Hybrid data-driven approach for truck travel time imputation

Abolfazl Karimpour, Amin Ariannezhad, Yao Jan Wu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Truck travel time data plays a critical role in freight performance measurement and is usually collected with probevehicle technologies. However, due to low sampling rates, truck data usually suffers from missing values. The primary purpose of this study is to develop a hybrid model to accurately impute missing truck travel time data by leveraging multiple data sources. The proposed model imputes missing values by considering the interaction, similarity, and differences of the data as well as incorporating available historical information. The hybrid model achieves robust results by combining both probe vehicle and loop detector data to impute continuous missing truck travel time data in sparse datasets. The proposed model was used to impute missing truck travel time data in the National Performance Measures Research Dataset (NPMRDS). The imputation performance of the proposed model was compared with several popular imputation models including historical, spline interpolation, random forest, and bootstrapping EM. The results indicated that the proposed model was capable of imputing missing data in sparse datasets, notably when the data was missing continuously. With ∼13% mean-absolute percentage error, the hybrid model outperformed other models in imputing an entire day of missing data.

Original languageEnglish (US)
Pages (from-to)1518-1524
Number of pages7
JournalIET Intelligent Transport Systems
Volume13
Issue number10
DOIs
StatePublished - Oct 1 2019

Fingerprint

Travel time
travel time
Trucks
travel
time
bootstrapping
Splines
performance measurement
Interpolation
performance
interpolation
Values
Sampling
Detectors
probe

ASJC Scopus subject areas

  • Transportation
  • Environmental Science(all)
  • Mechanical Engineering
  • Law

Cite this

Hybrid data-driven approach for truck travel time imputation. / Karimpour, Abolfazl; Ariannezhad, Amin; Wu, Yao Jan.

In: IET Intelligent Transport Systems, Vol. 13, No. 10, 01.10.2019, p. 1518-1524.

Research output: Contribution to journalArticle

Karimpour, Abolfazl ; Ariannezhad, Amin ; Wu, Yao Jan. / Hybrid data-driven approach for truck travel time imputation. In: IET Intelligent Transport Systems. 2019 ; Vol. 13, No. 10. pp. 1518-1524.
@article{d6c0e15939c1430caaa7eb672e406604,
title = "Hybrid data-driven approach for truck travel time imputation",
abstract = "Truck travel time data plays a critical role in freight performance measurement and is usually collected with probevehicle technologies. However, due to low sampling rates, truck data usually suffers from missing values. The primary purpose of this study is to develop a hybrid model to accurately impute missing truck travel time data by leveraging multiple data sources. The proposed model imputes missing values by considering the interaction, similarity, and differences of the data as well as incorporating available historical information. The hybrid model achieves robust results by combining both probe vehicle and loop detector data to impute continuous missing truck travel time data in sparse datasets. The proposed model was used to impute missing truck travel time data in the National Performance Measures Research Dataset (NPMRDS). The imputation performance of the proposed model was compared with several popular imputation models including historical, spline interpolation, random forest, and bootstrapping EM. The results indicated that the proposed model was capable of imputing missing data in sparse datasets, notably when the data was missing continuously. With ∼13{\%} mean-absolute percentage error, the hybrid model outperformed other models in imputing an entire day of missing data.",
author = "Abolfazl Karimpour and Amin Ariannezhad and Wu, {Yao Jan}",
year = "2019",
month = "10",
day = "1",
doi = "10.1049/iet-its.2018.5469",
language = "English (US)",
volume = "13",
pages = "1518--1524",
journal = "IET Intelligent Transport Systems",
issn = "1751-956X",
publisher = "Institution of Engineering and Technology",
number = "10",

}

TY - JOUR

T1 - Hybrid data-driven approach for truck travel time imputation

AU - Karimpour, Abolfazl

AU - Ariannezhad, Amin

AU - Wu, Yao Jan

PY - 2019/10/1

Y1 - 2019/10/1

N2 - Truck travel time data plays a critical role in freight performance measurement and is usually collected with probevehicle technologies. However, due to low sampling rates, truck data usually suffers from missing values. The primary purpose of this study is to develop a hybrid model to accurately impute missing truck travel time data by leveraging multiple data sources. The proposed model imputes missing values by considering the interaction, similarity, and differences of the data as well as incorporating available historical information. The hybrid model achieves robust results by combining both probe vehicle and loop detector data to impute continuous missing truck travel time data in sparse datasets. The proposed model was used to impute missing truck travel time data in the National Performance Measures Research Dataset (NPMRDS). The imputation performance of the proposed model was compared with several popular imputation models including historical, spline interpolation, random forest, and bootstrapping EM. The results indicated that the proposed model was capable of imputing missing data in sparse datasets, notably when the data was missing continuously. With ∼13% mean-absolute percentage error, the hybrid model outperformed other models in imputing an entire day of missing data.

AB - Truck travel time data plays a critical role in freight performance measurement and is usually collected with probevehicle technologies. However, due to low sampling rates, truck data usually suffers from missing values. The primary purpose of this study is to develop a hybrid model to accurately impute missing truck travel time data by leveraging multiple data sources. The proposed model imputes missing values by considering the interaction, similarity, and differences of the data as well as incorporating available historical information. The hybrid model achieves robust results by combining both probe vehicle and loop detector data to impute continuous missing truck travel time data in sparse datasets. The proposed model was used to impute missing truck travel time data in the National Performance Measures Research Dataset (NPMRDS). The imputation performance of the proposed model was compared with several popular imputation models including historical, spline interpolation, random forest, and bootstrapping EM. The results indicated that the proposed model was capable of imputing missing data in sparse datasets, notably when the data was missing continuously. With ∼13% mean-absolute percentage error, the hybrid model outperformed other models in imputing an entire day of missing data.

UR - http://www.scopus.com/inward/record.url?scp=85072726463&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072726463&partnerID=8YFLogxK

U2 - 10.1049/iet-its.2018.5469

DO - 10.1049/iet-its.2018.5469

M3 - Article

AN - SCOPUS:85072726463

VL - 13

SP - 1518

EP - 1524

JO - IET Intelligent Transport Systems

JF - IET Intelligent Transport Systems

SN - 1751-956X

IS - 10

ER -