Artificial bee colony directive for continuous optimization

Abstract

The artificial bee colony (ABC) algorithm, a relatively new swarm intelligence optimization technique, has been shown to be a competitive alternative to other population-based algorithms. This paper fundamentally modifies the solution search equations of the ABC in a manner that sends bee agents in search of three types of search regions that improve convergence speeds and proposes an innovative artificial bee colony directive (ABCD) algorithm. Moreover, this paper validates the ABCD algorithm by showing better performance by improving two familiar ABC variants in experimental tests. In addition, 10 applicable search strategies that adopt the proposed three search-region types are presented. The proposed ABCD not only improves the original ABC and its subsequently improved versions but is also useful for setting the search regions of other swarm intelligence algorithms.