Digitala Vetenskapliga Arkivet

Change search
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
Bayesian Models for Spatiotemporal Data from Transportation Networks
Linköping University, Department of Computer and Information Science, The Division of Statistics and Machine Learning. Linköping University, Faculty of Arts and Sciences.ORCID iD: 0000-0001-9025-6701
2023 (English)Doctoral thesis, comprehensive summary (Other academic)Alternative title
Bayesianska modeller för spatiotemporal data från transportnätverk (Swedish)
Abstract [en]

Urbanization has caused a historical transformation at a global scale, and humanity is moving towards a fully connected society where cities will concentrate population, infrastructure and economic activity. A key element in the cities’ infrastructure is the transportation system, as it facilitates the mobility of people and goods. Transportation systems are constantly generating data from, e.g., GPS, sensors and cameras, and the statistical modeling is challenging due to the complex structure and dynamics of the system, and the inherent uncertainty. In this thesis, we develop Bayesian models with applications to transportation. We specifically focus on models that can be trained on spatiotemporal data coming from transport networks to make predictions on, e.g., bus delays or the actual network topology. Special attention has been given to model scalability issues and uncertainty quantification. We have used real-world data from transportation systems in every study to keep a balance between statistical rigor, novelty, and applicability. 

The thesis consists of four papers. The first study presents a state-of-the-art probabilistic latent network model to forecast multilayer dynamic graphs. The model uses stochastic blockmodeling to reduce the computational burden, and is illustrated on a sample of 10-year data from four major airlines within the US air transportation system. In the second paper, we develop a robust model for real-time bus travel time prediction that departs from Gaussian assumptions by using Student-t errors, and show how Bayesian inference naturally allows for predictive uncertainty quantification in a highly stochastic environment. Experiments are performed using data from high-frequency buses in Stockholm, Sweden. The third paper shows the potential of multi-output Gaussian processes to tackle network-wide travel time prediction in an urban area. We develop a responsive online model based on a coregionalized covariance and test its accuracy on real data from GPS-equipped taxis. Finally, we propose a novel regularization strategy for the vector autoregressive model that is based on a graphical spike-and-slab prior, and present a case study with real airline delay data to assess its predictive performance and analyze network patterns related to the propagation of delays across airports. 

Abstract [sv]

Urbaniseringen har orsakat en historisk förändring på en global skala, och mänskligheten går mot ett uppkopplat globalt nätverkssamhälle där städer kommer att koncentrera befolkning, infrastruktur och ekonomisk aktivitet. Ett nyckelelement i städernas infrastruktur är transportsystemet, eftersom det underlättar rörligheten av människor och varor. Transportsystem genererar ständigt data från tex. GPS, sensorer och kameror, och den statistiska modelleringen är utmanande på grund av systemets komplexa struktur och dynamik, samt dess naturliga osäkerheter.

I denna avhandling utvecklar vi Bayesianska modeller med tillämpningar för transporter. Vi fokuserar specifikt på modeller som kan tränas på spatiotemporala data från transportnätverk för att göra prediktioner av t ex. bussförseningar eller verklig nätverkstopologi. Särskild uppmärksamhet har ägnats åt modellskalbarhetsfrågor och kvantifiering av osäkerhet. Vi har använt data från riktiga transportsystem i varje studie för att skapa en balans mellan statistisk korrekthet, praktiskt tillämpbarhet och vetenskaplig höjd. Avhandlingen består av fyra artiklar. Den första artikeln presenterar en probabilistisk latent nätverksmodell för att prognostisera dynamiska grafer med multipla lager. Modellen använder stokastisk blockmodellering för att minska beräkningsbördan, och illustreras på ett datamaterial bestånde av tio års data från fyra stora flygbolag inom det amerikanska lufttransportsystemet. I den andra artikeln utvecklar vi en robust modell för realtidsprognoser av bussförseningar genom att använda Student-t fördelning och vi visar hur Bayesiansk inferens ger en naturlig kvantifiering av osäkerhet i en mycket stokastisk miljö. Experiment utförs med hjälp av högfrekventa data från bussar i Stockholm. Den tredje artikeln visar potentialen hos fler-dimensionella Gaussiska processer för att generera nätverksövergripande prediktioner av trafikflöden i en tätortsmiljö. Vi utvecklar en responsiv onlinemodell baserad på en co-regionaliserad kovariansstruktur och utvärderar prognosförmåga på verkliga data från GPS-utrustade taxibilar. Slutligen föreslår vi en ny regularisering av den vektorautoregressiva modellen via en nätverksbaserad variabelsselektionsprior, och presenterar en fallstudie på verkliga data över förseningar i kommersiell flygtrafik där vi utvärderar prediktiv förmåga och analyserar nätverksmönster för hur förseningar sprids mellan flygplatser.  

