A brief traffic jam, a stuck door, or a large number of passengers getting on and off at a stop – even minor delays in train and bus schedules can cause major problems. A new artificial intelligence (AI) could aid in the design of schedules that are less vulnerable to minor disruptions.
A team from Martin Luther University Halle-Wittenberg (MLU), the Fraunhofer Institute for Industrial Mathematics ITWM, and the University of Kaiserslautern developed it. The publishing of these findings is available in “Transportation Research Part C: Emerging Technologies.”
The team was looking for a quick and easy way to see how well timetables can compensate for minor, unavoidable disruptions and delays. This is known as robustness in technical terms. Until now, such timetable optimizations have needed complex computer simulations that calculate the routes of a large number of passengers under various scenarios.
A single simulation can easily take several minutes to run. To optimize timetables, however, many thousands of such simulations are required. Our new method allows us to estimate the robustness of a timetable within milliseconds, says Professor Matthias Müller-Hannemann of MLU’s Institute of Computer Science.
For training their artificial intelligence, the researchers from Halle and Kaiserslautern used a variety of methods for evaluating timetables. The team put the new AI to the test with timetables for Göttingen and parts of southern Lower Saxony and got very good results.
Müller-Hannemann claims that delays cannot be avoided. They occur, for instance, when there is a traffic jam during rush hour, when a train door jam, or when a particularly large number of passengers board or disembark at a stop.
When transfers are tightly scheduled, even a few minutes of delay can cause passengers to miss their connections. In the worst-case scenario, says co-author Ralf Rückert, they miss the last connection of the day. Another effect is that vehicle rotations can be disrupted, causing subsequent journeys to begin late, and the problem continues to worsen.
There are only a few options for preventing such delays ahead of time: Travel times between stops and wait times at stops could be calculated more generously, and larger time buffers at terminal stops and between subsequent trips could be planned. All of this, however, comes at the expense of economic efficiency.
The new method may now be used to optimize timetables for achieving a very good balance between passenger needs, like quick connections and few transfers, timetable robustness against disruptions, and the external economic conditions of the transportation companies.
The Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) funded the study as part of the research unit “Integrated Planning for Public Transport.”
Source link
