In the transport sector, the global trend is facing a major shift toward sustainability, efficiency, and technological advancement, with rail transport playing a crucial role in this trend. Consequently, railway systems worldwide are evolving, with Automatic Train Protection (ATP) systems becoming a key element in these transformations. Many rail networks focus on integrating ATP into their existing infrastructure or replacing their existing ATP systems with the new ones. However, the integration of trackside and onboard components in ATP, which must be compatible to function effectively, presents challenges in ensuring coordinated deployment while avoiding disruptions to railway operations. This thesis develops an optimization framework to generate ATP migration strategies. It particularly focuses on timeline alignment between ATP trackside and onboard installation, while incorporating key operational constraints and requirements to ensure its practical applicability. Non-dominated Sorting Genetic Algorithm II (NSGA-II) is integrated into the framework and customized to find near-optimal ATP migration strategies aimed at minimizing migration costs and duration. The framework is validated using case studies derived from Thailand's mainline network and from the literature to demonstrate its applicability. The results show that the developed framework successfully finds a trade-off that balances migration costs and duration while satisfying all constraints. In addition, solution refinement and parameter reduction mechanisms are embedded to enhance the solution quality and usability of the framework. The results show that the solutions obtained with these mechanisms outperformed those generated without them. This confirms the potential of the proposed framework as a practical decision support tool to generate ATP migration strategies.
Apitchaya Sujindaparn (Fri,) studied this question.