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2023. , p. 38
Series
Linköping Studies in Arts and Sciences, ISSN 0282-9800 ; 848Linköping Studies in Statistics, ISSN 1651-1700 ; 17
Keywords [en]
Bayesian statistics, Transportation networks, Spatiotemporal data, Machine learning
Keywords [sv]
Bayesiansk statistik, Transportnätverk, Spatiotemporal data, Maskininlärning
National Category
Probability Theory and Statistics
Identifiers
URN: urn:nbn:se:liu:diva-191153DOI: 10.3384/9789180750363ISBN: 9789180750356 (print)ISBN: 9789180750363 (electronic)OAI: oai:DiVA.org:liu-191153DiVA, id: diva2:1729371
Public defence
2023-02-17, Ada Lovelace, Building B, Campus Valla, Linköping, 13:15 (English)
Opponent
Supervisors
Note

Funding agencies: This work was partially supported by the Wallenberg AI, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation, Sweden. The computations were enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC), partially funded by the Swedish Research Council through grant agreement no. 2018-05973.

Available from: 2023-01-20 Created: 2023-01-20 Last updated: 2023-02-15Bibliographically approved
List of papers
1. A multilayered block network model to forecast large dynamic transportation graphs: An application to US air transport
Open this publication in new window or tab >>A multilayered block network model to forecast large dynamic transportation graphs: An application to US air transport
2022 (English)In: Transportation Research Part C: Emerging Technologies, ISSN 0968-090X, E-ISSN 1879-2359, Vol. 137, article id 103556Article in journal (Refereed) Published
Abstract [en]

Dynamic transportation networks have been analyzed for years by means of static graph-based indicators in order to study the temporal evolution of relevant network components, and to reveal complex dependencies that would not be easily detected by a direct inspection of the data. This paper presents a state-of-the-art probabilistic latent network model to forecast multilayer dynamic graphs that are increasingly common in transportation and proposes a community-based extension to reduce the computational burden. Flexible time series analysis is obtained by modeling the probability of edges between vertices through latent Gaussian processes. The models and Bayesian inference are illustrated on a sample of 10-year data from four major airlines within the US air transportation system. Results show how the estimated latent parameters from the models are related to the airlines’ connectivity dynamics, and their ability to project the multilayer graph into the future for out-of-sample full network forecasts, while stochastic blockmodeling allows for the identification of relevant communities. Reliable network predictions would allow policy-makers to better understand the dynamics of the transport system, and help in their planning on e.g. route development, or the deployment of new regulations.

Place, publisher, year, edition, pages
Oxford, United Kingdom: Elsevier, 2022
Keywords
Transportation networks, Multilayer graphs, Air transport, Machine learning
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:liu:diva-182829 (URN)10.1016/j.trc.2022.103556 (DOI)000777337100005 ()2-s2.0-85124142579 (Scopus ID)
Note

Funding: Wallenberg AI, Autonomous Systems and Software Program (WASP) - Knut and Alice Wallenberg Foundation, Sweden

Available from: 2022-02-09 Created: 2022-02-09 Last updated: 2023-01-20Bibliographically approved
2. Robust Real-Time Delay Predictions in a Network of High-Frequency Urban Buses
Open this publication in new window or tab >>Robust Real-Time Delay Predictions in a Network of High-Frequency Urban Buses
2022 (English)In: IEEE transactions on intelligent transportation systems (Print), ISSN 1524-9050, E-ISSN 1558-0016, Vol. 23, no 9, p. 16304-16317Article in journal (Refereed) Published
Abstract [en]

