PREDICTION OF REAL-TIME TRAIN ARRIVAL TIMES ALONG THE SWEDISH SOUTHERN MAINLINE
(2022) 18th International Conference on Railway Engineering Design and Operation, COMPRAIL 2022 In WIT Transactions on the Built Environment 213. p.135-143- Abstract
Real-time train arrival time prediction is crucial for providing passenger information and timely decision support. The paper develops methods to simultaneously predict train arrival times at downstream stations, including direct multiple output liner regression (DMOLR) and seemingly unrelated regression (SUR) models. To capture correlations of prediction equations, two bias correction terms are tested: (1) one-step prior prediction error and (2) upstream prediction errors. The models are validated on high-speed trains operation data along the Swedish Southern Mainline from 2016 to 2020. The results show that the DMOLR model slightly outperforms the SUR. The DMOLR’s prediction performance improves up to 0.32% and 24.03% in term of RMSE... (More)
Real-time train arrival time prediction is crucial for providing passenger information and timely decision support. The paper develops methods to simultaneously predict train arrival times at downstream stations, including direct multiple output liner regression (DMOLR) and seemingly unrelated regression (SUR) models. To capture correlations of prediction equations, two bias correction terms are tested: (1) one-step prior prediction error and (2) upstream prediction errors. The models are validated on high-speed trains operation data along the Swedish Southern Mainline from 2016 to 2020. The results show that the DMOLR model slightly outperforms the SUR. The DMOLR’s prediction performance improves up to 0.32% and 24.03% in term of RMSE and R2 respectively when upstream prediction errors are considered.
(Less)
- author
- Tiong, Kah Yong
LU
; Ma, Zhenliang
and Palmqvist, Carl William
LU
- organization
- publishing date
- 2022-11-30
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- direct multiple output liner regression, seemingly unrelated regression, train arrival time predictions
- host publication
- Computers in Railways XVIII : Railway Engineering Design and Operation - Railway Engineering Design and Operation
- series title
- WIT Transactions on the Built Environment
- editor
- Passerini, Giorgio and Mera, Jose Manuel
- volume
- 213
- pages
- 9 pages
- publisher
- WIT Press
- conference name
- 18th International Conference on Railway Engineering Design and Operation, COMPRAIL 2022
- conference location
- Virtual, Online, Spain
- conference dates
- 2022-09-21 - 2022-09-23
- external identifiers
-
- scopus:85144007572
- ISSN
- 1746-4498
- 1743-3509
- ISBN
- 9781784664794
- DOI
- 10.2495/CR220121
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2022 WIT Press.
- id
- b4cf0c1d-9f28-4318-8427-690d18ee483c
- date added to LUP
- 2022-12-26 10:42:10
- date last changed
- 2024-05-30 16:58:03
@inproceedings{b4cf0c1d-9f28-4318-8427-690d18ee483c, abstract = {{<p>Real-time train arrival time prediction is crucial for providing passenger information and timely decision support. The paper develops methods to simultaneously predict train arrival times at downstream stations, including direct multiple output liner regression (DMOLR) and seemingly unrelated regression (SUR) models. To capture correlations of prediction equations, two bias correction terms are tested: (1) one-step prior prediction error and (2) upstream prediction errors. The models are validated on high-speed trains operation data along the Swedish Southern Mainline from 2016 to 2020. The results show that the DMOLR model slightly outperforms the SUR. The DMOLR’s prediction performance improves up to 0.32% and 24.03% in term of RMSE and R<sup>2</sup> respectively when upstream prediction errors are considered.</p>}}, author = {{Tiong, Kah Yong and Ma, Zhenliang and Palmqvist, Carl William}}, booktitle = {{Computers in Railways XVIII : Railway Engineering Design and Operation}}, editor = {{Passerini, Giorgio and Mera, Jose Manuel}}, isbn = {{9781784664794}}, issn = {{1746-4498}}, keywords = {{direct multiple output liner regression; seemingly unrelated regression; train arrival time predictions}}, language = {{eng}}, month = {{11}}, pages = {{135--143}}, publisher = {{WIT Press}}, series = {{WIT Transactions on the Built Environment}}, title = {{PREDICTION OF REAL-TIME TRAIN ARRIVAL TIMES ALONG THE SWEDISH SOUTHERN MAINLINE}}, url = {{http://dx.doi.org/10.2495/CR220121}}, doi = {{10.2495/CR220121}}, volume = {{213}}, year = {{2022}}, }