kth.sePublications KTH
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Digital Twin for Urban Transportation: Architecture, Technology, Modeling and Applications in Stockholm
KTH, School of Architecture and the Built Environment (ABE), Civil and Architectural Engineering, Transport planning.ORCID iD: 0009-0007-8372-6622
2026 (English)Licentiate thesis, comprehensive summary (Other academic)
Sustainable development
SDG 3: Good Health and Well-Being, SDG 11: Sustainable cities and communities, SDG 13: Climate action
Abstract [en]

This licentiate thesis develops and prototypically demonstrates Digital Twin (DT) architectures for urban transportation with a focus on road-traffic emissions, Public Transportation (PT) and data-driven path flow estimation to support DT simulation workflows. To advance current transport DTs from isolated sensor dashboards and disconnected simulations, the thesis addresses three research objectives: 1) to formulate a transferable DT architecture for network-level road-traffic emission nowcasting, forecasting, and retrospective analysis, 2) to design an open, automated, and extendable DT development pipeline for PT, and 3) to investigate the robustness and transferability of Partial Least Squares Regression (PLSR) for reconstructing path flows from link flow observations. 

Paper I develops and demonstrates an emission-oriented DT framework that integrates camera-based sensing, demand estimation, hybrid traffic simulation, and interactive 3D visualization. Traffic cameras are processed using a computer-vision pipeline that detects and classifies vehicles and extracts speed and acceleration used to nowcast emissions at camera locations. To further estimate network-level emissions, the camera data is used to estimate dynamic OD demand as input for microscopic simulation. A Unity-based 3D platform integrates sensor and simulation output using MQTT-based data streaming. This platform enables joint exploration of near-real-time emissions at sensor locations and simulated emission estimates in the surrounding network. A case study in Kista, Stockholm, illustrates the framework’s ability to support both emission nowcasting and scenario analysis, for example by assessing changes in network-level emissions under reduced parking availability.

Since DTs integrate multiple complex components, their development is often resource-intensive and time-consuming. To lower entry barriers Paper II proposes an open, automated DT development pipeline for PT that relies on open data, open standards, and open source software. PT operations are represented using GTFS data, where its static component serves as input for microscopic traffic simulation to enable joint simulation of PT vehicles and private traffic interactions. The GTFS real-time feeds enable both monitoring of current PT operations through low-latency visualization and retrospective analysis of events using a database storing historic observations. The link- and vehicle-level traffic data is displayed together with automatically derived OpenStreetMap building models in a Cesium-based web platform. This interactive visualization allows users to switch between nowcasting, scenario-based forecasting, and playback of historical operations within a 3D spatial context facilitating informed decision making. A case study in Kista, Stockholm demonstrates the pipeline’s technical feasibility by showcasing real-time PT operations and simulation-based scenarios visualized in the 3D interactive DT platform.

Paper III formulates and evaluates a PLSR-based path flow estimator as a data-driven alternative to conventional OD matrix estimation (ODME) within DT simulation workflows. Estimating a larger number of OD pairs from a smaller number of link counts, as well as collinearities in the observations render the path flow estimation problem ill-posed. PLSR learns a low-dimensional latent representation that maximizes the covariance between observed link flows and path flows, providing implicit regularization for the ill-posed inverse problem. While the method was used for similar problem settings, it has not yet been used for path flow estimation. Thus, this study evaluates its suitability by assessing its stability and transferability using a synthetic test network and controlled data-generating processes that reflect practically relevant OD and route choice structures. The experiments indicate that PLSR achieves the lowest reconstruction errors when variability in path flows is dominated by OD demand fluctuations. Increased path choice randomness, however, is reducing recoverability at first, but the performance stabilizes once strong path competition regimes are entered. The experiments further indicate that PLSR transfers reliably when fitting and deployment share the same correlation regimes, but performance deteriorates near regime boundaries where OD-driven correlations give way to path choice competition-based correlations. These findings suggest that PLSR can serve as a fast, data-driven path flow estimator in DT contexts. Though, to ensure reliable estimates over time it requires continuous monitoring of the underlying correlation regimes to detect shifts and retrain the model when needed.

