Dung Beetle with Reflection Cuckoo Catfish Optimizer for Numerical Optimization and Reservoir Production Optimization

As engineering systems grow in complexity, reliable metaheuristic optimizers are increasingly essential. While swarm intelligence algorithms are widely applied, recent approaches like the Cuckoo Catfish Optimizer (CCO) can experience premature convergence due to limited local exploitation and simplistic boundary handling. To address these limitations, this paper proposes the Dung Beetle with Reflection CCO (DBRCCO), integrating two principal mechanisms. First, an adaptive local search strategy inspired by dung beetle foraging is incorporated to intensify exploitation within dynamically contracting regions. Second, a momentum-preserving reflecting boundary mechanism replaces traditional clamping, maintaining population diversity near constraint edges. DBRCCO is evaluated against eight contemporary metaheuristic algorithms using the 29 CEC2017 benchmark functions and a reservoir production optimization problem. Statistical analyses indicate that DBRCCO achieves competitive performance, securing a Friedman ranking of 1.5172 (p<0.05). In the reservoir application, DBRCCO improves the mean Net Present Value (NPV) by 12.54% while reducing variance by over 72% relative to the standard CCO. These findings suggest that DBRCCO offers a stable and effective alternative for complex optimization tasks.