An Enhanced Spotted Hyena Optimization Algorithm and its Application to Engineering Design Scenario

Abstract

The Spotted Hyena Optimization (SHO) algorithm is inspired by simulating the predatory behavior of spotted hyenas. While the mathematical model of the SHO algorithm is simple and optimal, it is easy to fall into local optimization and causes premature convergence compared to some metaheuristic algorithms. To the end, we propose an enhanced Spotted Hyena Optimization algorithm, a hybrid SHO algorithm using Elite Opposition-Based Learning coupled with the Simplex Method called EOBL-SM-SHO. The EOBL-SM-SHO algorithm combines the characteristics of the simplex method’s geometric transformations (reflection, inside contraction, expansion, and outside contraction) with more practical information on elite opposition-based learning strategy. They can significantly strengthen the SHO algorithm’s search range and augment the hyena population’s diversity. Furthermore, we employ eleven benchmark functions and three engineering design issues to gauge the effectiveness of the EOBL-SM-SHO algorithm. Our extensive experimental results unveil that EOBL-SM-SHO achieves better accuracy and convergence rate than the state-of-the-art algorithms (e.g., Artificial Gorilla Troops Optimizer (GTO), Cuckoo Search (CS), Farmland Fertility Algorithm (FFA), Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), Spotted Hyena Optimizer (SHO)).