Driving style identification and its association with risky driving behaviors among truck drivers based on GPS, load condition, and in-vehicle monitoring data

Abstract

This study provides an approach to identify driving style of truck drivers by using GPS, load condition, and in-vehicle monitoring data and investigates the association of driving styles with risky driving behaviors from macro and micro perspectives. The naturalistic driving data used in this study were collected from 4,357 trucks in Hangzhou, China over three months in 2021. Six driving volatility parameters and six warning parameters were used to characterize the driving styles. Then, three driving styles under the two load conditions were identified using k-means clustering methods and principal component analysis. Finally, one-way MANOVA and ANOVA were used to analyze the relationship between driving styles and driving risk. It was found that truck drivers have different thresholds for aggressive and cautious driving style under different load conditions. Truck drivers who exhibited aggressive driving under both load conditions exhibited high driving risk. Although most truck drivers exhibited safe or normal driving under both conditions, the few who exhibited aggressive driving contribute to a disproportionate driving risk. These results can help distinguish differences in truck drivers’ driving styles under different load conditions, thus providing a more comprehensive safety assessment of truck drivers’ performance for monitoring purposes.

Keywords:

• truck drivers
• driving style
• load conditions
• naturalistic driving data
• in-vehicle monitoring system
• risky driving behavior

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The research is sponsored by the National Natural Science Foundation of China (No. 52172342) and supported by the Open Project of Key Laboratory of the Ministry of Public Security for Road Traffic Safety of China (Grant No. 2021ZDSYSKFKT02), and also sponsored by Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX22_0280).