Multi-Strategy Improved Pied Kingfisher Optimizer for Solving Constrained Optimization Problems.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• Constrained optimization problems (COPs) are prevalent in engineering and industry.
• COPs present challenges due to complex constraints and potential for local optima.
• Existing algorithms may struggle with the complexity and dimensionality of real-world COPs.

Purpose of the Study:

• To introduce a novel metaheuristic algorithm, the multi-strategy improved pied kingfisher optimizer (MSIPKO).
• To enhance the performance of the standard pied kingfisher optimizer (PKO) for constrained optimization.
• To provide an effective tool for solving complex, high-dimensional, and multi-constrained optimization tasks.

Main Methods:

• Incorporation of a reverse differential crossover for improved global exploration and diversity.
• Implementation of an enhanced diving-fishing operator for stronger local exploitation.
• Integration of an improved commensalism phase to enrich search directions and robustness.

Main Results:

• MSIPKO demonstrated superior optimization accuracy, convergence speed, and stability on benchmark functions and engineering problems.
• The algorithm outperformed several state-of-the-art methods, especially on high-dimensional, nonlinear, and multi-constrained problems.
• MSIPKO achieved competitive or better solutions with fewer function evaluations, highlighting its efficiency.

Conclusions:

• MSIPKO is a robust and efficient metaheuristic for tackling complex constrained optimization problems.
• The proposed enhancements significantly improve the capabilities of the original PKO.
• The algorithm shows promise for real-world applications and future extensions to multi-objective and large-scale optimization.