Elephant Herding Optimization Using Multi-Search Strategy for Continuous Optimization Problems

The elephant herding optimization (EHO), which imitates social behaviors of the elephants, is recently proposed a swarm intelligence and population-based optimization algorithm. Although EHO is a good at local search, it is not effective on the global search due to the rapid loss of population diversity. In the basic EHO method, a single solution search equation is used for the generating the new individuals. Therefore, it is insufficient on the solving the problems which have different characteristics and the exploring the search space effectively. In this study, in order to overcome these problems and to provide a balance between exploration and exploitation, elephant herding optimization using multi-search strategy (Multi-EHO) has been proposed which inspired by the search strategies of the most well-known optimization techniques. For the comparison of the proposed method and the basic EHO, the CEC2015 benchmark set with 15 different functions is used. In addition, to validate the performance of Multi-EHO, the proposed method is compared with the grey wolf optimizer (GWO) and the whale optimization algorithm (WOA) proposed in recent years. Experimental results show that the proposed method has more successful and more robust performance than other methods.