EKEO: An Enhanced Kangaroo Escape Optimizer with Balanced Search for Global Optimization and Engineering Design.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Biomimetic Computing

Background:

• Kangaroo Escape Optimizer (KEO) is a biomimetic metaheuristic.
• KEO can suffer from premature convergence and loss of diversity in complex problems.

Purpose of the Study:

• To propose an Enhanced Kangaroo Escape Optimizer (EKEO).
• To address exploration-exploitation balance limitations in KEO.
• To improve performance on complex, multimodal, and constrained optimization problems.

Main Methods:

• Integration of Differential Evolution Mutation (DEM) for local exploitation.
• Integration of Quasi-Oppositional Learning (QOL) for diversity preservation and global exploration.
• Rigorous benchmarking on 23 classical functions, CEC 2019 suite, and four real-world engineering problems.

Main Results:

• EKEO consistently outperforms 11 state-of-the-art and classical metaheuristics.
• Demonstrated superior solution quality, convergence speed, and robustness.
• Validated practical applicability and constraint-handling effectiveness on engineering design problems.

Conclusions:

• The principled integration of DEM and QOL creates a robust optimization framework.
• EKEO offers a generalizable design principle for hybrid biomimetic metaheuristics.
• EKEO is a reliable and versatile tool for complex constrained engineering optimization.