Providing transport users and operators with accurate forecasts on travel times is challenging due to a highly stochastic traffic environment. Public transport users are particularly sensitive to unexpected waiting times, which negatively affect their perception on the system's reliability. In this paper we develop a robust model for real-time bus travel time prediction that departs from Gaussian assumptions by using Student-t errors. The proposed approach uses spatiotemporal characteristics from the route and previous bus trips to model short-term effects, and date/time variables and Gaussian processes for long-run forecasts. The model allows for flexible modeling of mean, variance and kurtosis spaces. We propose algorithms for Bayesian inference and for computing probabilistic forecast distributions. Experiments are performed using data from high-frequency buses in Stockholm, Sweden. Results show that Student-t models outperform Gaussian ones in terms of log-posterior predictive power to forecast bus delays at specific stops, which reveals the importance of accounting for predictive uncertainty in model selection. Estimated Student-t regressions capture typical temporal variability between within-day hours and different weekdays. Strong spatiotemporal effects are detected for incoming buses from immediately previous stops, which is in line with many recently developed models. We finally show how Bayesian inference naturally allows for predictive uncertainty quantification, e.g. by returning the predictive probability that the delay of an incoming bus exceeds a given threshold.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2022
Keywords
Predictive models; Uncertainty; Real-time systems; Spatiotemporal phenomena; Delays; Data models; Probabilistic logic; Intelligent transportation systems; bus arrival time predictions; spatiotemporal networks; probabilistic modeling; robustness
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:liu:diva-183002 (URN)10.1109/tits.2022.3149656 (DOI)000758176500001 ()2-s2.0-85124827488 (Scopus ID)
Funder
Swedish Research Council, 2020-02846
Note

Funding: Wallenberg AI, Autonomous Systems and Software Program (WASP) by the Knut and Alice Wallenberg Foundation; Swedish Research CouncilSwedish Research CouncilEuropean Commission [2020-02846]

Available from: 2022-02-17 Created: 2022-02-17 Last updated: 2023-03-06Bibliographically approved
3. Urban Network Travel Time Prediction via Online Multi-Output Gaussian Process Regression
Open this publication in new window or tab >>Urban Network Travel Time Prediction via Online Multi-Output Gaussian Process Regression
2017 (English)In: 2017 IEEE 20TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS (ITSC), Institute of Electrical and Electronics Engineers (IEEE) , 2017Conference paper, Published paper (Refereed)
Abstract [en]

The paper explores the potential of Multi-Output Gaussian Processes to tackle network-wide travel time prediction in an urban area. Forecasting in this context is challenging due to the complexity of the traffic network, noisy data and unexpected events. We build on recent methods to develop an online model that can be trained in seconds by relying on prior network dependences through a coregionalized covariance. The accuracy of the proposed model outperforms historical means and other simpler methods on a network of 47 streets in Stockholm, by using probe data from GPS-equipped taxis. Results show how traffic speeds are dependent on the historical correlations, and how prediction accuracy can be improved by relying on prior information while using a very limited amount of current-day observations, which allows for the development of models with low estimation times and high responsiveness.

Place, publisher, year, edition, pages
Institute of Electrical and Electronics Engineers (IEEE), 2017
Series
IEEE International Conference on Intelligent Transportation Systems-ITSC, ISSN 2153-0009
National Category
Transport Systems and Logistics
Identifiers
urn:nbn:se:liu:diva-148663 (URN)10.1109/ITSC.2017.8317796 (DOI)000432373000201 ()2-s2.0-85046289474 (Scopus ID)9781538615263 (ISBN)
Conference
IEEE 20th International Conference on Intelligent Transportation Systems (ITSC), Yokohama, Japan, 16-19 Oct. 2017
Available from: 2018-06-18 Created: 2018-06-18 Last updated: 2023-01-20Bibliographically approved

Open Access in DiVA

fulltext(5204 kB)750 downloads
File information
File name FULLTEXT01.pdfFile size 5204 kBChecksum SHA-512
fb2ac70a9192bf2334597b6c6adb0788ac45ff0a5268d456a35126b54055381f6c7444ac26586f82fa236bbd62fb0038afdd68fc8aaff388b48cb0c8bcc9142f
Type fulltextMimetype application/pdf
Order online >>

Other links

Publisher's full text

Search in DiVA

By author/editor
Rodriguez Déniz, Héctor
By organisation
The Division of Statistics and Machine LearningFaculty of Arts and Sciences
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar
Total: 754 downloads
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

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 4442 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