Rank charged system search algorithm for optimization and operations research.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Operations Research

Background:

• Complex optimization problems require efficient and robust algorithms.
• Existing charged system search (CSS) algorithms have limitations in balancing exploration and exploitation.

Purpose of the Study:

• Introduce CSSRank, an improved CSS algorithm for enhanced efficiency in complex optimization.
• Evaluate CSSRank's performance against existing methods and benchmark suites.
• Assess CSSRank's applicability to real-world clustering and reservoir operation optimization problems.

Main Methods:

• Developed CSSRank by integrating rank-based reduction selection and ranking-based mutation strategies.
• Tested CSSRank on standard benchmark functions, CEC 2014, and CEC 2024 suites.
• Applied CSSRank to UCI clustering datasets and reservoir operation optimization problems.

Main Results:

• CSSRank outperformed many existing methods on CEC 2014 and showed competitive results on CEC 2024.
• Achieved higher clustering accuracy and reliable objective values on UCI datasets.
• Provided superior engineering solutions for reservoir operation optimization, improving cost and efficiency.

Conclusions:

• CSSRank demonstrates effectiveness, versatility, and reliability in theoretical and practical optimization tasks.
• The algorithm offers a strong candidate for solving complex problems in optimization and operations research.
• CSSRank provides a robust and scalable solution for diverse optimization challenges.