Improved Cooperative Group Search Optimization Based on Divide-and-Conquer Strategy

Evolutionary Computing (EC) approaches have been widely applied on optimization problems, given their flexibility and capabilities to deal with difficult environments. In this context, Group Search Optimizer (GSO) was proposed as a nature-inspired algorithm based on animal searching Behaviour and group living theory to solve continuous optimization problems. Cooperation has been applied successfully to improve the performance of population-based methods, such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO). In this paper, two new cooperative group search optimization models are presented, based on multiple GSO groups, employing divide-and-conquer strategies. Experiments were performed on 14 benchmark functions to evaluate the performance of the proposed algorithms in comparison to other well-known EC methods from literature. Experimental results showed that the proposed approaches are able to achieve better results than standard GSO in most of the test functions.