Abstract
Synchronized timetables are crucial for public transportation, yet optimizing them remains complex. While Genetic Algorithms (GAs) address this challenge effectively, their success depends on parameter tuning. In this study was examined how the GA parameters affect schedule synchronization, using a stochastic simulation model to evaluate the fitness of alternative solutions. It is found that the larger population sizes improve solution quality, and higher crossover probabilities enhance overall performance. Conversely, increasing mutation rates or the number of attempted mutations reduces the algorithm’s ability to converge on optimal timetables. Those findings provide researchers and practitioners with data-driven insights for efficiently deploying GAs in public transit optimization.