Which multiobjective evolutionary algorithm reigns supreme? This research provides a systematic comparison of various evolutionary approaches to multiobjective optimization. The study employs six carefully selected test functions, each designed with specific features that can cause difficulties in the evolutionary optimization process, primarily related to convergence toward the Pareto-optimal front, including multimodality and deception. The experimental results reveal a hierarchy among the algorithms and highlight the importance of elitism in enhancing evolutionary multiobjective search. These insights are crucial for researchers and practitioners seeking to select the most effective techniques for addressing complex optimization problems.
This comparative study of multiobjective evolutionary algorithms aligns perfectly with the scope of Evolutionary Computation, a journal dedicated to advancements in computational intelligence. By rigorously testing various algorithms, this research contributes to the journal's mission of advancing the field and providing valuable insights for researchers and practitioners.