Predicting unsafe driving risk among commercial truck drivers using machine learning: Lessons learned from the surveillance of 20 million driving miles. (September 2021)

• Record Type: Journal Article
• Authors: Mehdizadeh, Amir; Alamdar Yazdi, Mohammad Ali; Cai, Miao; Hu, Qiong; Vinel, Alexander; Rigdon, Steven E.; Davis, Karen; Megahed, Fadel M.
• Is Part Of: Accident analysis and prevention. Volume 159(2022)
• Publication Date: 2021-09
• Subjects: Artificial intelligence -- Big data analytics -- Crash risk prediction -- Naturalistic driving -- Near crashes -- Safety critical events; Accidents -- Prevention -- Periodicals; Accident Prevention -- Periodicals; Accidents -- Prévention -- Périodiques; 363.106
• DOI: 10.1016/j.aap.2021.106285
• Languages: English
• ISSNs: 0001-4575

Abstract

Highlights: The occurrence of SCEs can be predicted in 30-minute trips/intervals. Predictors capturing kinematic and traffic conditions were the most predictive. Developed machine learning models are generalizable to new drivers. Developed models can be deployed without retraining for a 2+ months-period. The developed code is freely available for practitioners and researchers. Abstract: The emergence of sensor-based Internet of Things (IoT) monitoring technologies have paved the way for conducting large-scale naturalistic driving studies, where continuous kinematic driver-based data are generated, capturing crash/near-crash safety critical events (SCEs) and their precursors. However, it is unknown whether the SCEs risk can be predicted to inform driver decisions in the medium term (e.g., hours ahead) since the literature has focused on SCE predictions either for a given road segment or for automated breaking applications, i.e., immediately before the event. In this paper, we examine the SCE data generated from 20+ million miles-driven by 496 commercial truck drivers to address three main questions. First, whether SCEs can be predicted using disparate driving-related data sources. Second, if so, what the relative importance of the different predictors examined is. Third, whether the prediction models can be generalized to new drivers and future time periods. We show that SCEs can be predicted 30 min in advance, using machine learning techniques and dependent variables capturing the driver's characteristics, weather conditions, and day/time categories, where an area under the curve (AUC) up to 76% can be achieved. Moreover, the predictive performance remains relatively stable when tested on new (i.e., not in the training set) drivers and a future two-month time period. Our results can inform dispatching and routing applications, and lead to the development of technological interventions to improve driver safety.