Abstract [sv]

Denna licentiatuppsats utvecklar och prototypiskt demonstrerar Digital Twin (DT)-arkitekturer för urbana transportsystem med fokus på vägtrafikutsläpp, kollektivtrafik och datadriven skattning av vägflöden för att stödja simuleringsbaserade DT-arbetsflöden. För att föra dagens transportrelaterade DT-lösningar vidare från isolerade sensordashboards och fristående simuleringar behandlar avhandlingen tre forskningsmål: 1) att formulera en överförbar DT-arkitektur för nätverksomfattande skattning i realtid, prognoser och retrospektiv analys av vägtrafikutsläpp, 2) att utforma en öppen, automatiserad och utbyggbar utvecklingspipeline för DT inom kollektivtrafik, samt 3) att undersöka robustheten och överförbarheten hos Partial Least Squares Regression (PLSR) för rekonstruktion av vägflöden från länktrafikobservationer.

Artikel I utvecklar och demonstrerar ett utsläppsorienterat DT-ramverk som integrerar kamerabaserad datainsamling, efterfrågeskattning, hybrid trafiksimulering och interaktiv 3D-visualisering. Trafikkameror behandlas med en datorseendepipeline som detekterar och klassificerar fordon samt extraherar hastighet och acceleration, vilka används för att skatta utsläpp i realtid vid kamerapositionerna. För att vidare uppskatta utsläpp på nätverksnivå används kameradata för att skatta dynamisk OD-efterfrågan som sedan används som indata till mikrosimulering. En Unity-baserad 3D-plattform integrerar sensor- och simuleringsdata via MQTT-baserad dataströmning. Plattformen möjliggör gemensam analys av nästan realtida utsläpp vid sensorplatser och simulerade utsläppsskattningar i det omgivande nätverket. En fallstudie i Kista, Stockholm, visar ramverkets förmåga att stödja både realtidsskattning av utsläpp och scenarioanalys, exempelvis genom att bedöma förändringar i nätverksomfattande utsläpp vid minskad parkeringstillgång.

Eftersom DT-system integrerar flera komplexa komponenter är deras utveckling ofta resurskrävande och tidsintensiv. För att sänka trösklarna föreslår Artikel II en öppen och automatiserad DT-utvecklingspipeline för kollektivtrafik som bygger på öppna data, öppna standarder och programvara med öppen källkod. Kollektivtrafikens drift representeras med GTFS-data, där den statiska delen används som indata till mikroskopisk trafiksimulering för att möjliggöra gemensam simulering av kollektivtrafikfordon och interaktioner med privat trafik. GTFS realtidsflöden möjliggör både övervakning av aktuell kollektivtrafikdrift genom visualisering med låg latens och retrospektiv analys av händelser genom en databas som lagrar historiska observationer. Trafikdata på länk- och fordonsnivå visualiseras tillsammans med automatiskt genererade OSM-baserade byggnadsmodeller i en Cesium-baserad webbplattform. Denna interaktiva visualisering gör det möjligt för användare att växla mellan realtidsövervakning, scenariobaserad prognostisering och uppspelning av historisk drift i en rumslig 3D-kontext som underlättar välgrundat beslutsfattande. En fallstudie i Kista, Stockholm, demonstrerar pipeline-lösningens tekniska genomförbarhet genom att visa realtida kollektivtrafikdrift och simuleringsbaserade scenarier i den interaktiva 3D-baserade DT-plattformen.

Artikel III formulerar och utvärderar en PLSR-baserad skattare av vägflöden som ett datadrivet alternativ till konventionell skattning av OD-matriser inom DT-baserade simuleringsarbetsflöden. Att skatta ett större antal OD-par utifrån ett mindre antal länkräkningar, i kombination med kollinearitet i observationerna, gör problemet att skatta vägflöden ill-posed. PLSR lär sig en lågdimensionell latent representation som maximerar kovariansen mellan observerade länkflöden och vägflöden, vilket ger en implicit regularisering av det ill-posed inversa problemet. Även om metoden har använts i liknande problemsammanhang har den ännu inte tillämpats för skattning av vägflöden. Studien utvärderar därför dess lämplighet genom att analysera dess stabilitet och överförbarhet med hjälp av ett syntetiskt testnätverk och kontrollerade datagenererande processer som speglar praktiskt relevanta OD- och vägvalsstrukturer. Experimenten visar att PLSR uppnår lägst rekonstruktionsfel när variationen i vägflöden domineras av variation i OD-efterfrågan. Ökad slumpmässighet i vägval minskar däremot initialt rekonstruerbarheten, men prestandan stabiliseras när systemen går in i regimer med stark konkurrens mellan vägar. Experimenten indikerar vidare att PLSR är pålitligt överförbar när träning och tillämpning delar samma korrelationsregimer, men att prestandan försämras nära regimgränser där OD-drivna korrelationer övergår i korrelationer som orsakas av konkurrens mellan vägval. Resultaten tyder på att PLSR kan fungera som en snabb och datadriven skattare av vägflöden i DT-sammanhang. För att säkerställa tillförlitliga skattningar över tid krävs dock kontinuerlig övervakning av de underliggande korrelationsregimerna för att upptäcka skiften och träna om modellen vid behov.

