Associations between drivers' safety records and driving styles: A naturalistic study. (September 2013)

• Record Type: Journal Article
• Authors: Merrikhpour, Maryam; Donmez, Birsen; Wang, Chongbo; Hayes, Benjamin; Grush, Bern
• Is Part Of: Proceedings of the Human Factors and Ergonomics Society ... Annual Meeting. Volume 57:Part 1(2013)
• Page Start: 1973
• Page End: 1977
• Publication Date: 2013-09
• Subjects: Human engineering -- Congresses; 620.8205
• DOI: 10.1177/1541931213571440
• Languages: English
• ISSNs: 1541-9312

Abstract

Identifying potentially at-risk drivers based on their driving styles may help target these individuals with effective countermeasures, such as individual counselling, triggering alarms for degraded driving style, and assessing risk for insurance. The primary objective of this research was to investigate the relation between riskiness (as measured through self-reported at-fault crash and moving violation records in the previous five years) and driving style, as measured through speed and acceleration profiles. A field trial was conducted with 40 participants in two age groups: younger (ages 25 to 35) and older (ages 45 to 65). Through an initial questionnaire, 19 drivers with worse safety records (defined as one or more at-fault traffic crashes in the last five years and/or two or more at fault crashes since licensure and/or two or more speeding tickets in the last five years) and 21 drivers with better safety records (defined as no at-fault crashes in the last five years and no more than one at-fault crash since licensure and no more than one speeding ticket in the last five years) were selected for participation. Within each group, age was approximately counterbalanced. Naturalistic data including vehicle position, speed, and acceleration were collected through a GPS-enabled telematics platform, which was installed in participants' vehicles. Data were collected from each participant over a one month period. No statistically significant relation was identified between safety record and driving style. However, three groups of drivers emerged in a cluster analysis, with one group exhibiting speeding and abrupt deceleration behaviours significantly less than the other two. These other two groups, which did exhibit riskier behaviours, differed regarding the extent of which behaviour they exhibited more. One group was more likely to exhibit speeding and the other more likely to exhibit abrupt acceleration and deceleration behaviours.