A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis

Highway-rail grade crossing (HRGC) crashes continue to be the major contributors to rail causalities in the United States and have been intensively researched in the past. Data-mining models focus on prediction while dominant general linear models focus on model and data fitness. Decision makers and...

Full description

Saved in:
Bibliographic Details
Main Authors: Pan Lu, Zijian Zheng, Yihao Ren, Xiaoyi Zhou, Amin Keramati, Denver Tolliver, Ying Huang
Format: Article
Language:English
Published: Wiley 2020-01-01
Series:Journal of Advanced Transportation
Online Access:http://dx.doi.org/10.1155/2020/6751728
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849684715014979584
author Pan Lu
Zijian Zheng
Yihao Ren
Xiaoyi Zhou
Amin Keramati
Denver Tolliver
Ying Huang
author_facet Pan Lu
Zijian Zheng
Yihao Ren
Xiaoyi Zhou
Amin Keramati
Denver Tolliver
Ying Huang
author_sort Pan Lu
collection DOAJ
description Highway-rail grade crossing (HRGC) crashes continue to be the major contributors to rail causalities in the United States and have been intensively researched in the past. Data-mining models focus on prediction while dominant general linear models focus on model and data fitness. Decision makers and traffic engineers rely on prediction models to examine at-grade crash frequency and make safety improvement. The gradient boosting (GB) model has gained popularity in many research areas. In this study, to fully understand the model performance on HRGC accident prediction performance, the GB model with functional gradient descent algorithm is selected to analyze crashes at highway-rail grade crossings (HRGCs) and to identify contributor factors. Moreover, contributors’ importance and partial-dependent relations are generated to further understand the relationship of identified contributors and HRGC crash likelihood to concur “black box” issues that most machine learning methods face. Furthermore, to fully demonstrate the model’s prediction performance, a comprehensive model prediction power assessment based on six measurements is conducted, and the prediction performance of the GB model is verified and compared with a decision tree model as a reference due to their popularity and comparable data availability. It is demonstrated that the GB model produces better prediction accuracy and reveals nonlinear relationships among contributors and crash likelihood. In general, HRGC crash likelihood is significantly impacted by several traffic exposure factors: highway traffic volume, railway traffic volume, and train travel speed and others.
format Article
id doaj-art-c032306037d24f3f839f2fe12c50bcd0
institution DOAJ
issn 0197-6729
2042-3195
language English
publishDate 2020-01-01
publisher Wiley
record_format Article
series Journal of Advanced Transportation
spelling doaj-art-c032306037d24f3f839f2fe12c50bcd02025-08-20T03:23:23ZengWileyJournal of Advanced Transportation0197-67292042-31952020-01-01202010.1155/2020/67517286751728A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash AnalysisPan Lu0Zijian Zheng1Yihao Ren2Xiaoyi Zhou3Amin Keramati4Denver Tolliver5Ying Huang6Department of Transportation, Logistics, and Finance, Upper Great Plains Transportation Institute, North Dakota State University NDSU, Dept 2880, P. O. Box 6050, Fargo, ND 58108-6050, USAGates Corporation, 1144 Fifteenth St. Suite 1400, Denver, CO 80202, USAUpper Great Plains Transportation Institute, North Dakota State University NDSU, Dept 2880 P. O. Box 6050, Fargo, ND 58108-6050, USAUpper Great Plains Transportation Institute, North Dakota State University NDSU, Dept 2880 P. O. Box 6050, Fargo, ND 58108-6050, USAUpper Great Plains Transportation Institute, North Dakota State University NDSU, Dept 2880 P. O. Box 6050, Fargo, ND 58108-6050, USAUpper Great Plains Transportation Institute, North Dakota State University NDSU, Dept 2880 P. O. Box 6050, Fargo, ND 58108-6050, USADepartment of Civil and Environmental Engineering, North Dakota State University NDSU, Dept 2470, P. O. Box 6050, Fargo, ND 58108-6050, USAHighway-rail grade crossing (HRGC) crashes continue to be the major contributors to rail causalities in the United States and have been intensively researched in the past. Data-mining models focus on prediction while dominant general linear models focus on model and data fitness. Decision makers and traffic engineers rely on prediction models to examine at-grade crash frequency and make safety improvement. The gradient boosting (GB) model has gained popularity in many research areas. In this study, to fully understand the model performance on HRGC accident prediction performance, the GB model with functional gradient descent algorithm is selected to analyze crashes at highway-rail grade crossings (HRGCs) and to identify contributor factors. Moreover, contributors’ importance and partial-dependent relations are generated to further understand the relationship of identified contributors and HRGC crash likelihood to concur “black box” issues that most machine learning methods face. Furthermore, to fully demonstrate the model’s prediction performance, a comprehensive model prediction power assessment based on six measurements is conducted, and the prediction performance of the GB model is verified and compared with a decision tree model as a reference due to their popularity and comparable data availability. It is demonstrated that the GB model produces better prediction accuracy and reveals nonlinear relationships among contributors and crash likelihood. In general, HRGC crash likelihood is significantly impacted by several traffic exposure factors: highway traffic volume, railway traffic volume, and train travel speed and others.http://dx.doi.org/10.1155/2020/6751728
spellingShingle Pan Lu
Zijian Zheng
Yihao Ren
Xiaoyi Zhou
Amin Keramati
Denver Tolliver
Ying Huang
A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis
Journal of Advanced Transportation
title A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis
title_full A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis
title_fullStr A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis
title_full_unstemmed A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis
title_short A Gradient Boosting Crash Prediction Approach for Highway-Rail Grade Crossing Crash Analysis
title_sort gradient boosting crash prediction approach for highway rail grade crossing crash analysis
url http://dx.doi.org/10.1155/2020/6751728
work_keys_str_mv AT panlu agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT zijianzheng agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT yihaoren agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT xiaoyizhou agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT aminkeramati agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT denvertolliver agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT yinghuang agradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT panlu gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT zijianzheng gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT yihaoren gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT xiaoyizhou gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT aminkeramati gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT denvertolliver gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis
AT yinghuang gradientboostingcrashpredictionapproachforhighwayrailgradecrossingcrashanalysis