Dynamic Traffic Management in Railway Traffic Networks

Abstract — Current practice in the operational-level management of railway traffic networks is mostly based on predefined rules and on the ability of traffic controllers and train dispatchers to detect and avoid conflicting situations. However, train operations are affected by many varying factors, such as driver behavior, passenger volumes, weather conditions, etc. These factors are results in delays. Delays in railway traffic networks include initial, primary, consecutive, knock- on delays. These delays can be partly absorbed by a stable and robust timetable. Timetable stability is rapidly gaining attention due to the increasingly saturated European railway infrastructure, where a slightly delayed train may cause a domino effect of consecutive delays over the entire network. The essential structure of railway traffic operating under a periodic railway timetable can be modeled as a linear system in max-plus algebra. Max-plus-linear systems can be characterized as discrete event systems in which only synchronization and no concurrency or choice occurs. In this paper we present a controlled railway system model using the switching max-plus-linear system. If the model is affine in the controls, the optimization problem can be recast as a mixedinteger linear programming problem.