Enhancing Road Safety Education through AI-Driven Tools: The Impact of Driving Simulations for Accident Prevention

Abstract

Road safety is a major global problem; notwithstanding the developments in safety technologies, driver behaviour and error remain integral contributors to accidents, underlining the demand for enhanced driver education. This study analyses the potential use of Artificial Intelligence (AI)-driven strategies to enhance road safety education by fostering philosophical understanding and practical driving skills. Using a quasi experimental methodology, this research involved 100 qualified oil and petrol tanker drivers who were assigned to two separate groups: the initial group underwent traditional instructor-led training. In contrast, the other group had AI-based instruction using AI-powered driving simulations. Driving experience, accident history, age, and driving hours showed no significant variations between the groups. The study's findings revealed that the AI-based group outperformed the traditional training group in reaction time (2.2 seconds vs. 3.1 seconds), accident frequency (1.0 vs. 3.4), road safety knowledge (22% vs. 12%), and driving errors (4 vs. 11 per simulation). Also, the AI group had 91% accuracy in hazard detection. The findings from the regression analysis revealed that fewer driving errors were strongly linked to a decline in accidents. Hazard detection accuracy and driving mistakes were connected to faster reaction times and increased knowledge. Based on the study's findings, further research is needed on the reliability, practical application, and long-term effects of AI in various driving scenarios.