Comparison of different mutation strategies applied to artificial bee colony algorithm

摘要

Artificial bee colony (ABC) algorithm is a simple but powerful swarm intelligence optimization algorithm which was successfully applied to a number of problems. In this paper we propose a new approach for extending ABC algorithm based on five mutation strategies "borrowed" from differential evolution (DE) algorithm in order to improve the exploitation process. We compared five different strategies with original ABC algorithm on standard benchmark functions for various numbers of problem variables. The experimental results show that the modified ABC algorithms are effective and outperform the original algorithms in most cases.