A Two-Stage Self-adaptive Model for Passenger Flow Prediction on Schedule-Based Railway System

Li, B; Guo, T; Li, R; Wang, Y; Gandomi, AH; Chen, F

A Two-Stage Self-adaptive Model for Passenger Flow Prediction on Schedule-Based Railway System

Li, B Guo, T

Li, R Wang, Y Gandomi, AH Chen, F

Permalink

Publisher:: SPRINGER INTERNATIONAL PUBLISHING AG
Publication Type:: Conference Proceeding
Citation:: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2022, 13282 LNAI, pp. 147-160
Issue Date:: 2022-01-01

Closed Access

	Filename	Description	Size
	A Two-Stage Self-adaptive Model for Passenger Flow Prediction on Schedule-Based Railway System.pdf	Published version	38.26 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	Li, R
dc.contributor.author	Wang, Y
dc.contributor.author	Gandomi, AH
dc.contributor.author	Chen, F https://orcid.org/0000-0003-4971-8729
dc.contributor.editor	Gama, J
dc.contributor.editor	Li, T
dc.contributor.editor	Yu, Y
dc.contributor.editor	Chen, E
dc.contributor.editor	Zheng, Y
dc.contributor.editor	Teng, F
dc.date	2022-05-16
dc.date.accessioned	2023-04-10T23:52:55Z
dc.date.available	2023-04-10T23:52:55Z
dc.date.issued	2022-01-01
dc.identifier.citation	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2022, 13282 LNAI, pp. 147-160
dc.identifier.isbn	9783031059803
dc.identifier.issn	0302-9743
dc.identifier.issn	1611-3349
dc.identifier.uri	http://hdl.handle.net/10453/169450
dc.description.abstract	Platform-level passenger flow prediction is crucial for addressing the overcrowding problem on platforms that endangered the passengers’ safety and experience in railway systems. Although some studies exist on this topic, it remains difficult to apply these methods in the real world e.g., the data deficiency in older railway systems, potential impacts of dynamic interchange passenger flows, real-time predictive ability. Thus, we propose a two-stage self-adaptive model for accurately and timely predicting platform-level flow. In the first stage, a self-attention-based prediction model is introduced to predict the next-day passenger flow based on the historical boarding record. The proposed decomposing components transferring the discrete boarding records into continuous patterns make the module able to deliver a robust minute-level prediction. In the second stage, a real-time fine-tuning model is developed to adjust the predicted flow based on the real-time emergencies in passenger flows. The combination of offline deep learning mechanism and real-time reallocation algorithm ensures the real-time response without loss of accuracy. The experiments show that our model can offer accurate predictions to trip planners for timetable design and provide timely decision support for controllers when emergencies happen, and our end-to-end framework has been applied to the railway system in Sydney, Australia.
dc.language	en
dc.publisher	SPRINGER INTERNATIONAL PUBLISHING AG
dc.relation	iMOVE Australia Limited1-019
dc.relation.ispartof	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
dc.relation.ispartof	26th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD)
dc.relation.ispartofseries	Lecture Notes in Artificial Intelligence
dc.relation.isbasedon	10.1007/978-3-031-05981-0_12
dc.rights	info:eu-repo/semantics/closedAccess
dc.subject.classification	Artificial Intelligence & Image Processing
dc.title	A Two-Stage Self-adaptive Model for Passenger Flow Prediction on Schedule-Based Railway System
dc.type	Conference Proceeding
utslib.citation.volume	13282 LNAI
utslib.location.activity	SW Jiaotong Univ, Chengdu, PEOPLES R CHINA
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	*
dc.date.updated	2023-04-10T23:52:51Z
pubs.finish-date	2022-05-19
pubs.publication-status	Published
pubs.start-date	2022-05-16
pubs.volume	13282 LNAI

Abstract:

Platform-level passenger flow prediction is crucial for addressing the overcrowding problem on platforms that endangered the passengers’ safety and experience in railway systems. Although some studies exist on this topic, it remains difficult to apply these methods in the real world e.g., the data deficiency in older railway systems, potential impacts of dynamic interchange passenger flows, real-time predictive ability. Thus, we propose a two-stage self-adaptive model for accurately and timely predicting platform-level flow. In the first stage, a self-attention-based prediction model is introduced to predict the next-day passenger flow based on the historical boarding record. The proposed decomposing components transferring the discrete boarding records into continuous patterns make the module able to deliver a robust minute-level prediction. In the second stage, a real-time fine-tuning model is developed to adjust the predicted flow based on the real-time emergencies in passenger flows. The combination of offline deep learning mechanism and real-time reallocation algorithm ensures the real-time response without loss of accuracy. The experiments show that our model can offer accurate predictions to trip planners for timetable design and provide timely decision support for controllers when emergencies happen, and our end-to-end framework has been applied to the railway system in Sydney, Australia.

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

http://hdl.handle.net/10453/169450