Performance Analysis of Current Multi-Objective Metaheuristic Optimization Algorithms for Unconstrained Problems

Abstract

Multi-objective optimization is a method used to produce suitable solutions for problems with more than one Objective. Various multi-objective optimization algorithms have been developed to apply this method to problems. In multi-objective optimization algorithms, the pareto optimal method is used to find the appropriate solution set over the problems. In the Pareto optimal method, the Pareto optimal set, which consists of the solutions reached by the multi-objective optimization, includes all the best solutions of the problems in certain intervals. For this reason, the Pareto optimal method is a very effective method to find the closest value to the optimum. In this study, the Multi-Objective Golden Sine Algorithm we developed (MOGoldSA), the recently published Multi-Objective Artificial Hummingbird Algorithm (MOAHA), and the Non-Dominant Sequencing Genetic Algorithm II (NSGA-II), which has an important place among the multi-objective optimization algorithms in the literature, are discussed. In order to see the performance of the algorithms on unconstrained comparison functions and engineering problems, performance comparisons were made on performance metrics