The Performance and Sensitivity of the Parameters Setting on the Best-so-far ABC

摘要

Artificial Bee Colony (ABC) is a metaheuristic technique in which a colony of artificial bees cooperates in finding good solutions in optimal search space. The algorithm is one of the Swarm Intelligence algorithms explored in recent literature. However, ABC can sometimes be a slow technique to converge. In order to improve its performance the modified version of ABC called Best-so-far ABC were proposed. The results demonstrated that the Best-so-far ABC can produce higher quality solutions with faster convergence than either the original ABC or the current state-of-the-art ABC-based algorithm. In this work, we aim to extend the performance analysis of the Best-so-far ABC algorithm by investigating the effect of each proposed modification to the overall performance as well as to present the sensitivity of the parameters setting on the algorithm.