A Comparative Study on Prominent Swarm Intelligence Methods for Function Optimization

Abstract

Optimization includes finding best available values of some objective function given a defined domain. Function optimization (FO) is the well-studied continuous optimization task which aim is to find best suited parameter values to get optimal value of a function. A number of techniques have been investigated in last few decades to solve FO and recently Swarm Intelligence (SI) methods, imitating power of the collective behavior of insects or animals, become popular to solve it. A number of SI methods have been developed on different time and tested on different test functions; therefore, it is important to compare the algorithms on a common test bench to identify their capability as well as best suited method for FO. The objective of this study is to draw a fair comparison among prominent SI methods in solving benchmark test functions. The SI methods considered in this study are Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC) optimization, Firefly Algorithm (FFA), Cuckoo Search Optimization (CSO), Group Search Optimization (GSO) and Grey Wolf Optimizer (GWO). Among the methods PSO is the pioneer and most popular in recent time; and GWO is the most recently developed method. The performance of the methods is compared in solving a suite of 22 well known benchmark test functions having different ranges, dimensions and types. Experimental results as well as analysis revealed that GWO is the overall best method among the SI methods and PSO is still promising to solve bench mark functions.