A Comparison of Metaheuristic Algorithms for Structural Optimization: Performance and Efficiency Analysis

Optimization challenges effectively mirror numerous complex real-world problems. These are typically addressed using two methodologies: approximate methods and exact methods. Within the approximate category, meta-heuristic methods are prominent. These algorithms employ systematic principles to avoid local optima or prevent their occurrence altogether. The evolution of computing technology has spurred the development of various meta-heuristic algorithms designed for precise and expedient resolution of optimization issues. This paper conducts a comparative analysis of meta-heuristic algorithms applied to structural optimization. It encompasses 11 distinct meta-heuristic optimization algorithms: atomic orbital search algorithm, chaos game optimization, energy valley optimizer, firefly algorithm, genetic algorithm, gray wolf optimization, harmony search, material generation algorithm, particle swarm optimization, social network search, and whale optimization algorithm. The efficacy of these algorithms is assessed through their application to structural problems such as trusses, cantilever-cracked beams, welded beams, tubular columns, and I-shaped beams. Evaluation criteria include computational efficiency, accuracy, and statistical analysis based on reports among the 50 iterations. The findings confirm the effectiveness of the algorithms, with particle swarm optimization and social network search frequently delivering superior results across most problem sets.