kth.sePublications
Change search
Link to record
Permanent link

Direct link
Publications (10 of 45) Show all publications
Skoufas, A., Jenelius, E., Cebecauer, M., Cats, O. & Burghout, W. (2024). Ex-post assessment of public transport on-board crowding induced by new urban development. In: : . Paper presented at Transportation Research Board (TRB) 103rd Annual Meeting, Washington DC, USA, 7-11 January 2024.
Open this publication in new window or tab >>Ex-post assessment of public transport on-board crowding induced by new urban development
Show others...
2024 (English)Conference paper, Poster (with or without abstract) (Refereed)
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344062 (URN)
Conference
Transportation Research Board (TRB) 103rd Annual Meeting, Washington DC, USA, 7-11 January 2024
Funder
Region Stockholm, RS 2022-0210
Note

QC 20240311

Available from: 2024-02-29 Created: 2024-02-29 Last updated: 2024-03-12Bibliographically approved
Burghout, W., Cebecauer, M., Danielsson, A., Gundlegård, D., Jenelius, E. & Rydergren, C. (2024). Multimodal Traffic Management: Project Report. Trafikverket
Open this publication in new window or tab >>Multimodal Traffic Management: Project Report
Show others...
2024 (English)Report (Other academic)
Abstract [sv]

Nya system för att kombinera transportsätt, till exempel Mobility as a Service (MaaS), ger nya möjligheter för trafikanter att växla mellan olika färdmedel. Samtidigt ger stora mängder data från såväl kollektivtrafiknätet som vägtrafiknätet samt multimodala data från mobilnäten i kombination med nya metoder för att uppskatta resmönster uppdelat på färdmedel möjligheter till en helt ny förståelse av multimodala resmönster i en stad. Att förstå hur multimodala resmönster utvecklas över tid ger nya möjligheter att utveckla effektiva verktyg för multimodal trafikledning.

Det övergripande målet med projektet är att möjliggöra förbättrad tillgänglighet i transportsystemen genom effektivare trafikledning. Mer specifikt syftar projektet till att utveckla nya metoder för att uppskatta multimodal efterfrågan samt färdmedelsval och ruttval för multimodal trafikledning. Vidare har potentiella effekter av multimodal trafikledning analyserats.

Projektet omfattar en litteraturstudie för analys av möjligheter och utmaningar med multimodal trafikledning. En explorativ analys baserad på oövervakat lärande har utförts för att identifiera typiska nätverksövergripande mobilitetsmönster. Val av rutt och färdmedel har predikterats med hjälp av statistiska modeller. Ett multimodalt dataset för fem veckor i Stockholm med storskalig mobilitetsdata för vägnätet och biljettdata för kollektivtrafiknätet har sammanställs för den explorativa analysen samt utvärderingen av rutt- och transportsättsmodellerna i samband med trafikledning.

Baserat på litteraturstudien kan vi dra slutsatsen att koordinerad ledning av väg och kollektivtrafik har potential att minska trängseln och säkerställa effektiv förflyttning av resenärer i ett storstadsområde. Det finns flera motiv för multimodal trafikledning, där de viktigaste är potentiellt ökad efterfrågan för kollektivtrafik, förbättrad robusthet för transportsystemet och bättre prioritering av trafikledningsåtgärder. De största utmaningarna är samarbete mellan intressenter, informationsdelning och datafusion.

Resultaten av den explorativa analysen baserad på oövervakad inlärning indikerar att klustring för att ta fram typdagar kan vara användbart vid scenarioutvärdering, men också fungera som input till korttidsprediktion, vilket ger en enkel och robust predikteringsmetod för länkflöden med ett MAPE-prediktionsfel på 10-15 %.

Ruttvalsanalysen visar att en modell baserad på en ruttuppsättning med genererade rutter är mer responsiv för restidsförändringar än en modell baserad på endast observerade rutter, vilket är användbart för att förutspå effekten av olika trafikledningsåtgärder. En ruttvalsmodell med enbart restid är en vanlig förenkling att använda för att prediktera ruttval, men resultatet i denna studie visar att inkludering av fler attribut avsevärt förbättrar modellernas prestanda.

