Starfish Optimization, Sand Cat Swarm Optimization, Weighted Average and Mirage Search Optimization Algorithms for Finding Global Optimal Solutions

Abstract

In this study, the performance of four contemporary meta-heuristic techniques—namely Starfish Optimization, Sandcat Swarm Optimization, Weighted Average Algorithm (WAA), and Mirage Search Optimization—is investigated across diverse optimization challenges. The central goal is to execute a rigorous, unbiased comparative analysis to identify the most proficient optimizer among them. To ensure a fair benchmark, each method was configured with identical population sizes and iteration limits across five distinct problems. Furthermore, to guarantee statistical reliability, 50 independent trials were conducted for every algorithm. The results achieved by the four applied algorithms are compared to each other using different criteria, including the Minimum Fitness, Average Fitness, Maximum Fitness, and Standard Deviation. The comparison of the four criteria's values across the four algorithms reveals that WAA completely outperforms the others in all aspects, especially in the minimum fitness value and the standard deviation index. Moreover, WAA demonstrates superior stability and faster convergence during the search phase. Consequently, WAA is identified as the most effective solution and is highly endorsed for addressing the optimization tasks explored in this work.