Pbest-Guided Artificial Bee Colony Algorithm for Global numerical Function Optimization

Artificial Bee Colony (ABC) Algorithm is a population-based optimization algorithm, which has been shown to be superior to other evolution algorithm, such as differential evolution (DE) algorithm, particle swarm optimization (PSO) algorithm. However, there are still some shortcomings of the ABC algorithm in search strategies which is good at exploration and bad at exploitation. Inspired by DE, a new search strategy which some inferior solutions are considered is proposed to improve optimization performance of ABC algorithm. In this paper, the new improved ABC algorithm is proposed combining some inferior information with ABC algorithm, namely “ABC/current-to-pbest” algorithm. Simulation results show that ABC/current-to-pbest algorithm is effective in accelerating convergence while avoiding pre-mature especially solving multimodal problems when compared with other population-based algorithms according to a set of 24 test functions.