Abstract
OptBees is an algorithm inspired by the processes of collective decision-making by bee colonies designed with the objective of generating and maintaining diversity, trading off exploitation (diversification) and exploration (intensification) and promoting a multimodal search, so that a broader coverage of promising regions of the search space can be achieved, allowing the determination of locally optimal solutions and/or multiple global optimal solutions. In this paper, the OptBees is presented in details and its performance is evaluated, in terms of global search, in all twenty-five minimization problems proposed for the Optimization Competition of Real Parameters of the CEC 2005 Special Session on Real-Parameter Optimization. The results obtained show that OptBees is competitive when the goal is just to obtain the best possible solution, without being necessary to determine locally optimal solutions and/or multiple global optimal solutions.