Aim and research questions 

This paper aims to investigate railway punctuality in the three Swedish metropolitan regions of Stockholm, Gothenburg, and Malmö through macroscopic simulation. The results will contribute to a more detailed knowledge of delays and punctuality in the three biggest metropolitan regions in Sweden, which gives support to finding actions to decrease delays and improve punctuality. Moreover, the need for further research concerning how to decrease the amount of primary delays and increase punctuality will be identified. This paper seeks to answer the following research questions: 

• How do the primary and secondary delay distributions, respectively, differ between the three metropolitan regions?
• How much would the primary delays have to decrease to achieve 95% punctuality? 

As part of the research, we have formed three hypotheses which will be proved or disproved. The hypotheses are: 

H1: The more and denser the traffic in a region, the more secondary delays will occur. Thus, we expect the Stockholm region, which is the biggest, to have the largest amount of secondary delays followed by the Gothenburg region and, lastly, the Malmö region. 

H2: Entry delays, run-time delays, and dwell-time delays do not contribute equally to the total amount of delays. 

H3: The different types of delays vary in share between the regions, e.g. more dwell time delays in the Stockholm region and more run-time delays in the Gothenburg region. This is in line with findings from (Palmqvist and Kristoffersson, 2022). 

State of the art 

The Swedish railway has many challenges. There is a high demand for travel and transport and wishes to further increase the number of trains operated to meet this demand, but at the same time, there are problems with a lack of capacity and insufficient punctuality, where the goal of 95% of the trains being punctual at their final destination is not reached even with the current traffic volume. To run even more trains leads to even more secondary delays, and to expand the traffic with acceptable quality the amount of primary delays has to decrease. 

Previous studies addressing various aspects of punctuality and delays through simulation have been performed, e.g. simulation of the punctuality in Skåne with a current timetable and with a 2025 timetable with increased traffic (Johansson et al., 2022b; Palmqvist et al., 2023). (Johansson et al., 2022a) performed a case study on the Southern Main Line in Sweden comparing the total punctuality for the case when the freight trains were allowed to depart before schedule to the case when they were not allowed early departures. 

Method 

The railway operations are simulated with the simulation tool PROTON (Sipilä, 2023), which is a macroscopic tool, i.e. not modelling all details of the infrastructure, vehicles, and timetable. The advantage of macroscopic simulation is that larger networks can be simulated without significantly increasing the computational time. The macroscopic simulation will use different scaling factors for 

the respective entry, run-time, and dwell-time delays. Turning trains will be handled for better modelling of secondary delays during the full day, compared to previous simulation studies with PROTON. The timetable from 2019 will be used since it is relatively new and not affected by operational changes due to the COVID-19 pandemic. The metropolitan regions will be simulated separately, meaning that delays in one region will not affect the simulation outcome in the other regions. 

The “Design of experiments” (DOE) tool of Circumscribed Central Composite Design is used to find the 20 relevant delay distribution scenarios to study. Thereafter, these scenarios are simulated, and a regression model is solved to find the scenarios where the shares of run time, dwell, and entry delays most likely are primary delays. 

Results and analysis 

The results are expected to be useful in prioritising efforts to improve punctuality in the three metropolitan regions. In addition, the results will give a deeper knowledge of the similarities and differences between types of delays occurring per region, and the interaction effects between the different types of delays; entry, run-time, and dwell-time delays. 

References 

Johansson, I., Palmqvist, C.-W., Sipilä, H., Warg, J., Bohlin, M., 2022a. Microscopic and macroscopic simulation of early freight train departures. J. Rail Transp. Plan. Manag. 21. https://doi.org/10.1016/j.jrtpm.2022.100295 

Johansson, I., Sipilä, H., Palmqvist, C.-W., 2022b. Simulating the Punctuality Impacts of Early Freight Train Departures, in: Proceedings of 13th World Congress on Railway Research (WCRR). Birmingham, UK. 

Palmqvist, C.-W., Johansson, I., Sipilä, H., 2023. A method to separate primary and secondary train delays in past and future timetables using macroscopic simulation. Transp. Res. Interdiscip. Perspect. 17, 100747. https://doi.org/10.1016/j.trip.2022.100747 

Palmqvist, C.W., Kristoffersson, I., 2022. A Methodology for Monitoring Rail Punctuality Improvements. IEEE Open J. Intell. Transp. Syst. 3, 388–396. https://doi.org/10.1109/OJITS.2022.3172509 

Sipilä, H., 2023. Simulations with PROTON and RailSys: Use of a macroscopic and microscopic railway simulation tool in Swedish applications (No. TRITA–ABE–RPT–2323).