Train drivers are in short supply – all over Europe this is slowing down the desired growth of the sustainable mode of rail transport.
Manual dispatching processes lead to major challenges in complex rail operations and, in addition to the general lack of skilled workers, contribute to a shortage of driving personnel.
This requires a high degree of flexibility in day-to-day operations, and individual preferences and lifestyles can hardly be matched in the planning process.
This is where the WILSON-LEARN project comes into play. Started in February 2020 by the partners Menlo79 GmbH, Havelländische Eisenbahn AG, SCI Verkehr GmbH, and Zuse Institute Berlin, it aims at using a machine learning algorithm to significantly increase the efficiency of operational personnel dispatch in rail freight transport.
The project is funded by the Federal Ministry of Transport and Digital Infrastructure (BMVI) with around 750,000 euros as part of the program “Future Rail Freight Transport”.
As part of the program “Future Rail Freight Transport”, the Federal Ministry of Transport and Digital Infrastructure promotes the testing and market introduction of innovations with the aim of increasing the profitability and capability of rail freight transport.
A machine learning algorithm is to be implemented and tested, which generates valid real-time suggestions for efficient, operational personnel dispatching in rail freight transport. This enables an increase in personnel productivity and counteracts staff shortages.
On the basis of successful dispatching decisions in the past, the algorithm learns which factors are decisive for sustainable dispatching and can thus intelligently support the dispatching of time-consuming processes.
Machine learning algorithms are particularly suitable for such a challenge because, compared to static optimizers, they are able to learn behavior and handle it more realistically.