Analysen av nätverksövergripande multimodala data för 5 veckor i Stockholm indikerar att det är möjligt att uppskatta hur transportsättsandelen mellan kollektivtrafik och andra transportslag varierar i tid och rum. En bättre förståelse för spatiotemporal variation av färdmedelsvalet är en viktig input till förbättrat beslutsstöd i multimodal trafikledning.

Abstract [en]

New systems for combining modes of transport, for example Mobility as a Service (MaaS), provide new opportunities for road users to switch between different means of transport. At the same time, large amounts of data from both the public transport network and the road traffic network as well as multimodal data from mobile networks in combination with new methods for estimating travel patterns divided by means of transport provide opportunities for a completely new understanding of multimodal travel patterns in a city. Understanding how multimodal travel patterns develop over time provides new opportunities to develop effective tools for multimodal traffic management.

The overall goal of the project is to enable improved accessibility in the transport systems through more efficient traffic management. More specifically, the project aims to develop new methods for estimating multimodal demand as well as mode of transport and route selection for multimodal traffic management. Furthermore, potential effects of multimodal traffic management should be analysed.

The project includes a literature survey for analysis of potential and challenges of multimodal traffic management. An explorative analysis based on unsupervised learning is performed for identification of typical network-wide mobility patterns. Route and mode choice is predicted using statistical models. A five-week multimodal dataset for Stockholm including large-scale mobility data for the road network and smartcard data for the public transport network is compiled for the explorative analysis as well as evaluation of the route and mode choice models in the context of traffic management.

Based on the literature survey, we can conclude that simultaneous management of road and public transport has the potential to reduce congestion and ensure efficient movement of travelers in an urban area. There are several motives for integrated management of multiple modes, where the most important are potential demand shifts to public transport, improved robustness for the transport system, and better prioritization of traffic management actions. The main challenges are collaboration between stakeholders, information sharing, and data fusion.

The results of the explorative analysis based on unsupervised learning indicate that day clustering can be useful in scenario evaluation, but also serve as input to short-term prediction providing a simple and robust prediction method with a MAPE prediction error of 10-15%.

The route choice analysis showed that a model based on a route set with generated routes is more responsive to travel time changes than a model based on only observed routes, which is useful for predicting the effect of traffic management actions. A route choice model with only travel time is a common simplification to use for prediction route choices. However, the result in this study shows that including more attributes significantly improves the performance of the models.

The analysis of network-wide multimodal data for 5 weeks in Stockholm indicates that it is possible to estimate how mode share between public transport and other modes of transport varies in space and time. A better understanding of spatiotemporal variation of mode share is an important input to improved decision support in multimodal traffic management.

Place, publisher, year, edition, pages
Trafikverket, 2024. p. 34
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-345790 (URN)
Funder
Swedish Transport Administration, TRV 2020/118663
Note

QC 20240425

Available from: 2024-04-19 Created: 2024-04-19 Last updated: 2024-04-25Bibliographically approved
Khan, M. A., Burghout, W., Cats, O., Jenelius, E. & Cebecauer, M. (2023). A comprehensive review of viability and operability of dynamic charging solutions for autonomous electric vehicles. In: : . Paper presented at 12th Annual Swedish Transport Research Conference, Stockholm, Sweden, 16-17 October 2023.
Open this publication in new window or tab >>A comprehensive review of viability and operability of dynamic charging solutions for autonomous electric vehicles
Show others...
2023 (English)Conference paper, Oral presentation only (Other academic)
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344137 (URN)
Conference
12th Annual Swedish Transport Research Conference, Stockholm, Sweden, 16-17 October 2023
Funder
Swedish Transport Administration, TRV 2022/8287
Note

