Population Control in Metaheuristic Algorithms: Can Fewer Be Better?

This chapter delves into the challenges faced by metaheuristic algorithms, particularly the inefficiencies arising from multiple search agents converging on similar solutions, thereby limiting exploration capabilities and escalating computational costs. Addressing this issue, the chapter introduces novel search operators integrated into the Differential Evolution (DE) algorithm. These operators facilitate dynamic adjustments to the number of search agents based on population diversity, enabling the algorithm to adapt its search behavior according to the exploration–exploitation balance required for optimal performance. The chapter presents empirical validations of the enhanced DE algorithm across 29 benchmark functions, including multimodal, unimodal, hybrid, and shifted functions, as well as real-world interplanetary trajectory design problems. Comparative analyses highlight the superior performance of the modified DE algorithm over existing state-of-the-art metaheuristic approaches.