Neighbourhood-Inspired Grey Wolf Optimizer

Abstract

Over the past few decades, swarm intelligence (SI)-based algorithms are performing well in the optimization field. Grey wolf optimizer (GWO) is a new addition for the class of SI-based algorithms. It mimes the social behaviour of grey wolves. In GWO, a solution is updated by incorporating the information from the first fittest, second fittest and third fittest solution of the search space. Half of the generations in GWO are dedicated to exploration and the rest half are used for exploitation. To enhance the exploration capability of GWO, this article proposes a modified variant of GWO. In the proposed method, the searching process is guided by an arbitrarily selected solution to the search space. The proposed strategy is termed as neighbourhood-inspired grey wolf optimizer (NIGWO). To validate the performance of NIGWO, 16 benchmark functions are considered and the results obtained are compared with the basic version of GWO, particle swarm optimization (PSO), power law-based local search in spider monkey optimization (PLSMO) and differential evolution (DE) algorithm. The obtained results validate the proposed NIGWO approach.KeywordsSwarm intelligenceGrey wolf optimizerNature-inspired algorithmOptimization