Train travel is considered to be a sustainable choice when going on a longer journey. For our society to fully reap the benefits of this means of transportation, careful resource planning is essential. For example, passengers need to be able to rely on the fact that trains are on time and train operators need to make efficient use of their vehicles and crew.

Control room, SJ. Photo credit SJ

Railways have an expensive infrastructure. The process of planning activities relevant to railways is highly complex and consists of many steps that have different planning horizons and involve various players. One part of this planning process is to schedule crew for trains that provide public railway transport. This research project therefore addresses the question how railway operators can handle disruptions and reschedule their crew efficiently. The long-term goal is to develop an optimisation-based decision support tool that can help planners to handle disruptions in a cost-efficient manner that also considers the schedules’ quality. 

Central station, Stockholm. Photo credit SJ

Efficient solution within a limited amount of time

Since disruptions are handled online, it is important that the decision support tool is able to propose a new, high-quality plan quickly. From an optimisation perspective, this means that the implemented methods need to offer an efficient solution within a limited amount of time. For this purpose, we intend to develop matheuristic methods that utilise machine learning to choose a suitable search space.


This project is conducted within KAJT (Capacity in the Railway Traffic System), a research programme for improved railway system performance.


SJ – a Swedish provider of public railway transport and owned by the Swedish government.

IVU – a German IT solutions provider for public transport and railway operators. They are active on a global scale and offer several optimisation-based decision support tools as part of their products.

TU Wien – The development of optimisation methods will be done in collaboration with Günther Raidl, in the Algorithms and Complexity group at TU Wien.

The Communications and Transport Systems (KTS) division at Linköping University – The interdisciplinary aspects of the project are addressed in dialogue with Anders Peterson at the Communications and Transport Systems division.