Understanding dwell times using automatic passenger count data: A quantile regression approach
(2024) In Journal of Rail Transport Planning and Management 29.- Abstract
- Accurately scheduling dwell times is vital to ensure punctual and reliable railway services, but the stochastic nature of dwell times makes this a non-trivial task. An important step towards scheduling accurate dwell times is to gain an in-depth understanding of the mechanics that influence dwell times, which is commonly done by modelling the mean dwell time. It is, however, of more interest to understand the conditional distribution of dwell times. The study presented here proposes the use of quantile regression to study the conditional distribution of dwell times at different percentile. To do so, a year's worth of highly detailed train operation and passenger count data is used. The results indicate that the use of quantile regression... (More)
- Accurately scheduling dwell times is vital to ensure punctual and reliable railway services, but the stochastic nature of dwell times makes this a non-trivial task. An important step towards scheduling accurate dwell times is to gain an in-depth understanding of the mechanics that influence dwell times, which is commonly done by modelling the mean dwell time. It is, however, of more interest to understand the conditional distribution of dwell times. The study presented here proposes the use of quantile regression to study the conditional distribution of dwell times at different percentile. To do so, a year's worth of highly detailed train operation and passenger count data is used. The results indicate that the use of quantile regression over ordinary least squares regression is justifiable and beneficial. Numerical examples show the importance of arrival punctuality on dwell times, whereas the effect of the volume of boarding passengers at the critical door is limited. The results of the model presented here can help steer the discourse towards scheduling dwell times that more accurately reflect the actual situation by taking station-specific parameters into account. Doing so will help to increase the punctuality of railways and with it the attractiveness and effectiveness of railways. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/e14e2095-c96e-4cad-b776-1dddd224e1e7
- author
- Kuipers, Ruben LU
- organization
- publishing date
- 2024-01-25
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Dwell time, Quantile regression, Automatic passenger count data, Passengers
- in
- Journal of Rail Transport Planning and Management
- volume
- 29
- pages
- 16 pages
- publisher
- Elsevier
- ISSN
- 2210-9706
- DOI
- 10.1016/j.jrtpm.2024.100431
- project
- Punctual Metropolitan Railways – an analysis of delays at stations and effects of altered dwell time planning approaches and effective traveller exchanges
- language
- English
- LU publication?
- yes
- id
- e14e2095-c96e-4cad-b776-1dddd224e1e7
- date added to LUP
- 2024-02-12 09:39:11
- date last changed
- 2024-02-13 12:42:57
@article{e14e2095-c96e-4cad-b776-1dddd224e1e7,
abstract = {{Accurately scheduling dwell times is vital to ensure punctual and reliable railway services, but the stochastic nature of dwell times makes this a non-trivial task. An important step towards scheduling accurate dwell times is to gain an in-depth understanding of the mechanics that influence dwell times, which is commonly done by modelling the mean dwell time. It is, however, of more interest to understand the conditional distribution of dwell times. The study presented here proposes the use of quantile regression to study the conditional distribution of dwell times at different percentile. To do so, a year's worth of highly detailed train operation and passenger count data is used. The results indicate that the use of quantile regression over ordinary least squares regression is justifiable and beneficial. Numerical examples show the importance of arrival punctuality on dwell times, whereas the effect of the volume of boarding passengers at the critical door is limited. The results of the model presented here can help steer the discourse towards scheduling dwell times that more accurately reflect the actual situation by taking station-specific parameters into account. Doing so will help to increase the punctuality of railways and with it the attractiveness and effectiveness of railways.}},
author = {{Kuipers, Ruben}},
issn = {{2210-9706}},
keywords = {{Dwell time; Quantile regression; Automatic passenger count data; Passengers}},
language = {{eng}},
month = {{01}},
publisher = {{Elsevier}},
series = {{Journal of Rail Transport Planning and Management}},
title = {{Understanding dwell times using automatic passenger count data: A quantile regression approach}},
url = {{http://dx.doi.org/10.1016/j.jrtpm.2024.100431}},
doi = {{10.1016/j.jrtpm.2024.100431}},
volume = {{29}},
year = {{2024}},
}