Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA

Driving at night is riskier in terms of crash involvement than it is during the day. Fortunately, it is clearly established that illumination on roadways can reduce the number and severity of nighttime crashes. However, state and municipal departments of transportation (DOTs) lack the available illu...

Full description

Saved in:
Bibliographic Details
Main Authors: Raja Daoud, Matthew Vechione, Okan Gurbuz, Prabha Sundaravadivel, Chi Tian
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Vehicles
Subjects:
Online Access:https://www.mdpi.com/2624-8921/7/1/20
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849340104634531840
author Raja Daoud
Matthew Vechione
Okan Gurbuz
Prabha Sundaravadivel
Chi Tian
author_facet Raja Daoud
Matthew Vechione
Okan Gurbuz
Prabha Sundaravadivel
Chi Tian
author_sort Raja Daoud
collection DOAJ
description Driving at night is riskier in terms of crash involvement than it is during the day. Fortunately, it is clearly established that illumination on roadways can reduce the number and severity of nighttime crashes. However, state and municipal departments of transportation (DOTs) lack the available illumination data. Therefore, the objective of this research is threefold, as follows: (i) to develop machine learning models that use readily available roadway characteristic data to predict the severity of nighttime crashes; (ii) determine the effect that illumination has on crash severity; and (iii) develop a tool to assist DOT decision makers in collecting illumination data. To accomplish this objective, we have extracted data from the Texas Department of Transportation (TxDOT) Crash Record Information System (CRIS) database, which was then further split into a training and a test dataset. Then, seven machine learning techniques, namely binary logistic regression, k-nearest neighbors, naïve Bayes, random forest, artificial neural network, Extreme Gradient Boosting (XGBoost), and a Long Short-Term Memory (LSTM) model, were all applied to the unseen test data. The random forest model produced the most promising results by predicting severe crashes with 97.6% accuracy. In addition, we conducted a pilot study to test the collection of illumination data using a light meter. In the future, we aim to complete the development of a smartphone application, which can be used in conjunction with the random forest model presented in this paper, to collect crowdsourced illumination data and predict nighttime crash hotspots. This may assist DOT decision makers to prioritize funding for illumination at the hot spots.
format Article
id doaj-art-ccceeee406524147bfcd5dc57bcb5a17
institution Kabale University
issn 2624-8921
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Vehicles
spelling doaj-art-ccceeee406524147bfcd5dc57bcb5a172025-08-20T03:43:58ZengMDPI AGVehicles2624-89212025-02-01712010.3390/vehicles7010020Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USARaja Daoud0Matthew Vechione1Okan Gurbuz2Prabha Sundaravadivel3Chi Tian4Department of Civil Engineering, The University of Texas, Tyler, TX 75799, USADepartment of Civil Engineering, The University of Texas, Tyler, TX 75799, USAResearch and Implementation Program, Texas A&M Transportation Institute, College Station, TX 75251, USADepartment of Electrical Engineering, The University of Texas, Tyler, TX 75799, USADepartment of Civil Engineering, The University of Texas, Tyler, TX 75799, USADriving at night is riskier in terms of crash involvement than it is during the day. Fortunately, it is clearly established that illumination on roadways can reduce the number and severity of nighttime crashes. However, state and municipal departments of transportation (DOTs) lack the available illumination data. Therefore, the objective of this research is threefold, as follows: (i) to develop machine learning models that use readily available roadway characteristic data to predict the severity of nighttime crashes; (ii) determine the effect that illumination has on crash severity; and (iii) develop a tool to assist DOT decision makers in collecting illumination data. To accomplish this objective, we have extracted data from the Texas Department of Transportation (TxDOT) Crash Record Information System (CRIS) database, which was then further split into a training and a test dataset. Then, seven machine learning techniques, namely binary logistic regression, k-nearest neighbors, naïve Bayes, random forest, artificial neural network, Extreme Gradient Boosting (XGBoost), and a Long Short-Term Memory (LSTM) model, were all applied to the unseen test data. The random forest model produced the most promising results by predicting severe crashes with 97.6% accuracy. In addition, we conducted a pilot study to test the collection of illumination data using a light meter. In the future, we aim to complete the development of a smartphone application, which can be used in conjunction with the random forest model presented in this paper, to collect crowdsourced illumination data and predict nighttime crash hotspots. This may assist DOT decision makers to prioritize funding for illumination at the hot spots.https://www.mdpi.com/2624-8921/7/1/20applied machine learningcrash severityilluminationsafetynighttime crashes
spellingShingle Raja Daoud
Matthew Vechione
Okan Gurbuz
Prabha Sundaravadivel
Chi Tian
Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA
Vehicles
applied machine learning
crash severity
illumination
safety
nighttime crashes
title Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA
title_full Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA
title_fullStr Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA
title_full_unstemmed Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA
title_short Comparison of Machine Learning Models to Predict Nighttime Crash Severity: A Case Study in Tyler, Texas, USA
title_sort comparison of machine learning models to predict nighttime crash severity a case study in tyler texas usa
topic applied machine learning
crash severity
illumination
safety
nighttime crashes
url https://www.mdpi.com/2624-8921/7/1/20
work_keys_str_mv AT rajadaoud comparisonofmachinelearningmodelstopredictnighttimecrashseverityacasestudyintylertexasusa
AT matthewvechione comparisonofmachinelearningmodelstopredictnighttimecrashseverityacasestudyintylertexasusa
AT okangurbuz comparisonofmachinelearningmodelstopredictnighttimecrashseverityacasestudyintylertexasusa
AT prabhasundaravadivel comparisonofmachinelearningmodelstopredictnighttimecrashseverityacasestudyintylertexasusa
AT chitian comparisonofmachinelearningmodelstopredictnighttimecrashseverityacasestudyintylertexasusa