Many-Objective Grey Wolf Optimizer (MaOGWO) for solving real-world problems

• Introduces MaOGWO, an algorithm for complex optimization problems. • MaOGWO outperforms NSGA-III and MaOPSO in benchmarks and real-world tests. • Innovative techniques in MaOGWO balance convergence and diversity. • Empirical analysis shows MaOGWO's effectiveness in various scenarios. • MaOGWO's verified on diverse mathematical and real-world problems. The Grey Wolf Optimizer (GWO) is an effective optimization algorithm that primarily focuses on single objective optimization and is characterized by its simplicity and fast convergence. However, adapting GWO for many-objective optimization requires addressing two critical challenges: keeping the solutions in sync while also promoting variety. In this study, a new Many-Objective Grey Wolf Optimizer (MaOGWO) is proposed which includes reference point strategies, niche preservation and information feedback to enhance convergence and diversity. A comprehensive comparative analysis was conducted to evaluate MaOGWO against four leading many-objective optimization algorithms: Some of them are Non-dominated Sorting Genetic Algorithm III (NSGA-III), Many-Objective Particle Swarm Optimizer (MaOPSO), Many-Objective Teaching Learning-Based Optimizer (MaOTLBO) and Many-Objective Gradient-Based Optimizer (MaOGBO). We evaluated the performance of these algorithms on the DTLZ1-DTLZ7 problem sets with 5, 10 and 15 objectives and five real-world many-objective optimization problems, namely RWMaOP1-RWMaOP5. The quality assessment metrics used in this research were generational distance (GD), inverted generational distance (IGD), spacing (SP), spread (SD), hypervolume (HV) and runtime (RT). The experimental outcomes revealed that the MaOGWO dominated the other algorithms in terms of convergence to the Pareto-optimal front and solution diversity for both synthetic and practical problems with lower computational costs.