Neighborhood-search-based enhanced multi-strategy collaborative artificial Bee colony algorithm for constrained engineering optimization

Since artificial bee colony (ABC) algorithm, one of swarm intelligent algorithms, was proposed, it has shown good superiority in addressing optimization problems, and has attracted widespread attention because of its simple structure and good global optimization ability. However, ABC still has the shortcomings of slower convergence and poorer exploitation for complex practical problems. To overcome these limitations, an enhanced algorithm of multi-strategy collaboration based on neighborhood search called EMABC-NS is proposed. Firstly, the information of global optimal individual in the current population and individuals in the neighborhood are employed to the search phase of employed bees and onlooker bees, respectively. Secondly, the modification rate MR is introduced to randomly perturb all dimensions of the solutions. Finally, the search strategy of scout bees is enhanced by integrating current optimal solution and stochastic solution through MR. 23 well-established benchmark functions and 5 engineering optimization problems are utilized to validate the performance of EMABC-NS. The experimental result reveals that EMABC-NS is more competitiveness compared with other outstanding competitors, and it ranks first in the Friedman test. Compared with the other five algorithms, the proposed algorithm is also proved to be effective in solving practical engineering problems.