A New Bonobo optimizer (BO) for Real-Parameter optimization

This paper presents a new metaheuristic optimization algorithm, namely Bonobo optimizer (BO). It is inspired from the social behaviour and reproductive strategies of Bonobos. Bonobos adopt fission-fusion social strategy, which is nothing but forming several groups of variable sizes and compositions within a community and after a short period of time, they again reunite with their own community. Furthermore, bonobos show mainly four types of reproductive strategies, such as promiscuous, restrictive, consortship and extra-group mating. These natural behaviours of bonobos are artificially modelled in the proposed algorithm to solve optimization problems. The novelty of the proposed BO lies with the updating mechanisms of searching agents and their associated parameters, and the selection method of the mating partners. The performance of the proposed BO has been examined on a set of twenty optimization problems with varying attributes, and the results are compared with the reported ones by using seven other metaheuristic algorithms, in the literature. The outcome of the experiment clearly shows the superior performance of the proposed BO in both the aspects of exploration and exploitation capabilities compared to that of the others. The source code of BO is available at: https://sites.google.com/site/softcomputinglaboratory/Home.