Understanding how drivers respond to feedback and incentives is crucial for designing data-driven interventions that enhance road safety. This study investigates driver profiling and behavioral change using high-resolution telematics data collected through the OSeven DrivingStar smartphone application within the O7Insurance project. The naturalistic driving experiment was divided into two main phases: a baseline period with personalized feedback (Phase A) and an incentive-based phase (Phase B) comprising two gamified driving challenges with distinct reward criteria. Using data from 86 active participants, K-means clustering identified three driver profiles—Low-Exposure Cautious, Balanced/Average, and High-Risk Drivers—based on exposure, harsh events, speeding, and mobile phone use. The Balanced/Average group exhibited statistically significant improvements during both challenges, reducing speeding frequency and intensity (e.g., from 4.8% to 3.7%, p < 0.01), while High-Risk Drivers achieved moderate reductions in speeding intensity (from 6.4 to 5.3 km/h, p < 0.05). Low-Exposure Cautious Drivers maintained stable, low-risk performance throughout. These findings demonstrate that incentive-based telematics schemes can effectively influence driving behavior, particularly among drivers with moderate risk levels. The study contributes to the growing body of research on gamified driver feedback by linking behavioral clustering with responsiveness to incentives, providing a foundation for adaptive and personalized road safety interventions.
| ID | pj284 |
| DOI | |
| Tags |
