Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems

Cheng Feng; Jane Hillston; Daniel Reijsbergen

doi:10.1007/978-3-319-43425-4_9

Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems

Cheng Feng, Jane Hillston, Daniel Reijsbergen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We study the problem of future bike availability prediction of a bike station through the moment analysis of a PCTMC model with time-dependent rates. Given a target station for prediction, the moments of the number of available bikes in the station at a future time can be derived by a set of moment equations with an initial set-up given by the snapshot of the current state of all stations in the system. A directed contribution graph with contribution propagation method is proposed to prune the PCTMC to make it only contain stations which have significant contribution to the journey ows to the target station. The underlying probability distribution of the available number of bikes is reconstructed through the maximum entropy approach based on the derived moments. The model is parametrized using historical data from Santander Cycles, the bike-sharing system in London. In the experiments, we show our
model outperforms the classic time-inhomogeneous queueing model on several performance metrics for bike availability prediction.

Original language	English
Title of host publication	Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016)
Subtitle of host publication	Québec City, Québec, Canada August 23-25 2016
Publisher	Springer
Pages	139-155
Number of pages	16
ISBN (Electronic)	978-3-319-43425-4
ISBN (Print)	978-3-319-43424-7
DOIs	https://doi.org/10.1007/978-3-319-43425-4_9
Publication status	Published - 3 Aug 2016
Event	13th International Conference on Quantitative Evaluation of SysTems - Quebec City, Canada Duration: 23 Aug 2016 → 25 Aug 2016 http://www.qest.org/qest2016/

Publication series

Name	Lecture Notes In Computer Science
Publisher	Springer, Cham
Volume	9826
ISSN (Print)	0302-9743

Conference

Conference	13th International Conference on Quantitative Evaluation of SysTems
Abbreviated title	QEST 2016
Country/Territory	Canada
City	Quebec City
Period	23/08/16 → 25/08/16
Internet address	http://www.qest.org/qest2016/

Access to Document

10.1007/978-3-319-43425-4_9

Feng et al 2016 Moment Based Probablitstic PredictionAccepted author manuscript, 346 KB

https://link.springer.com/chapter/10.1007/978-3-319-43425-4_9

Cite this

Feng, C., Hillston, J., & Reijsbergen, D. (2016). Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems. In Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016) : Québec City, Québec, Canada August 23-25 2016 (pp. 139-155). (Lecture Notes In Computer Science; Vol. 9826). Springer. https://doi.org/10.1007/978-3-319-43425-4_9

@inproceedings{73b37cc1b4a34a14ab5fbb95d99762ac,

title = "Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems",

abstract = "We study the problem of future bike availability prediction of a bike station through the moment analysis of a PCTMC model with time-dependent rates. Given a target station for prediction, the moments of the number of available bikes in the station at a future time can be derived by a set of moment equations with an initial set-up given by the snapshot of the current state of all stations in the system. A directed contribution graph with contribution propagation method is proposed to prune the PCTMC to make it only contain stations which have significant contribution to the journey ows to the target station. The underlying probability distribution of the available number of bikes is reconstructed through the maximum entropy approach based on the derived moments. The model is parametrized using historical data from Santander Cycles, the bike-sharing system in London. In the experiments, we show ourmodel outperforms the classic time-inhomogeneous queueing model on several performance metrics for bike availability prediction.",

author = "Cheng Feng and Jane Hillston and Daniel Reijsbergen",

year = "2016",

month = aug,

day = "3",

doi = "10.1007/978-3-319-43425-4_9",

language = "English",

isbn = "978-3-319-43424-7",

series = "Lecture Notes In Computer Science",

publisher = "Springer",

pages = "139--155",

booktitle = "Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016)",

address = "United Kingdom",

note = "13th International Conference on Quantitative Evaluation of SysTems , QEST 2016 ; Conference date: 23-08-2016 Through 25-08-2016",

url = "http://www.qest.org/qest2016/",

}

Feng, C, Hillston, J & Reijsbergen, D 2016, Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems. in Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016) : Québec City, Québec, Canada August 23-25 2016. Lecture Notes In Computer Science, vol. 9826, Springer, pp. 139-155, 13th International Conference on Quantitative Evaluation of SysTems , Quebec City, Quebec, Canada, 23/08/16. https://doi.org/10.1007/978-3-319-43425-4_9

Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems. / Feng, Cheng; Hillston, Jane; Reijsbergen, Daniel.
Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016) : Québec City, Québec, Canada August 23-25 2016. Springer, 2016. p. 139-155 (Lecture Notes In Computer Science; Vol. 9826).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems

AU - Feng, Cheng

AU - Hillston, Jane

AU - Reijsbergen, Daniel

PY - 2016/8/3

Y1 - 2016/8/3

N2 - We study the problem of future bike availability prediction of a bike station through the moment analysis of a PCTMC model with time-dependent rates. Given a target station for prediction, the moments of the number of available bikes in the station at a future time can be derived by a set of moment equations with an initial set-up given by the snapshot of the current state of all stations in the system. A directed contribution graph with contribution propagation method is proposed to prune the PCTMC to make it only contain stations which have significant contribution to the journey ows to the target station. The underlying probability distribution of the available number of bikes is reconstructed through the maximum entropy approach based on the derived moments. The model is parametrized using historical data from Santander Cycles, the bike-sharing system in London. In the experiments, we show ourmodel outperforms the classic time-inhomogeneous queueing model on several performance metrics for bike availability prediction.

AB - We study the problem of future bike availability prediction of a bike station through the moment analysis of a PCTMC model with time-dependent rates. Given a target station for prediction, the moments of the number of available bikes in the station at a future time can be derived by a set of moment equations with an initial set-up given by the snapshot of the current state of all stations in the system. A directed contribution graph with contribution propagation method is proposed to prune the PCTMC to make it only contain stations which have significant contribution to the journey ows to the target station. The underlying probability distribution of the available number of bikes is reconstructed through the maximum entropy approach based on the derived moments. The model is parametrized using historical data from Santander Cycles, the bike-sharing system in London. In the experiments, we show ourmodel outperforms the classic time-inhomogeneous queueing model on several performance metrics for bike availability prediction.

UR - http://www.qest.org/qest2016/

U2 - 10.1007/978-3-319-43425-4_9

DO - 10.1007/978-3-319-43425-4_9

M3 - Conference contribution

SN - 978-3-319-43424-7

T3 - Lecture Notes In Computer Science

SP - 139

EP - 155

BT - Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016)

PB - Springer

T2 - 13th International Conference on Quantitative Evaluation of SysTems

Y2 - 23 August 2016 through 25 August 2016

ER -

Feng C, Hillston J, Reijsbergen D. Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems. In Proceedings for the 13th International Conference on Quantitative Evaluation of SysTems (QEST 2016) : Québec City, Québec, Canada August 23-25 2016. Springer. 2016. p. 139-155. (Lecture Notes In Computer Science). doi: 10.1007/978-3-319-43425-4_9

Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

QUANTICOL - A Quantitative Approach to Management and Design of Collective and Adaptive Behaviours (RTD)

Cite this

Moment-based Probabilistic Prediction of Bike Availability for Bike-Sharing Systems

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Projects

QUANTICOL - A Quantitative Approach to Management and Design of Collective and Adaptive Behaviours (RTD)

Cite this