Driving Style Analyses for Car-sharing Users Utilizing Low-frequency Trajectory Data

Car-sharing is a representative of the sharing economy in the field of transportation, which provides drivers with more economical and flexible approaches for car ownership and usage. Currently the car-sharing users are mainly young and novice drivers with frequent risky driving behaviors and traffic crashes. The risky driving behaviors have led to a sharp increase in the cost of maintenance and insurance for car-sharing enterprises, and the crashes have also increased the operational risks of roadway traffic. Therefore, it is urgent to conduct analyses for driving style classifications in order to identify potential high-risk drivers. In this study, we conducted a driving style analysis based on the operation data of a car-sharing project located in Shanghai. Rather than the high resolution driving behavior data that adopted in the majority driving style studies, low-frequency trajectory data were utilized here. Relative speeding time ratios of freeway, urban expressway and urban road were used as analysis variables and K-Means clustering technique was used to classify the drivers’ driving styles. A total of three categories were concluded, which are aggressive, calm and novice with the percentages of 28.33%, 66.67%, and 5.00% correspondingly. Then, for the purpose of understanding the different driving styles, comparison analyses were further conducted from the aspects of vehicle operation features and personal information. The results suggested that drivers have substantial differences in their vehicle operation characteristics. The aggressive drivers tend to drive faster, have higher speeding tendency, better speed stability and higher skills; while the novice drivers reported highest speeding tendency on freeway, lowest operating speed and speed stability. Besides, no statistically significant differences in age, gender or violation between driving styles were identified.