The design and implementation of a real-time Fleet Management and Vehicle Tracking System including hardware and software components for improved driver and vehicle monitoring is presented in this work. At the vehicle end, a Raspberry Pi fitted with an accelerometer, gyroscope, and GPS collects motion and location data and forwards it via a REST API to a web application. Built using Flask and SQLite with SQLAlchemy, the online platform provides modules including a dashboard, realtime car tracking via Leaflet.js, trip histories with route maps, and driver and vehicle performance reports. Harsh driving events such as acceleration, braking and cornering are detected by sensor thresholds and recorded in a SQLite database, enabling both live monitoring and retrospective analysis. A safety score that balances distance travelled against event frequency is computed for each vehicle and driver. These metrics are input to an SVM risk-prediction model, which groups drivers and vehicles into risk levels to guide preventive maintenance and safety actions. The platform is designed to scale with fleet size and to provide live updates during operation.