Order based emigrant creation strategy for parallel artificial bee colony algorithm

Abstract

Artificial Bee Colony (ABC) algorithm inspired by the foraging behaviors of real honey bees is one of the most important swarm intelligence based optimization algorithms. When considering the robust and phase divided structure of the ABC algorithm, it is clearly seen that ABC algorithm is intrinsically suitable for parallelization. In this paper, we proposed a new emigrant creation strategy for parallel ABC algorithm. The proposed model named order based emigrant creation strategy depends on sending best food source in a subcolony after modifying it with another food source chosen sequentially from the same subcolony at each migration time. Experimental studies on a set of numerical benchmark functions showed that parallel ABC algorithm powered by the newly proposed model significantly improved quality of the final solutions and convergence performance when compared with standard serial ABC algorithm and parallel ABC algorithm for which the best food sources in the subcolonies directly used as emigrants.