Multiobjective Atomic Orbital Search (MOAOS) for Global and Engineering Design Optimization

In the real world, many optimization problems have such levels of uncertainty and complexity that a single objective function cannot represent all the characteristics of the considered system. Hence, multi-objective optimization algorithms are needed to account for multiple aspects of the problem, represented by multiple objective functions, to achieve reasonable and useful results through the optimization procedure. In this paper, we introduce the multi-objective version of a recently-developed single-objective metaheuristic algorithm known as Atomic Orbital Search (AOS), which will be called Multi-Objective Atomic Orbital Search (MOAOS). To this end, the general aspects and main searching loop of the AOS algorithm are modified to make it capable of dealing with problems with multiple objectives. For the performance evaluation of this algorithm, the mathematical benchmark problems ZDT and DTLZ, alongside several real-world engineering design problems and the CEC- 2020 MMO test problems, are utilized. Based on the results obtained in this study, we can conclude that MOAOS is capable of producing either superior or closely comparable results when evaluated in competition with alternative state-of-the-art metaheuristic methods.