Improved multi-strategy artificial bee colony algorithm

Abstract

Artificial bee colony ABC algorithm is a nature-inspired metaheuristic based on imitating the foraging behaviour of bee, which is widely used in solving complex multi-dimensional optimisation problems. In order to overcome the drawbacks of standard ABC, such as slow convergence and low solution accuracy, we propose an improved multi-strategy artificial bee colony algorithm MSABC. According to the type of position information in ABC, three basic search mechanisms are summarised, the mechanisms include searching around the individual, the random neighbour and the global best solution. Then, the basic search mechanisms are improved to obtain three search strategies. Each bee randomly selects a search strategy to produce a candidate solution under the same probability in each iteration. Thus these strategies can make a good balance between exploration and exploitation. Finally, the experiments are conducted on eight classical functions. Results show that our algorithm performs significantly better than several recently proposed similar algorithms in terms of the convergence speed and solution accuracy.