Place, publisher, year, edition, pages
Stockholm: KTH Royal Institute of Technology, 2026. , p. 44
Series
TRITA-ABE-DLT ; 2616
Keywords [en]
Digital Twin, Traffic Emissions, Public Transport, Path flow estimation
Keywords [sv]
Digital tvilling, Trafikutsläpp, Kollektivtrafik, Beräkning av trafikflöde
National Category
Transport Systems and Logistics
Research subject
Transport Science, Transport Systems
Identifiers
URN: urn:nbn:se:kth:diva-381058ISBN: 978-91-8106-604-3 (print)OAI: oai:DiVA.org:kth-381058DiVA, id: diva2:2058883
Presentation
2026-06-08, W25, Teknikringen 78A, KTH Campus, public video conference link https://kth-se.zoom.us/j/61606441432, Stockholm, 14:00 (English)
Opponent
Supervisors
Projects
GEMINI: DiGital twin for Emission MonItoring aNd predIction – Kista CaseDIRAC: DynamIc uRban roAd trafiC noise simulation model using passive and publicly available dataT-Twin: Digital Twin Sandbox for Network-Level Traffic Control
Note

QC 20260511

Available from: 2026-05-11 Created: 2026-05-08 Last updated: 2026-05-19Bibliographically approved
List of papers
1. Digital Twin for urban car traffic emission: A case study in Kista, Stockholm
Open this publication in new window or tab >>Digital Twin for urban car traffic emission: A case study in Kista, Stockholm
Show others...
2026 (English)In: Journal of Intelligent and Connected Vehicles, E-ISSN 2399-9802Article in journal (Refereed) Epub ahead of print
Abstract [en]

The commitment to decarbonization is motivating urban planners to adopt emerging techniques that advance sustainability. Road traffic emissions remain a major source of greenhouse gases and pollutants, requiring precise, near-real-time monitoring for effective mitigation policies. This study introduces the design and demonstration of a Digital Twin (DT) platform for road traffic emission nowcasting and forecasting. The focus is on establishing a streamlined technical architecture and showcasing how the system can utilize multi-source data from IoT sensors and simulation to provide a high spatio-temporal resolution view of emissions. As a proof of concept, the platform leverages traffic camera data as IoT input, highlighting its potential for simultaneous emission and Origin Destination Matrix Estimation (ODME). A case study in Kista, Stockholm, illustrates the platform’s capabilities through a 3D interactive visualization in Unity. This demonstration serves as a first step toward a fully validated emission monitoring system, providing a scalable and modular framework that can be adapted for related applications, such as congestion analysis and noise monitoring. 

Place, publisher, year, edition, pages
Tsinghua University Press, 2026
Keywords
Traffic emission, Digital twin, Computer vision, Simulation, Intelligent transportation system
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:kth:diva-380468 (URN)10.26599/jicv.2026.9210079 (DOI)
Funder
Digital FuturesTrenOp, Transport Research Environment with Novel Perspectives
Note

QC 20260430

Available from: 2026-04-29 Created: 2026-04-29 Last updated: 2026-05-08Bibliographically approved
2. Open source Digital Twin development pipeline for Public Transport: A case study in Stockholm
Open this publication in new window or tab >>Open source Digital Twin development pipeline for Public Transport: A case study in Stockholm
2026 (English)In: Journal of Public Transportation, ISSN 1077-291X, Vol. 28, article id 100149Article in journal (Refereed) Published
Abstract [en]

