Performance Comparison of Recent Metaheuristic Algorithms on Engineering Design Optimization Problems

Abstract

Metaheuristic algorithms have been extensively applied in a variety of complex engineering design optimization problems (EDOPs) due to their capability of yielding near-optimal solutions without excessive computational times. The aim of this study is to investigate the performance comparison among seven novel metaheuristic optimization algorithms: Artificial Hummingbird Algorithm (AHA), Artificial Protozoa Optimizer (APO), African Vultures Optimization Algorithm (AVOA), Electric Eel Foraging Optimization (EEFO), Mountain Gazelle Optimizer (MGO), Pied Kingfisher Optimizer (PKO), and Quadratic Interpolation Optimization (QIO). This comparison is performed with twelve engineering design optimization problems evaluating the best, worst, mean, and standard deviation of their results. We also use non-parametric statistical tests such as the Friedman rank test and Wilcoxon signed rank test to finally compare the performance of algorithms. The results show the merits and demerits of each algorithm, which give us clues on their suitability for different engineering design problems. According to Friedman rank test, EEFO surpasses the other algorithms in these EDOPs. In addition, it performs statistically better than AVOA and QIO according to Wilcoxon signed rank test.