Skip to main content
Back to KTH start page

Travel pattern analysis from public transport smart cards

Case description

The Division of Transport Planning, KTH, has collaborated with the Region Stockholm Public Transport Administration in a sequence of projects to develop and apply analysis methods for automated public transport data, especially ticket validations (the Access system), passenger counts, vehicle positions and timetables. Data and methods cover all major modes of public transport in Stockholm: bus, commuter train, metro, tram and light rail.

The Access data is a particularly rich source of information on users’ travel patterns. The data contain every ”tap-in” in the Stockholm public transport system (ca 1.2 million data points per day), including the time, location, ticket type and anonymized card id. Prior to our projects, the Access data were not used for analysis to any great extent by the Public Transport Administration due to a lack of tools for extracting the information. In the project FairAccess (2018-2019) we developed and applied a series of algorithms to infer tap-out locations, vehicles, travel times and transfers. Thereafter, we constructed time-dependent origin-destination matrices for which segmentations can be performed with respect to geographical and user product features. We demonstrated the approach and algorithms developed by performing a before-after analysis of the fare scheme change from zone-based to flat fares (Cats et al., 2019, Kholodov et al., 2021).

Data-driven supply and demand modeling

In the project Unravelling Mobility Patterns using Longitudinal Smart Card Data (2020-2021) we advanced the analysis capabilities by discovering the prevailing demand patterns and identifying distinctive user profiles from the data. To this end we performed a series of clustering studies. The results offer a more nuanced understanding of different user segments. Moreover, we selected as a case study the impacts of a particular infrastructural investment, the opening of the Citybanan commuter train railway, on how different user groups travel (Cats et al., 2021, Kolkowski et al., 2023). KTH news article in Swedish.

User segmentation

When the Covid-19 pandemic caused dramatic changes in travel behavior during 2020 and onward, we applied the developed analysis tools to evaluate the effects on public transport ridership (Jenelius and Cebecauer, 2020) and the equity effects in the Stockholm region (Almlöf et al., 2021). In particular, we studied the propensity to stop travelling by public transport during COVID-19 and found that socioeconomic status affected the change in behaviour during the pandemic and that exposure to the virus was determined by citizens’ socioeconomic class. KTH news article in Swedish.

Travel patterns, socio-economics, covid-19

Impact

There is a development project at the Public Transport Administration of Stockholm to incorporate KTH:s journey inference algorithm in-house. The algorithm, the Administration believes, is seen as a crucial tool for future work in journey forecasting and forming revenue strategies and general planning purposes.

The study Almlöf et al. (2021) has been used at the Public Transport Administration to assess a sensitivity case for studies of future travel demand:

  • A sensitivity model is built into the Administration's forecast framework (PT demand assigned on an APT Visum network).
  • Sensitivity cases have been included in a project regarding the delayed arrival of new metro lines in Stockholm (the metro lines are planned to be ready in 2025, the project focus was to find mitigating measures to handle demand in the period between 2025 and the actual opening of new lines)
  • In the long-term plan for public transport development (Region Stockholm, 2021), the sensitivity analysis is used to understand potential differences on regional and local levels between given long-term effects of Covid-19.

The inferred journeys have been used in a development project to find an operable method of assessing the competitiveness of public transport to use as a goal-setting tool in the Administration's work. This project is presented in a master thesis and a full journal paper (Klar and Rubensson, 2024). In addition, the method is currently assessed for incorporation in the Administrations's ongoing strategic work.

The analysis of equity effects during Covid-19 received considerable attention in the media, including Dagens Nyheter (Ritzén, 2021) and Mitti (Johansson, 2021). The work has also spurred political discussion. In particular, the study Almlöf et al. (2021) was cited by politicians from Socialdemokraterna in a proposal to analyze the ability of public transport to promote equity in the Stockholm region (Plambeck, 2021).

Contact person

Isak Rubensson
Samhällsekonom, tekn. Dr
Strategisk utveckling
Analys
E-post: isak.jarlebring-rubensson@regionstockholm.se

References

Almlöf, E., Rubensson, I., Cebecauer, M. & Jenelius, E. (2021). Who continued travelling by public transport during COVID-19? : Socioeconomic factors explaining travel behaviour in Stockholm 2020 based on smart card data. European Transport Research Review, 13(1). 

Cats, O., Ferranti, F., Rubensson, I., Cebecauer, M., Kolkowski, L. & Jenelius, E. (2021). Unravelling Mobility Patterns using Longitudinal Smart Card Data : Final report for Trafik och Region 2019SLL-KTH research project. KTH Royal Institute of Technology.

Cats, O., Rubensson, I., Cebecauer, M., Kholodov, Y., Vermeulen, A., Jenelius, E. & Susilo, Y. (2019). How fair is the fare? Estimating travel patterns and the impacts of fare schemes for different user groups in Stockholm based on smartcard data : Final report for Trafik och Region 2018 SLL-KTH research project. KTH Royal Institute of Technology.

Jenelius, E. & Cebecauer, M. (2020). Impacts of COVID-19 on public transport ridership in Sweden: Analysis of ticket validations, sales and passenger counts. Transportation Research Interdisciplinary Perspectives, 8.

Johansson, I. (2021). ”Ojämlik kollektivtrafik får S att rasa”, Mitti, published 24 August 2021. https://www.mitti.se/nyheter/ojamlik-kollektivtrafik-far-s-att-rasa/repuhr!cYXWTgtsmHShvJ2qrsmg/

Kholodov, Y., Jenelius, E., Cats, O., van Oort, N., Mouter, N., Cebecauer, M. & Vermeulen, A. (2021). Public transport fare elasticities from smartcard data: Evidence from a natural experiment. Transport Policy, 105, 35-43.

Klar, R. & Rubensson, I. (2024) "Spatio-Temporal Investigation of Public Transport Demand Using Smart Card Data". Applied Spatial Analysis and Policy 17, 241–268.

Kolkowski, L., Cats, O., Dixit, M., Verma, T., Jenelius, E., Cebecauer, M. & Rubensson, I. J. (2023). Measuring activity-based social segregation using public transport smart card data. Journal of Transport Geography, 110.

Plambeck, J. (2021). ”Svar på skrivelse från Socialdemokraterna om jämlikhetsanalys” Region Stockholm Trafikförvaltningen Strategisk utveckling. Tjänsteutlåtande TN 2021-0880. 

Region Stockholm (2021). "Kollektivtrafikplan 2050 Remisshandling" Presented at the transport committee (Trafiknämden) Region Stockholm May 2021.

Ritzén, J. (2021) ”Tydliga klasskillnader i SL-trafiken under pandemin”, Dagens Nyheter, published 12 August 2021.  https://www.dn.se/sthlm/tydliga-klasskillnader-i-sl-trafiken-under-pandemin/