Pastoralist Optimization Algorithm: A Novel Nature-Inspired Metaheuristic Optimization Algorithm

Abstract

This paper proposes the development of a novel optimization algorithm called the Pastoralist Optimization Algorithm (POA) inspired by the pastoralists herding strategies. The strategies are scouting, camp selection, camping, herd splitting and merging. These strategies were modelled mathematically and used to develop the POA. The performance of the algorithm was evaluated by testing the algorithm on 10 unimodal and multimodal test benchmark functions. This is to measure the algorithm exploitative, explorative, convergence speed as well as the ability to escape being trapped in a local optimum solution. Also, a nonparametric statistical test (Wilcoxon rank sum tests) was carried out to ascertain the statistical significance level of the proposed algorithm results. The experimental results obtained show that the algorithm is very competitive and obtain better results in most cases when compared with some similar existing state-of-the-art nature-inspired metaheuristic optimization algorithms. Also, it is statistical proven that the results are very significant.