QC 20240304

Available from: 2024-03-04 Created: 2024-03-04 Last updated: 2024-03-04Bibliographically approved
Ngo, H. N., Kaddoum, E., Cebecauer, M., Jenelius, E. & Goursolle, A. (2023). Considering Multi-Scale Data for Continuous Traffic Prediction Using Adaptive Multi-Agent System. In: Proceedings 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC): . Paper presented at 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, 24-28 September 2023 (pp. 1835-1842). Institute of Electrical and Electronics Engineers (IEEE)
Open this publication in new window or tab >>Considering Multi-Scale Data for Continuous Traffic Prediction Using Adaptive Multi-Agent System
Show others...
2023 (English)In: Proceedings 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Institute of Electrical and Electronics Engineers (IEEE) , 2023, p. 1835-1842Conference paper, Published paper (Refereed)
Abstract [en]

Accurate traffic prediction is essential for effective traffic management and planning. However, traffic prediction models are challenged by various factors such as complex spatiotemporal dependencies in traffic data. Recently, researchers have explored the new approach known as stream analysis since it can continuously update models by capturing new behaviors from the traffic data stream. However, applying this approach specifically raises the question about the balance between model complexity and model flexibility for dynamic updates. ADRIP - Adaptive multi-agent system for DRIving behaviors Prediction proposed in [1], [2] has combined the dynamic clustering and the multi-agent system approach to solve this challenge. This system has been applied to predict traffic dynamics at the road segment level. In this paper, we aim to extend ADRIP to complete its functionality for traffic prediction at the network level. Experiments for multi-scale traffic data are conducted to compare extended ADRIP with well-known clustering models.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2023
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344059 (URN)10.1109/ITSC57777.2023.10422536 (DOI)001178996701126 ()2-s2.0-85186495074 (Scopus ID)
Conference
2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain, 24-28 September 2023
Funder
Swedish Transport Administration, TRV2020/118663
Note

Part of proceedings ISBN 979-8-3503-9946-2

QC 20240304

Available from: 2024-02-29 Created: 2024-02-29 Last updated: 2024-06-20Bibliographically approved
Skoufas, A., Cebecauer, M., Burghout, W. & Jenelius, E. (2023). Generating and evaluating route choice sets for large multimodal public transport networks: A case study for Stockholm Region. In: : . Paper presented at 12th Annual Swedish Transport Research Conference, Stockholm, Sweden, 16-17 October 2023.
Open this publication in new window or tab >>Generating and evaluating route choice sets for large multimodal public transport networks: A case study for Stockholm Region
2023 (English)Conference paper, Oral presentation only (Other academic)
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344138 (URN)
Conference
12th Annual Swedish Transport Research Conference, Stockholm, Sweden, 16-17 October 2023
Funder
Swedish Transport Administration, TRV 2022/33324TrenOp, Transport Research Environment with Novel Perspectives
Note

QC 20240304

Available from: 2024-03-04 Created: 2024-03-04 Last updated: 2024-03-04Bibliographically approved
Skoufas, A., Cebecauer, M., Burghout, W. & Jenelius, E. (2023). Generating and Evaluating Route Choice Sets for Large Multimodal Public Transport Networks: A Case Study for Stockholm Region. In: 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC): . Paper presented at 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbo, Spain (pp. 2926-2931). Institute of Electrical and Electronics Engineers (IEEE)
Open this publication in new window or tab >>Generating and Evaluating Route Choice Sets for Large Multimodal Public Transport Networks: A Case Study for Stockholm Region
2023 (English)In: 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Institute of Electrical and Electronics Engineers (IEEE) , 2023, p. 2926-2931Conference paper, Published paper (Refereed)
Abstract [en]

