Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning

J Khiari; L Moreira-Matias; Vitor Cerqueira; Oded Cats

doi:10.1007/978-3-319-31753-3_44

Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning

J Khiari, L Moreira-Matias, Vitor Cerqueira, Oded Cats

Transport and Planning

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

16 Citations (Scopus)

41 Downloads (Pure)

Abstract

The efficiency of Public Transportation (PT) Networks is a major goal of any urban area authority. Advances on both location and communication devices drastically increased the availability of the data generated by their operations. Adequate Machine Learning methods can thus be applied to identify patterns useful to improve the Schedule Plan. In this paper, the authors propose a fully automated learning framework to determine the best Schedule Coverage to be assigned to a given PT network based on Automatic Vehicle location (AVL) and Automatic Passenger Counting (APC) data. We formulate this problem as a clustering one, where the best number of clusters is selected through an ad-hoc metric. This metric takes into account multiple domain constraints, computed using Sequence Mining and Probabilistic Reasoning. A case study from a large operator in Sweden was selected to validate our methodology. Experimental results suggest necessary changes on the Schedule coverage. Moreover, an impact study was conducted through a large-scale simulation over the affected time period. Its results uncovered potential improvements of the schedule reliability on a large scale.

Original language	English
Title of host publication	Advances in Knowledge Discovery and Data Mining
Editors	J. Bailey, L. Khan, T. Washio, G. Dobbie, J. Huang, R. Wang
Publisher	Springer
Pages	552-564
Number of pages	13
Volume	9651
ISBN (Electronic)	978-3-319-31753-3
ISBN (Print)	978-3-319-31752-6
DOIs	https://doi.org/10.1007/978-3-319-31753-3_44
Publication status	Published - 2016
Event	The 20th Pacific Asia Conference on Knowledge Discovery and Data Mining - Auckland, New Zealand Duration: 19 Apr 2016 → 22 Apr 2016 Conference number: 20

Publication series

Name	Lecture Notes in Computer Science (LNCS)
Publisher	Springer
Volume	9651
ISSN (Electronic)	0302-9743

Conference

Conference	The 20th Pacific Asia Conference on Knowledge Discovery and Data Mining
Abbreviated title	PAKDD 2016
Country/Territory	New Zealand
City	Auckland
Period	19/04/16 → 22/04/16

Keywords

Unsupervised learning
Public transportation
Big data
Schedule plan
Schedule coverage
Sequence mining
Probabilistic reasoning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-319-31753-3_44

bus schedule coverage2Accepted author manuscript, 390 KB

Cite this

Khiari, J., Moreira-Matias, L., Cerqueira, V., & Cats, O. (2016). Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning. In J. Bailey, L. Khan, T. Washio, G. Dobbie, J. Huang, & R. Wang (Eds.), Advances in Knowledge Discovery and Data Mining (Vol. 9651, pp. 552-564). (Lecture Notes in Computer Science (LNCS); Vol. 9651). Springer. https://doi.org/10.1007/978-3-319-31753-3_44

@inproceedings{b7ec5d96f18f4d3da845a97f018423a1,

title = "Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning",

abstract = "The efficiency of Public Transportation (PT) Networks is a major goal of any urban area authority. Advances on both location and communication devices drastically increased the availability of the data generated by their operations. Adequate Machine Learning methods can thus be applied to identify patterns useful to improve the Schedule Plan. In this paper, the authors propose a fully automated learning framework to determine the best Schedule Coverage to be assigned to a given PT network based on Automatic Vehicle location (AVL) and Automatic Passenger Counting (APC) data. We formulate this problem as a clustering one, where the best number of clusters is selected through an ad-hoc metric. This metric takes into account multiple domain constraints, computed using Sequence Mining and Probabilistic Reasoning. A case study from a large operator in Sweden was selected to validate our methodology. Experimental results suggest necessary changes on the Schedule coverage. Moreover, an impact study was conducted through a large-scale simulation over the affected time period. Its results uncovered potential improvements of the schedule reliability on a large scale.",

keywords = "Unsupervised learning, Public transportation, Big data, Schedule plan, Schedule coverage, Sequence mining, Probabilistic reasoning",

author = "J Khiari and L Moreira-Matias and Vitor Cerqueira and Oded Cats",

year = "2016",

doi = "10.1007/978-3-319-31753-3_44",

language = "English",

isbn = "978-3-319-31752-6",

volume = "9651",

series = "Lecture Notes in Computer Science (LNCS)",

publisher = "Springer",

pages = "552--564",

editor = "J. Bailey and L. Khan and T. Washio and G. Dobbie and J. Huang and R. Wang",

booktitle = "Advances in Knowledge Discovery and Data Mining",

note = "The 20th Pacific Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2016 ; Conference date: 19-04-2016 Through 22-04-2016",

