An artificial algae algorithm with stigmergic behavior for binary optimization

Abstract In this study, we focus on modification of the artificial algae algorithm (AAA), proposed for solving continuous optimization problems, for binary optimization problems by using exclusive-or (xor) logic operator and stigmergic behavior. In the algorithm, there are four processes sequentially realized for solving continuous problems. In the binary version of the algorithm, three of them are adapted in order to overcome the structure of binary optimization problems. In the initialization, the colonies of AAA are set to either zero or one with equal probability. Secondly, helical movement phase is used for obtaining candidate solutions and in this phase, the xor operator and stigmergic behavior are utilized for obtaining binary candidate solutions. The last modified phase is adaptation, and randomly selected binary values in the most starved solution are likened to biggest colony obtained so far. The proposed algorithm is applied to solve well-known uncapacitated facility location problems and numeric benchmark problems. Obtained results are compared with state-of-art algorithms in swarm intelligence and evolutionary computation field. Experimental results show that the proposed algorithm is superior to other techniques in terms of solution quality, convergence characteristics and robustness.