Title: Mutual Operator Patterns in Engineering Metaheuristics: A Review of Classical Engineering Design Optimization Problems
Authors: Iman Mousa Shaheed, Mustafa Kadhim Taqi,
Volume: 10
Issue: 6
Pages: 104-117
Publication Date: 2026/06/28
Abstract:
Optimization problems arising from engineering design applications exhibit nonlinear objective functions, multimodal search spaces, interactivity among variables, and tight constraints on feasible solutions; all of which have made classical approaches to deterministic optimization infeasible. Consequently, metaheuristic algorithms have emerged as effective means for tackling classic engineering benchmark problems such as pressure vessel design, tension and compression spring design, welded beam design, and speed reducer design. However, recent developments in the fast-growing field of hybrid metaheuristics have been plagued with excessive development of optimization algorithms whose novel methodology revolves merely around the search metaphor used. In this context, the present paper conducts a review of metaheuristics for engineering design optimization with emphasis on hybridization at the operator level. The review critically evaluates the use of main operator types that make up the structure of today's engineering optimizers, including: exploration operators, exploitation operators, constraint handling operators, repair operators, diversity preserving operators, and adaptative control operators. Additionally, common operator usage patterns observed in successful metaheuristics are identified, such as: cooperation between exploration and exploitation operators, diversified search with repair operators, chaotic initialization, Lévy flight perturbation, memetic refinement, and adaptive operator sequencing. Also, the study covers new trends regarding hyper-heuristics, reinforcement learning guided operators' selection, and adaptive operator ecosystems. The findings of the analysis indicate that operator cooperation, compatibility and adaptiveness, not the metaheuristic algorithm's metaphor, play a predominant role in achieving optimal results. Lastly, several challenges concerning open questions and directions for future research in this area are discussed, such as those relating to operator compatibility theory, explainability, benchmarking inconsistency, scalability issues, and autonomous optimization systems.