algorithms. © 2010 Springer-Verlag.