}

Khiari, J, Moreira-Matias, L, Cerqueira, V & Cats, O 2016, Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning. in J Bailey, L Khan, T Washio, G Dobbie, J Huang & R Wang (eds), Advances in Knowledge Discovery and Data Mining . vol. 9651, Lecture Notes in Computer Science (LNCS), vol. 9651, Springer, pp. 552-564, The 20th Pacific Asia Conference on Knowledge Discovery and Data Mining, Auckland, New Zealand, 19/04/16. https://doi.org/10.1007/978-3-319-31753-3_44

Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning. / Khiari, J; Moreira-Matias, L; Cerqueira, Vitor et al.
Advances in Knowledge Discovery and Data Mining . ed. / J. Bailey; L. Khan; T. Washio; G. Dobbie; J. Huang; R. Wang. Vol. 9651 Springer, 2016. p. 552-564 (Lecture Notes in Computer Science (LNCS); Vol. 9651).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning

AU - Khiari, J

AU - Moreira-Matias, L

AU - Cerqueira, Vitor

AU - Cats, Oded

N1 - Conference code: 20

PY - 2016

Y1 - 2016

N2 - The efficiency of Public Transportation (PT) Networks is a major goal of any urban area authority. Advances on both location and communication devices drastically increased the availability of the data generated by their operations. Adequate Machine Learning methods can thus be applied to identify patterns useful to improve the Schedule Plan. In this paper, the authors propose a fully automated learning framework to determine the best Schedule Coverage to be assigned to a given PT network based on Automatic Vehicle location (AVL) and Automatic Passenger Counting (APC) data. We formulate this problem as a clustering one, where the best number of clusters is selected through an ad-hoc metric. This metric takes into account multiple domain constraints, computed using Sequence Mining and Probabilistic Reasoning. A case study from a large operator in Sweden was selected to validate our methodology. Experimental results suggest necessary changes on the Schedule coverage. Moreover, an impact study was conducted through a large-scale simulation over the affected time period. Its results uncovered potential improvements of the schedule reliability on a large scale.

AB - The efficiency of Public Transportation (PT) Networks is a major goal of any urban area authority. Advances on both location and communication devices drastically increased the availability of the data generated by their operations. Adequate Machine Learning methods can thus be applied to identify patterns useful to improve the Schedule Plan. In this paper, the authors propose a fully automated learning framework to determine the best Schedule Coverage to be assigned to a given PT network based on Automatic Vehicle location (AVL) and Automatic Passenger Counting (APC) data. We formulate this problem as a clustering one, where the best number of clusters is selected through an ad-hoc metric. This metric takes into account multiple domain constraints, computed using Sequence Mining and Probabilistic Reasoning. A case study from a large operator in Sweden was selected to validate our methodology. Experimental results suggest necessary changes on the Schedule coverage. Moreover, an impact study was conducted through a large-scale simulation over the affected time period. Its results uncovered potential improvements of the schedule reliability on a large scale.

KW - Unsupervised learning

KW - Public transportation

KW - Big data

KW - Schedule plan

KW - Schedule coverage

KW - Sequence mining

KW - Probabilistic reasoning

U2 - 10.1007/978-3-319-31753-3_44

DO - 10.1007/978-3-319-31753-3_44

M3 - Conference contribution

SN - 978-3-319-31752-6

VL - 9651

T3 - Lecture Notes in Computer Science (LNCS)

SP - 552

EP - 564

BT - Advances in Knowledge Discovery and Data Mining

A2 - Bailey, J.

A2 - Khan, L.

A2 - Washio, T.

A2 - Dobbie, G.

A2 - Huang, J.

A2 - Wang, R.

PB - Springer

T2 - The 20th Pacific Asia Conference on Knowledge Discovery and Data Mining

Y2 - 19 April 2016 through 22 April 2016

ER -

Automated Setting of Bus Schedule Coverage Using Unsupervised Machine Learning

Abstract

Publication series

Conference

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this