Performance Analysis of Four Metaheuristic Algorithms on Benchmark Functions

Various metaheuristic algorithms inspired by nature are used to solve optimization problems. With the increasing number of metaheuristics, their performance on problems is gradually improving. In this paper, the performance analysis of the newly proposed metaheuristics Artificial Rabbit Optimization Algorithm (ARO), African Vulture Optimization Algorithm (AVOA), Prairie Dog Optimization Algorithm (PDO) and the well-known Genetic Algorithm (GA) were performed for the first time. ARO is modeled after rabbits’ behavioral patterns, such as detour foraging and random hiding. AVOA is developed based on the navigation and competitive behaviors of African vultures. The newly proposed final metaheuristic PDO is inspired by the survival struggle of prairie dogs. As for the popular GA, it is based on survival of the fittest. Unimodal and multimodal test functions were used during the analysis. According to the simulation results, AVOA performed better and generated more successful results compared to the others 22 times in the mean and best values. AVOA was followed by PDO and ARO, proving that the newly proposed metaheuristics will be successful on different problems.