A hybrid swarm intelligence optimizer based on particles and artificial bees for high-dimensional search spaces

Real-world problems can present search spaces with hundreds of dimensions and swarm intelligence algorithms have been developed to solve this type of problems. Particle Swarm Optimization (PSO) presents a fast convergence in continuous problems, but it can not maintain diversity along the search process. On the other hand, Artificial Bee Colony (ABC) presents the capability to generate diversity when the guide bees are in the exploration mode. We propose in this paper to introduce a mechanism based on the ABC to generate diversity in an adaptive PSO approach and analyze its performance in high dimensional search spaces. The swarm switches its behavior depending on the dispersion of the swarm. We evaluated our proposal in a well known set of 20 benchmark functions recently proposed in 2010 and our proposal achieved better performance than PSO, APSO and ABC in most of the cases.