Digital Twins (DTs) offer a wide range of functionalities including retrospective analyses of past conditions, real-time monitoring and control, as well as the evaluation of future scenarios. These capabilities make DTs a valuable tool in the Public Transport (PT) domain. To enable the functionalities demanded by traffic planners and operators, DTs rely on the integration of aggregated and disaggregated data from real-world sources and simulations. Informative visualizations of these data sources within the street network, combined with 3D representations of the built environment, provide the necessary contextualization to facilitate decision-making. However, since DTs integrate this large set of different components, their development is typically time- and resource-intensive which represents a barrier for their adoption. To make DTs more widely accessible and promote their use in the PT field, we propose and implement a DT development pipeline that integrates real-world data collection and processing, simulation, and 3D visualization in a unified framework. The pipeline is based on open data and open source software to enable the implementation of PT DTs with low barriers while allowing for extensibility to other urban-related applications. We demonstrate the DT’s functionalities with a prototypical implementation for the region of Kista in Stockholm, and discuss the potential as well as the limitations of PT DTs for practical use cases from a conceptual perspective.

Place, publisher, year, edition, pages
Elsevier BV, 2026
Keywords
Digital twin, Public transport, 3D visualization, Open source, Open data
National Category
Transport Systems and Logistics
Research subject
Transport Science, Transport Systems
Identifiers
urn:nbn:se:kth:diva-380467 (URN)10.1016/j.jpubtr.2026.100149 (DOI)001693266100001 ()2-s2.0-105035172165 (Scopus ID)
Funder
Digital FuturesTrenOp, Transport Research Environment with Novel Perspectives
Note

QC 20260430

Available from: 2026-04-29 Created: 2026-04-29 Last updated: 2026-05-08Bibliographically approved
3. Partial Least Squares Regression for Ill-Posed Path Flow Estimation from Partial Link Traffic Counts
Open this publication in new window or tab >>Partial Least Squares Regression for Ill-Posed Path Flow Estimation from Partial Link Traffic Counts
2026 (English)Conference paper, Oral presentation with published abstract (Refereed)
Abstract [en]

The problem of estimating path flows from partially observed link counts is ill-posed, as many different path flow vectors can produce the same set of link measurements. Nevertheless, reliable online path flow estimation is critical for real-time monitoring and Digital Twin applications. Existing approaches typically rely on strong prior assumptions about path flows or apply dimensionality reduction techniques that treat link flows solely as predictors, without explicitly accounting for their relationship to the unobserved path flows. We evaluate the applicability of Partial Least Squares Regression (PLSR), a supervised dimensionality reduction method that fits a low-dimensional mapping between observed link flows and unobserved path flows by maximizing their covariance, thereby providing implicit regularization. We assess its stability and transferability using a synthetic network under multiple data-generating processes designed to represent practically relevant demand levels and path-choice patterns, including 1) independent OD demand with fixed path shares, 2) fixed OD demand with multinomial path choices, and 3) correlated OD demand via a Poisson log-normal distribution, combined with path cost uncertainty. Results indicate that the estimation accuracy depends on the underlying demand and path flow correlation structure and may deteriorate under substantial shifts in the path choice variability.

Keywords
ill-posed path flow estimation, Partially observed link counts, Partial least squares regression (PLSR), Origin–destination demand
National Category
Transport Systems and Logistics
Research subject
Transport Science
Identifiers
urn:nbn:se:kth:diva-381001 (URN)
Conference
IEEE International Conference on Intelligent Transportation Systems (ITSC), Naples, Italy, Sep 15 – 18, 2026
Available from: 2026-05-07 Created: 2026-05-07 Last updated: 2026-05-12

Open Access in DiVA

fulltext(1710 kB)110 downloads
File information
File name FULLTEXT01.pdfFile size 1710 kBChecksum SHA-512
1a87cc9a4e611e5e12d1ec2eb8927869f150cbc2ce3031a8fca8fe92e2a0869babde45b12b2041558ceecf8dcb65020e2d96f3e08ebc99e97f798e2941b6f7a5
Type fulltextMimetype application/pdf

Authority records

Jostmann, Jonas

Search in DiVA

By author/editor
Jostmann, Jonas
By organisation
Transport planning
Transport Systems and Logistics

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 967 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf