Validation of a smartphone telematics algorithm for classifying driver trips

• Published in: Transportation research interdisciplinary perspectives 2024-05, Vol.25, p.101109, Article 101109
• Main Authors: Ebert, Jeffrey P., Xiong, Ruiying A., Patel, Arjun, Abdel-Rahman, Dina, McDonald, Catherine C., Kit Delgado, M.
• Format: Article
• Language: English
• Subjects: Accuracy; Driver classification algorithm; Driver risk score; Mobile telematics application; Naturalistic study; Usage-based insurance
• ISSN: 2590-1982; 2590-1982
• DOI: 10.1016/j.trip.2024.101109

•Auto insurers and researchers use mobile apps to measure risky driving behavior.•These apps depend on accurate driver/non-driver classifications.•Our field study found these classifications to be 97 % accurate.•This should give insurers, researchers, and users confidence in driver risk scores. This study assessed the accuracy of a smartphone telematics algorithm that classifies car trips as driver or non-driver. Participants’ trips were measured for 4 weeks by Way to Drive, a research telematics application that uses the same data algorithms as leading auto-insurance companies. At the end of each week, participants completed a survey prompting them to review trips within the app and report time and nature of any misclassified trips. Overall accuracy of driver vs. non-driver classification was high (96.5 %, SD = 5.1 %). Sensitivity, the percentage of actual driver trips classified as such, was also high (97.5 %, SD = 4.6 %). Specificity, the percentage of non-driver trips classified as such, was slightly lower and more variable (91.2 %, SD = 14.8 %). The algorithm’s accuracy was generally robust to a variety of phone characteristics, vehicle features, and driving habits.