Particle Swarm Optimization with Average Individuals Distance-Incorporated Exploitation

Particle swarm optimization (PSO) is a popular optimization technique known for its simplicity and effectiveness. This paper introduces a variant that achieves a better balance between exploration and exploitation, named DiPSO. DiPSO incorporates a novel strategy based on trends in mean distance between individuals for local exploitation control. Experiments on 29 benchmark functions demonstrate that DiPSO consistently outperforms the state-of-the-art variant of PSO. Convergence analysis reveals that DiPSO achieves faster convergence and superior solutions. These results highlight the effectiveness of DiPSO in solving optimization problems.