Graphical timetables based on forecasted data
When analysing future infrastructure- and traffic changes, it’s important to consider the effect on trains’ running times. The running time depends on technical properties, but also on track congestion and the graphical timetable. In GRAPRO we investigate if optimization models can be used to generate graphical timetables from forecasted data.
Aim and goal
The purpose of the project is to investigate if and how mathematical optimization models and methods can be used construct graphical timetables based on forecasted data. If more graphical timetables can be produced by the analysts at the Swedish Transport Administration, the quality of impact assessments could be improved, which in the long run could lead to more beneficial infrastructure developments.
Currently the analysts at the Swedish Transport Administration have to make a graphical timetable by hand, or use a mathematical formula, to estimate the run time effects of an infrastructure change. Constructing a timetable by hand is time consuming, and the mathematical formula doesn’t capture certain combination effects. If mathematical optimization models and methods are to support the analyst, the optimization model must be able to handle certain aspects that are particularly important in a prognosis setting, such as e.g. large geographical areas, complex stations and "good" trains. It is also important to investigate how optimization could be used in situations with high forecasted capacity utilization.
The project will, in close cooperation with the Swedish Transport Administration, adapt and advance previously developed optimization models for timetable generation, and also investigate which solution methods that are suitable for use in a prognosis setting. The model should also be demonstrated in an example case.
If the project results show that optimization can be used to generate graphical timetables, then the method could be adopted by the analysts at the Swedish Transport Administration. This would improve the run time estimations for specific traffic and infrastructure changes, which in turn could improve the quality of impact assessments and in the long run lead to more beneficial infrastructure developments.