Predictive Analytics Based Driver Drowsiness Detection and Health Monitoring System

Extended driving hours without rest impose significant strain on drivers, leading to drowsiness and a substantial rise in fatal accidents each year. The existing Driver drowsiness detection systems primarily rely on real-time monitoring and alerts, which may be insufficient due to delayed driver reaction times, impaired decision-making, and potential loss of vehicle control, especially when the driver has deteriorating health conditions. To address these limitations, this paper presents a predictive analytics-based drowsiness prediction system that utilizes real-time monitoring to provide warnings minutes before potential drowsiness or any signs of impending health conditions occur. The system continuously monitors eye blink patterns and heart rate, which enables early detection of fatigue. It provides personalized interventions and scheduled break recommendations for improved alertness and a healthy lifestyle. Our evaluation demonstrates a significant improvement with 100% accuracy in drowsiness prediction compared to traditional methods, thereby enhancing driver safety and mitigating fatigue-related accidents.