Best Supported Emigrant Creation For Parallel Implementation Of Artificial Bee Colony Algorithm

Abstract

: Artificial Bee Colony algorithm inspired by the foraging behavior of real honey bees is one of the mostpopular swarm intelligence based optimization techniques. Like other population based evolutionary computationapproaches, Artificial Bee Colony algorithm is intrinsically suitable for distributed architectures. However,determining which food source should be chosen to distribute between subcolonies and communication topologyapplied still remain as an important problem for parallel implementations. In this study, new schema for increasingthe quality of the distributed source by changing best solution is presented. The proposed model is adapted to ringmigration topology and its eflectivenes is compared with conventional ring based topology in which best food sourcesin each subpopulation are distributed and the original sequential counterpart. Comparative results show that theproposed model increased the quality ofthe solutions and early convergence speed while protecting the speedup gain.