Abstract
This paper proposes a novel nature-inspired algorithm, called the crocodile optimization algorithm (COA), which mimics the hunting strategies of crocodiles. Two important hunting processes of crocodiles are built, i.e., premeditation and waiting, during which the crocodile individuals gain and share information so that they can trace the prey; attacking and hunting, in this phase, crocodiles attacking and hunting their prey by implementing the “death roll” strategies. The search mechanisms of the proposed COA are differently compared to the existing methods inspired by the hunting behavior of crocodiles. The performance of the proposed COA is validated by utilizing twenty-nine standard test functions, including unimodal functions, multimodal functions, fixed-dimension multimodal functions, and composite functions, with qualitative and quantitative analysis, and its practical effectiveness in solving real-world problems is evaluated using five engineering optimization problems. The simulation results are compared with 2 algorithms also inspired by the hunting behavior of crocodiles and 9 other algorithms. The results and analysis suggest that COA is a competitive technique in handling unimodal, multimodal, and composite problems, and the Friedman ranking test statistical results revealed that COA is an excellent method for solving different kinds of complex problems. Finally, the outcomes of five engineering applications highlight the superiority and potential of COA in solving challenging real-world problems.