THE MODIFIED MARINE PREDATOR ALGORITHM FOR SOLVING OPTIMIZATION PROBLEMS (MMPA)

This study introduces the Modified Marine Predators Algorithm (MMPA) by incorporating a Dynamic Adjustment of Balancing Factor (R) to enhance its exploration and exploitation balance. The proposed factor is dynamically computed using a gradient-based equation that evolves with optimization. It is integrated into the transaction phase of the algorithm to guide search agents toward optimal solutions adaptively. The Modified Marine Predators Algorithm was evaluated on 23 benchmark functions, encompassing unimodal, multimodal, and composite landscapes, to assess its efficiency and robustness. Comparative results indicate that the modified algorithm outperforms the standard MPA and other state-of-the-art algorithms regarding convergence speed, solution precision, and the ability to avoid local optima. These findings demonstrate the effectiveness of the proposed modifications in solving complex optimization challenges.