Identification of Choice Sets (CSs) is a crucial step towards the estimation of public transport route choice models. However, identification of reliable CSs is a challenging task as the routes considered by travelers are not directly observed. To handle this issue, this study adopts an existing Choice Set Generation Methodology (CSGM) for the identification of CSs by using General Transit Feed Specification (GTFS) data. The final feasible CSs are then compared to the actual passengers' choices observed in Smart Card Data (SCD) by using three validation metrics; passenger and network coverage as well as network efficiency. The aim of the study is to shed light on the performance of a CSGM in Stockholm's multimodal transit network by using data retrieved by different Intelligent Transportation System (ITS) applications focused on operations and on ridership automated data collection. However, the use of CSGM for large networks raises scalability and computational issues. In this direction, the study also contributes in the implementation of the CSGM in a larger network compared to existing case studies.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2023
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344058 (URN)10.1109/ITSC57777.2023.10422419 (DOI)001178996702139 ()2-s2.0-85186489833 (Scopus ID)
Conference
2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbo, Spain
Funder
TrenOp, Transport Research Environment with Novel Perspectives
Note

QC 20240306

Part of ISBN: 979-8-3503-9946-2

Available from: 2024-02-29 Created: 2024-02-29 Last updated: 2024-06-18Bibliographically approved
Cebecauer, M., Gundlegård, D., Jenelius, E. & Burghout, W. (2023). High-resolution public transport mode share estimation from mobile network and smart card data. In: : . Paper presented at 12th Annual Swedish Transport Research Conference, Stockholm, Sweden, 16-17 October 2023.
Open this publication in new window or tab >>High-resolution public transport mode share estimation from mobile network and smart card data
2023 (English)Conference paper, Oral presentation only (Other academic)
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344122 (URN)
Conference
12th Annual Swedish Transport Research Conference, Stockholm, Sweden, 16-17 October 2023
Funder
Swedish Transport Administration, TRV 2023/62202
Note

QC 20240304

Available from: 2024-03-01 Created: 2024-03-01 Last updated: 2024-03-04Bibliographically approved
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, Article ID 103642.
Open this publication in new window or tab >>Measuring activity-based social segregation using public transport smart card data
Show others...
2023 (English)In: Journal of Transport Geography, ISSN 0966-6923, E-ISSN 1873-1236, Vol. 110, article id 103642Article in journal (Refereed) Published
Abstract [en]

While social segregation is often assessed using static data concerning residential areas, the extent to which people with diverse background travel to the same destinations may offer an additional perspective on the extent of urban segregation. This study further contributes to the measurement of activity-based social segregation between multiple groups using public transport smart card data. In particular, social segregation is quantified using the ordinal information theory index to measure the income group mix at public transport journey destination zones. The method is applied to the public transport smart card data of Stockholm County, Sweden. Applying the index on 2017-2020 data sets for a selected week, shows significant differences between income groups' segregation along the radial public transport corridors following the opening of a major rail project in the summer of 2017. The overall slight decrease in segregation over the years can be linked to declining segregation in the city center as a travel destination and its public transport hubs. Increasing zonal segregation is observed in suburban and rural zones with commuter train stations. This method helps to quantify social segregation, enriching the analysis of urban segregation and can aid in evaluating policies based on the dynamics of social life.

Place, publisher, year, edition, pages
Elsevier BV, 2023
Keywords
Social segregation, Public transport, Ex-post transport appraisal, Smart card data
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-333756 (URN)10.1016/j.jtrangeo.2023.103642 (DOI)001032026400001 ()2-s2.0-85163927737 (Scopus ID)
Note

QC 20230810

Available from: 2023-08-10 Created: 2023-08-10 Last updated: 2023-08-10Bibliographically approved
Cebecauer, M., Jenelius, E., Gundlegård, D. & Burghout, W. (2023). Revealing representative day-types in transport networks using traffic data clustering. Journal of Intelligent Transportation Systems / Taylor & Francis, 1-24
Open this publication in new window or tab >>Revealing representative day-types in transport networks using traffic data clustering
2023 (English)In: Journal of Intelligent Transportation Systems / Taylor & Francis, ISSN 1547-2450, E-ISSN 1547-2442, p. 1-24Article in journal (Refereed) Published
Abstract [en]

