Train Timetabling Model for Double-Tracks; Stochastic Programming Approach

Railway systems just like any other systems require optimized and exact planning to use maximum system capacity. Train timetabling is one the most important programs in railway systems that has a great deal of impact on other operations and programs. In years, there have been a lot of researchers who have been studying about this subject. Meanwhile, over recent years and because of the railway systems getting more complicated, using of uncertainty models are more focused by researchers. Stochastic programming is one the oldest current approaches for considering the uncertainty between all other techniques. In fact stochastic programming intends to realize models and results by considering stochastic parameters in real world. After studying the current models in the literature, we presented a two stage stochastic program according to Iranian railways conditions and constraints. The model intends to minimize trip times of trains at the first stage and minimize delays which occur over segment running time fluctuations at the second stage. The model has been implemented on one of the double tracks in Iranian railways, and the result has been compared with the current time table at the end. The results show that it still is necessary to add slack times in current time tables to minimize delays over segment running time fluctuations.