Crayfish Optimization Algorithm

This study aims to improve the performance of the Crayfish Optimization Algorithm (COA), a swarm intelligence algorithm recently introduced in the literature, on various test functions with fixed and variable dimensions. Optimization can be defined as making a system as efficient as possible at the least cost, within certain constraints. Numerous optimization algorithms have been designed in the literature to obtain the best solutions for specific problems. The most critical aspects in solving these problems are modeling the problem correctly, determining the parameters and constraints, and selecting an appropriate meta-heuristic algorithm for solving the objective function. Not every algorithm is suitable for every problem structure. While some algorithms solve fixed-dimension test functions better, others may perform better on variable-dimension test functions. In this study, the COA algorithm's performance was evaluated on 10 test functions previously used in the literature, consisting of three fixed-dimension functions (Schaffer Function, Himmelblau Function, Kowalik Function) and seven variable-dimension functions, including one unimodal (Elliptic Function) and six multimodal functions (Non-Continuous Rastrigin Function, Alpine Function, Levy Function, Weierstrass Function, Michalewicz Function, Dixon & Price Function). The solution values obtained for each of the selected functions were compared with the solutions obtained using the Harris Hawks Optimizer (HHO), the Charged System Search Algorithm (CSS), and the Backtracking Search Optimization Algorithm (BSA).