부산대학교 도서관

A well-designed timetable of a railway system not only saves the travel time for passengers but also reduces the operating cost of the railway system. However, planning a railway timetable is a complicated process with many tradeoffs and constraints, and thus it is challenging to solve the timetabling problem manually. In this paper, we target an optimization problem for the periodic timetable of a railway line. We integrate stop planning, service planning, and scheduling in a periodic timetabling problem and model it as a Mixed Integer Linear Programming (MILP) formulation to minimize the average travel delay of passengers. We then develop a genetic algorithm supported by a scheduling heuristic to solve the problem for better scalability and efficiency. A case study based on real-world data of the Taiwan High Speed Rail (THSR) shows that the developed algorithm efficiently reduces the average travel delay of passengers, compared with a multi-stage optimization approach and an existing timetable. The study also demonstrates the benefit of the integrated formulation.