Randomized grey wolf optimizer (RGWO) with randomly weighted coefficients

Meta-heuristic algorithms are quiet widely used in engineering and science. In this paper, a novel variant of grey-wolf optimizer, randomized gray-wolf optimizer, is proposed where the effect of random coefficients of the best agents are taken into consideration. The proposed variant is tested on a set of 12 standard benchmark functions and compared with grey-wolf optimizer and particle swarm optimization on different dimensional search spaces. To statistically validate and analyse the convergence behaviour of variant, wilcoxon rank sum test along with convergence graphs over 30 runs is studied. The experimental study signify that the proposed variant not only enhances the performance but has comparable convergence rate, especially in case of multi-modal problems.