Horse herd optimization algorithm: A nature-inspired algorithm for high-dimensional optimization problems

This paper proposes a new meta-heuristic algorithm inspired by horses' herding behavior for high-dimensional optimization problems. This method, called the Horse herd Optimization Algorithm (HOA), imitates the social performances of horses at different ages using six important features: grazing, hierarchy, sociability, imitation, defense mechanism and roam. The HOA algorithm is created based on these behaviors, which has not existed in the history of studies so far. A sensitivity analysis is also performed to obtain the best values of coefficients used in the algorithm. HOA has a very good performance in solving complex problems in high dimensions, due to the large number of control parameters based on the behavior of horses at different ages. The proposed algorithm is compared with popular nature-inspired optimization algorithms, including grasshopper optimization algorithm (GOA), sine cosine algorithm (SCA), multi-verse optimizer (MVO), moth-flame optimizer (MFO), dragonfly algorithm (DA), and grey wolf optimizer (GWO). Solving several high-dimensional benchmark functions (up to 10,000 dimensions) shows that the proposed algorithm is highly efficient for high-dimensional global optimization problems. The HOA algorithm also outperforms the mentioned popular optimization algorithms for the case of accuracy and efficiency with lowest computational cost and complexity. (C) 2020 Elsevier B.V. All rights reserved.

Automated summary

This paper introduces a new meta-heuristic algorithm called Horse Herd Optimization Algorithm (HOA), inspired by horses' behavior, which shows excellent performance in high-dimensional optimization problems. By imitating the behavior features of horses at different ages, HOA has a large number of control parameters leading to efficient solving of complex problems.