Recognition of spatio-temporal traffic patterns at the network-wide level plays an important role in data-driven intelligent transport systems (ITS) and is a basis for applications such as short-term prediction and scenario-based traffic management. Common practice in the transport literature is to rely on well-known general unsupervised machine-learning methods (e.g., k-means, hierarchical, spectral, DBSCAN) to select the most representative structure and number of day-types based solely on internal evaluation indices. These are easy to calculate but are limited since they only use information in the clustered dataset itself. In addition, the quality of clustering should ideally be demonstrated by external validation criteria, by expert assessment or the performance in its intended application. The main contribution of this paper is to test and compare the common practice of internal validation with external validation criteria represented by the application to short-term prediction, which also serves as a proxy for more general traffic management applications. When compared to external evaluation using short-term prediction, internal evaluation methods have a tendency to underestimate the number of representative day-types needed for the application. Additionally, the paper investigates the impact of using dimensionality reduction. By using just 0.1% of the original dataset dimensions, very similar clustering and prediction performance can be achieved, with up to 20 times lower computational costs, depending on the clustering method. K-means and agglomerative clustering may be the most scalable methods, using up to 60 times fewer computational resources for very similar prediction performance to the p-median clustering.

Place, publisher, year, edition, pages
Informa UK Limited, 2023
Keywords
Cluster validity, clustering, day clustering, dimensionality reduction, external indices, internal indices, network-wide prediction
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-333918 (URN)10.1080/15472450.2023.2205020 (DOI)000989310600001 ()
Funder
Swedish Transport Administration, TRV 2020/118663
Note

QC 20230815

Available from: 2023-08-14 Created: 2023-08-14 Last updated: 2023-09-21Bibliographically approved
Cebecauer, M., Gundlegård, D., Jenelius, E. & Burghout, W. (2023). Spatio-Temporal Public Transport Mode Share Estimation and Analysis Using Mobile Network and Smart Card Data. In: 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC): . Paper presented at 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain (pp. 2543-2548). Institute of Electrical and Electronics Engineers (IEEE)
Open this publication in new window or tab >>Spatio-Temporal Public Transport Mode Share Estimation and Analysis Using Mobile Network and Smart Card Data
2023 (English)In: 2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Institute of Electrical and Electronics Engineers (IEEE) , 2023, p. 2543-2548Conference paper, Published paper (Refereed)
Abstract [en]

Public transport plays a vital role in society and the urban environment. However, knowledge of its spatial and temporal shares is often limited to traditional travel surveys. Recently, there has been substantial progress in mobility data collection, including data from traffic, public transport, and mobile phones. Especially mobile network data is a large-scale and affordable source of high-level mobility records. Similarly, public transport smart cards or ticket validation data are being collected and made available in major cities. The contribution of this study is to unveil the potential of estimating public transport shares, by merging mobile and smart card data. Stockholm, Sweden, is used as a case study. We analyze and discuss spatio-temporal patterns of estimated public transport shares for Stockholm, using descriptive and cluster analysis. The typical representative day-types are revealed and analyzed. Finally, a regression analysis considering the weather and socioeconomic context is conducted. It provides a highly explanatory and predictive understanding of which factors impact the share of public transport in Stockholm. To conclude, combined mobile and smart card data offers a cost-efficient, large-scale, low spatio-temporal aggregation (capturing daily and hourly variations) alternative to traditional travel surveys for analyzing PT shares.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2023
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-344060 (URN)10.1109/ITSC57777.2023.10422199 (DOI)001178996702083 ()2-s2.0-85186524253 (Scopus ID)
Conference
2023 IEEE 26th International Conference on Intelligent Transportation Systems (ITSC), Bilbao, Spain
Funder
Swedish Transport Administration, TRV 2020/118663
Note

QC 20240301

Part of ISBN 979-8-3503-9946-2

Available from: 2024-02-29 Created: 2024-02-29 Last updated: 2024-06-19Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-8499-0843

Search in DiVA

Show all publications