Data-driven delay analysis with applications to railway networks

Li, B; Guo, T; Wang, Y; Chen, F

Data-driven delay analysis with applications to railway networks

Li, B Guo, T

Wang, Y

Chen, F

Permalink

Publisher:: Wiley
Publication Type:: Chapter
Citation:: Advances in Data Science and Analytics: Concepts and Paradigms, 2023, pp. 115-143
Issue Date:: 2023-10-31

Closed Access

	Filename	Description	Size
	21579662_11974736530005671.pdf		4.16 MB	Adobe PDF	View/Open

Copyright Clearance Process

Recently Added
In Progress
Closed Access

This item is closed access and not available.

Full metadata record

Field	Value	Language
dc.contributor.author	Li, B
dc.contributor.author	Guo, T https://orcid.org/0000-0001-5130-3237
dc.contributor.author	Wang, Y https://orcid.org/0000-0002-6815-0879
dc.contributor.author	Chen, F https://orcid.org/0000-0003-4971-8729
dc.date.accessioned	2023-11-09T21:48:13Z
dc.date.available	2023-11-09T21:48:13Z
dc.date.issued	2023-10-31
dc.identifier.citation	Advances in Data Science and Analytics: Concepts and Paradigms, 2023, pp. 115-143
dc.identifier.isbn	9781119791881
dc.identifier.uri	http://hdl.handle.net/10453/173286
dc.description.abstract	Reliability is one of the key evaluation criteria in railway service. Many factors contribute to the measure, such as delays spanning over spatiotemporal dimensions. A common method used to improve reliability is to design a better timetable to reduce the mutual influence between trains. Recently, machine learning shows great potential in improving the effectiveness and efficiency of decision-making to increase operational performance. For railway system management, this seems like an opportunity worth further exploration. In this chapter, we focus on analyzing railway disruptions and their impact on the traffic in the whole network. Specifically, we build three different data-driven models: (1) a conditional Bayes Net Model with the introduction of Markov property for delay propagation prediction, (2) a primary delay tracking back model, and (3) a dwell improvement evaluation model which can estimate the potential network benefits of the dwell improvement on single or multiple platforms. All the delay analysis models have been validated and deployed to the railway network in the Great Sydney area and the outcome of this application of the intelligent delay analysis technology significantly reduces delay-caused losses, increases the operation efficiency, and enables the train operating system to meet performance metrics and recover from incidents.
dc.format.extent	14
dc.language	en
dc.publisher	Wiley
dc.relation	iMOVE Australia Limited1-019
dc.relation.ispartof	Advances in Data Science and Analytics: Concepts and Paradigms
dc.relation.isbasedon	10.1002/9781119792826.ch6
dc.rights	info:eu-repo/semantics/closedAccess
dc.title	Data-driven delay analysis with applications to railway networks
dc.type	Chapter
pubs.organisational-group	/University of Technology Sydney
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology
pubs.organisational-group	/University of Technology Sydney/Faculty of Engineering and Information Technology/A/DRsch The Data Science Institute
utslib.copyright.status	closed_access	*
pubs.consider-herdc	false
dc.date.updated	2023-11-09T21:48:11Z
pubs.place-of-publication	USA
pubs.publication-status	Published
dc.location	USA

Abstract:

Reliability is one of the key evaluation criteria in railway service. Many factors contribute to the measure, such as delays spanning over spatiotemporal dimensions. A common method used to improve reliability is to design a better timetable to reduce the mutual influence between trains. Recently, machine learning shows great potential in improving the effectiveness and efficiency of decision-making to increase operational performance. For railway system management, this seems like an opportunity worth further exploration. In this chapter, we focus on analyzing railway disruptions and their impact on the traffic in the whole network. Specifically, we build three different data-driven models: (1) a conditional Bayes Net Model with the introduction of Markov property for delay propagation prediction, (2) a primary delay tracking back model, and (3) a dwell improvement evaluation model which can estimate the potential network benefits of the dwell improvement on single or multiple platforms. All the delay analysis models have been validated and deployed to the railway network in the Great Sydney area and the outcome of this application of the intelligent delay analysis technology significantly reduces delay-caused losses, increases the operation efficiency, and enables the train operating system to meet performance metrics and recover from incidents.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/173286