Key points are not available for this paper at this time.
The performance of genetic algorithms is significantly influenced by their parameters.However, parameters are traditionally determined based on the designer's empirical knowledge.While experienced designers may quickly find effective parameter combinations, others often need extensive experimentation.This study addresses these issues by proposing an efficient parameter optimization method leveraging the computational power of genetic algorithms.Multiple data and Pareto sets ensure that selected parameters work efficiently across diverse data.Our proposed parameter optimization method is applied to a genetic algorithm to find the optimal search area for search and rescue units (SRUs).We compare our proposed method to genetic algorithms that use optimized combinations of parameters and to genetic algorithms from previous studies.As a result, the proposed approach outperforms while reducing solution search time by approximately 65%.In addition, the genetic algorithm that underwent parameter optimization using three datasets performed better than the one that underwent parameter optimization using only one dataset.These experimental results indicate that our proposed method effectively optimizes parameters.Our parameter optimization method can be applied to the various studies of genetic algorithms.This implies that we have enhanced the deployment of SRU search areas and can anticipate performance improvements for other problems.
Hong et al. (Thu,) studied this question.