A hybrid artificial bee colony algorithm with modified search model for numerical optimization

Artificial bee colony (ABC) is an effective optimization algorithm, which has been used in various practical applications. However, the standard ABC suffers from low accuracy of solutions and slow convergence rate. To address these issues, a hybrid ABC (called HABC) is proposed in this paper. In HABC, two improved strategies are utilized. First, a new search model is designed based on the best-of-random mutation scheme. Second, new solutions are generated by updating multiple dimensions. To verify the performance of HABC, twelve numerical optimization problems are tested in the experiments. Results of HABC are compared the standard ABC and two other improved ABC versions. The comparison show that our approach can effectively improve the optimization performance.