Crocodile Ambush Optimization Algorithm: A new bio-inspired metaheuristic algorithm for solving optimization problems
Xinpeng Xu
Abstract
This paper presents the Crocodile Ambush Optimization Algorithm (CAOA), a novel metaheuristic inspired by the energy-saving and ambush hunting behavior of crocodiles. CAOA integrates adaptive energy decay modeling, stochastic leader selection, and threshold-based solution reinitialization to achieve a balanced trade-off between exploration and exploitation. The algorithm is evaluated on a subset of 29 representative benchmark functions selected from the CEC 2017 test suite under various dimensional settings (30, 50, 100), and its performance is compared against several classical and modern algorithms, including PSO, GWO, DE, SCA, WOA, SSA, HHO, and HGS. Experimental results show that CAOA achieves superior or competitive convergence accuracy across unimodal, multimodal, and composite functions, while maintaining a lower average runtime. Furthermore, CAOA is applied to two real-world constrained engineering design problems, where it successfully identifies feasible, high-quality solutions under nonlinear constraints. These results demonstrate that CAOA is a robust, efficient, and adaptable optimizer for both benchmark and real-world problems.