Adaptive predator prey algorithm for many objective optimization.

Area of Science:

• Computational intelligence
• Optimization algorithms
• Engineering design

Background:

• Many-objective optimization presents challenges in balancing solution diversity and convergence, especially in high-dimensional spaces.
• Existing metaheuristic methods often struggle with complex, conflicting objectives in real-world engineering problems.

Purpose of the Study:

• To introduce the Many-Objective Marine Predator Algorithm (MaOMPA), an enhanced version of the Marine Predators Algorithm (MPA).
• To address the limitations of traditional algorithms in many-objective optimization tasks.

Main Methods:

• MaOMPA integrates elitist, non-dominated sorting, and crowding distance mechanisms.
• It utilizes a predator-prey interaction model to balance exploration and exploitation.
• The algorithm was evaluated on benchmark test functions and real-world engineering problems.

Main Results:

• MaOMPA demonstrated superior performance compared to state-of-the-art algorithms.
• Evaluations used generational distance, hypervolume, and coverage metrics.
• The algorithm achieved a robust balance between exploration and exploitation.

Conclusions:

• MaOMPA is an effective and reliable tool for engineers and researchers tackling complex optimization problems.
• The proposed algorithm shows significant improvements in handling many-objective optimization scenarios.
• Source code is publicly available for further research and application.