Wildebeest herd optimization: A new global optimization algorithm inspired by wildebeest herding behaviour

This paper proposes a new metaheuristic global optimization algorithm inspired by Wildebeest herding behavior called Wildebeest Herd Optimization (WHO) algorithm. WHO algorithm mimics the way nomadic Wildebeest herds search vast areas of grasslands efficiently for regions of high food density. The WHO algorithm models five principal Wildebeest behaviors: firstly Wildebeests have limited eyesight and can only search for food locally, secondly Wildebeests stick to the herd to escape predators, thirdly Wildebeest herd as a whole migrates to regions of high food availability based on historical knowledge of annual grass growth rates and rainfall patterns, fourthly Wildebeests move out of crowded overgrazed regions and finally Wildebeests move to avoid starvation. The WHO algorithm is compared to Physics inspired, Swarm based, Biologically inspired and Evolution inspired global optimization algorithms on an extended test suite of benchmark optimization problems including rotated, shifted, noisy and high dimensional problems. Extensive simulation results indicate that the WHO algorithm proposed in this paper significantly outperforms state-of-the-art popular metaheuristic optimization algorithms like Particle Swarm Optimization Algorithm (PSO), Genetic Algorithm (GA), Gravitational Search Algorithm (GSA), Artificial Bee Colony Algorithm (ABC) and Simulated Annealing (SA) on shifted, high dimensional and large search range problems.