Application of Soft Computing Techniques in Modeling Train Delays

Train delays are the time between scheduled and actual arrival of the train. They have a great influence on the timetable and technological processes related to the train traffic. A model for calculating train delays can be used in the process of railway operations and timetable planning, and operational management. The model for train delays is based on the soft computing techniques. Neural Networks model and Adaptive Network-based Fuzzy Inference System model are trained and verified by the data collected from train dispatcher’s and infrastructure manager’s database. Model is tested on Rakovica station in Serbian Railways.