Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Comparison of Train Delay Prediction Models

Tiong, Kah Yong LU ; Ma, Zhenliang and Palmqvist, Carl William LU orcid (2024) 26th Euro Working Group on Transportation Meeting, EWGT 2024 In Transportation Research Procedia 86. p.900-907
Abstract

The growing need for intelligent transportation systems (ITS) to provide advanced train-related information has driven the development of train delay prediction models. The study used a case study in Sweden to illustrate how various prediction models can be evaluated under various aspects and dimensions in order to determine their effectiveness in practical implementation. The outputs of three prediction models—Manual Method, Digital Graph System, and AI Solution—are collected over a three-month period across 178 train stations to serve as the study’s primary data sources. The results show that AI Solution outperforms the models currently used in the railway industry. It also finds that AI Solution can be used as reliable support tools... (More)

The growing need for intelligent transportation systems (ITS) to provide advanced train-related information has driven the development of train delay prediction models. The study used a case study in Sweden to illustrate how various prediction models can be evaluated under various aspects and dimensions in order to determine their effectiveness in practical implementation. The outputs of three prediction models—Manual Method, Digital Graph System, and AI Solution—are collected over a three-month period across 178 train stations to serve as the study’s primary data sources. The results show that AI Solution outperforms the models currently used in the railway industry. It also finds that AI Solution can be used as reliable support tools for train operators and passengers by providing reliable information on expected train movement, thereby enabling end users to make better decisions and take better actions in response to potential train delays.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
data-driven prediction, performance evaluation, Train delay prediction
host publication
Transportation Research Procedia
series title
Transportation Research Procedia
volume
86
pages
8 pages
publisher
Elsevier
conference name
26th Euro Working Group on Transportation Meeting, EWGT 2024
conference location
Lund, Sweden
conference dates
2024-09-04 - 2024-09-06
external identifiers
  • scopus:105048894041
ISSN
2352-1457
DOI
10.1016/j.trpro.2026.07.016
language
English
LU publication?
yes
additional info
Publisher Copyright: Copyright © 2026. Published by Elsevier B.V.
id
c951ca51-6e3e-4d0d-bc55-1fba1fde4ece
date added to LUP
2026-09-10 09:08:58
date last changed
2026-09-14 11:59:24
@inproceedings{c951ca51-6e3e-4d0d-bc55-1fba1fde4ece,
  abstract     = {{<p>The growing need for intelligent transportation systems (ITS) to provide advanced train-related information has driven the development of train delay prediction models. The study used a case study in Sweden to illustrate how various prediction models can be evaluated under various aspects and dimensions in order to determine their effectiveness in practical implementation. The outputs of three prediction models—Manual Method, Digital Graph System, and AI Solution—are collected over a three-month period across 178 train stations to serve as the study’s primary data sources. The results show that AI Solution outperforms the models currently used in the railway industry. It also finds that AI Solution can be used as reliable support tools for train operators and passengers by providing reliable information on expected train movement, thereby enabling end users to make better decisions and take better actions in response to potential train delays.</p>}},
  author       = {{Tiong, Kah Yong and Ma, Zhenliang and Palmqvist, Carl William}},
  booktitle    = {{Transportation Research Procedia}},
  issn         = {{2352-1457}},
  keywords     = {{data-driven prediction; performance evaluation; Train delay prediction}},
  language     = {{eng}},
  pages        = {{900--907}},
  publisher    = {{Elsevier}},
  series       = {{Transportation Research Procedia}},
  title        = {{Comparison of Train Delay Prediction Models}},
  url          = {{http://dx.doi.org/10.1016/j.trpro.2026.07.016}},
  doi          = {{10.1016/j.trpro.2026.07.016}},
  volume       = {{86}},
  year         = {{2024}},
}