The increasing availability of telematics and smartphone-based sensing technologies has transformed the study of driver behavior, offering new avenues for personalized feedback and incentive-based interventions. This paper presents an empirical investigation into driver profiling and behavioral change through an incentive-driven naturalistic driving experiment, implemented within the O7Insurance project of OSeven Telematics.
Data were collected from 116 participants over a multi-phase experiment using the OSeven smartphone application, which continuously monitored trip-level indicators such as speeding, harsh acceleration and braking events, and mobile phone use. The experiment included two stages: an initial feedback-only phase (Phase A), where drivers received personalized behavioral reports, and a subsequent gamified phase (Phase B), introducing performance-based challenges and monetary rewards for safe driving. All data were anonymized and processed in compliance with GDPR, ensuring user privacy.
The findings indicate that incentive-based schemes can effectively influence driving behavior, particularly among mid-risk drivers who have both awareness of feedback and room for improvement. In contrast, high-risk drivers may require more intensive, personalized interventions to achieve substantial behavioral change, while low-risk drivers benefit primarily from continued reinforcement. These insights reinforce the potential of clustering techniques not only for descriptive profiling but also for adaptive feedback design and risk-based policy formulation.
From a broader perspective, this research highlights the value of combining behavioral analytics, gamification, and telematics to enhance road safety (Kontaxi et al., 2025a; Kontaxi et al., 2025b; Singh & Kathuria, 2021). The integration of feedback and incentive mechanisms through smartphone platforms represents a scalable and cost-effective strategy for driver engagement. Future research should extend observation periods, incorporate contextual traffic data, and explore the persistence of behavioral improvements after incentives are removed. Additionally, coupling telematics data with psychometric assessments and machine learning models could further refine the understanding of driver responsiveness to feedback and rewards.
Overall, this study demonstrates that data-driven profiling and incentive-based telematics applications can contribute meaningfully to road safety management, usage-based insurance models, and personalized driver support systems, aligning technological innovation with behavioral science for safer mobility.
| ID | pc596 |
| Presentation | |
| Full Text | |
| Tags |
