Performance comparison of Genetic Algorithm, Differential Evolution and Particle Swarm Optimization towards benchmark functions

Abstract

Genetic algorithm (GA), Differential Evolution (DE) and Particle Swarm Optimization (PSO) are always implemented to solve different kinds of complex optimization problems. Each method contains its own advantages and the performance varies based on different case studies. There are many Soft Computing (SC) methods which can generate different result for the same optimization problems. However, no exact result is produced because random function is usually applied in SC methods. The performance maybe is affected by the parameter setting or operations inside each method. Therefore, the motivation of this paper is to compare the performance of GA, DE and PSO by using the same parameters setting and optimization problems. The experiments can prove that although same parameters setting are applied, but different fitness and time can be obtained. Based on the result, GA was proven to perform better compared to DE and PSO in obtaining highest number of best minimum fitness and faster than both methods.