Large-scale public transport origin-destination matrix estimation via weighted graph and traffic features integration

Zhao, D; Mihaita, A; Ou, Y; Grzybowska, H; Shafiei, S; Qin, K; Dia, H; Tan, G

Large-scale public transport origin-destination matrix estimation via weighted graph and traffic features integration

Zhao, D Mihaita, A

Ou, Y Grzybowska, H Shafiei, S Qin, K Dia, H Tan, G

Permalink

Publication Type:: Conference Proceeding
Citation:: Transportation Research Board 103rd Annual Meeting TRB, 2023
Issue Date:: 2023-01-08

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Download Accepted versionAdobe PDF (4.25 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Zhao, D
dc.contributor.author	Mihaita, A https://orcid.org/0000-0001-7670-5777
dc.contributor.author	Ou, Y
dc.contributor.author	Grzybowska, H
dc.contributor.author	Shafiei, S
dc.contributor.author	Qin, K
dc.contributor.author	Dia, H
dc.contributor.author	Tan, G
dc.date	2023-01-08
dc.date.accessioned	2024-08-03T01:23:05Z
dc.date.available	2024-08-03T01:23:05Z
dc.date.issued	2023-01-08
dc.identifier.citation	Transportation Research Board 103rd Annual Meeting TRB, 2023
dc.identifier.uri	http://hdl.handle.net/10453/180006
dc.description.abstract	Estimation of the large-scale demand estimation for public transport in different cities can vary depending on the public transport network, public transport modes, and traffic data. To overcome the issue of traffic data shortage and effectively estimate the Origin-Destination (OD) matrix, we use the most accessible data: total boarding and alighting and public transport timetable, to capture the public transport dynamic patronage, and from this perspective, we establish a dynamic and microscopic OD matrix for public transport. In this paper, we propose a new method to model the dynamic large-scale stop-by-stop OD demand for public transport by developing a boosting of the gravity model via graph theory and Shannon's entropy. First, we propose a novel cost matrix estimation method that considers various sources of travel cost features extracted from both the traffic flow information in the traffic network and topological information in the graph network. Secondly, we proposed an ``Ensemble Cost Matrix Weighted by Entropy'' method to estimate the best weights of importance for each feature using Shannon's Entropy to maximise the performance of the cost matrix in OD matrix estimation. Third, we validate the proposed approach using real smart-card data. Last, by comparing the effectiveness of our proposed method with the traditional deterrence function-oriented methods, we prove that our proposed cost matrix estimation method cooperated OD matrix modelling method is superior in accurately OD matrix estimation to traditional methods by almost 54.46\% according to RMSE, 84.44\% according to MAPE, and 85.09\% according to MAE.
dc.language	en
dc.relation	http://purl.org/au-research/grants/arc/LP180100114
dc.relation.ispartof	Transportation Research Board 103rd Annual Meeting TRB 2023
dc.rights	info:eu-repo/semantics/openAccess
dc.title	Large-scale public transport origin-destination matrix estimation via weighted graph and traffic features integration
dc.type	Conference Proceeding
utslib.location.activity	Washington D.C.
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/All Manual Groups
pubs.organisational-group	University of Technology Sydney/All Manual Groups/Data Science Institute (DSI)
pubs.organisational-group	University of Technology Sydney/Strength - DSI - Data Science Institute
utslib.copyright.status	open_access	*
pubs.consider-herdc	false
dc.date.updated	2024-08-03T01:22:59Z
pubs.finish-date	2023-01-12
pubs.publication-status	Accepted
pubs.start-date	2023-01-08

Abstract:

Estimation of the large-scale demand estimation for public transport in different cities can vary depending on the public transport network, public transport modes, and traffic data. To overcome the issue of traffic data shortage and effectively estimate the Origin-Destination (OD) matrix, we use the most accessible data: total boarding and alighting and public transport timetable, to capture the public transport dynamic patronage, and from this perspective, we establish a dynamic and microscopic OD matrix for public transport. In this paper, we propose a new method to model the dynamic large-scale stop-by-stop OD demand for public transport by developing a boosting of the gravity model via graph theory and Shannon's entropy. First, we propose a novel cost matrix estimation method that considers various sources of travel cost features extracted from both the traffic flow information in the traffic network and topological information in the graph network. Secondly, we proposed an ``Ensemble Cost Matrix Weighted by Entropy'' method to estimate the best weights of importance for each feature using Shannon's Entropy to maximise the performance of the cost matrix in OD matrix estimation. Third, we validate the proposed approach using real smart-card data. Last, by comparing the effectiveness of our proposed method with the traditional deterrence function-oriented methods, we prove that our proposed cost matrix estimation method cooperated OD matrix modelling method is superior in accurately OD matrix estimation to traditional methods by almost 54.46\% according to RMSE, 84.44\% according to MAPE, and 85.09\% according to MAE.

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

http://hdl.handle.net/10453/180006