Fleet fuel telemetry

The client suspected losses but had no data to back it up. We installed capacitive level sensors in tanks cross-referenced with GPS, transmitting every 30 seconds via LTE. The backend correlates consumption with mileage and route, detecting anomalies such as apparent refueling without movement or level drops off-route. The React dashboard filters by vehicle, driver, route and time window, and exports evidence for driver meetings or disciplinary processes. Integration with the existing fueling system closed the loop: every recorded refuel is verified against the actual tank rise. In a later phase we added an isolation forest model over historical fueling and consumption patterns: the model learns each vehicle's normal behavior (expected consumption given route, time and load) and flags deviations for review. This let us scale fraud detection without relying on hard rules and cut false positives by around 60% versus the initial fixed thresholds. Inference runs on the backend hourly and alerts route into the same operations dashboard.

Key metrics

• ✓ −22% in total fuel consumption
• ✓ 6 critical leak points identified in month 1
• ✓ ROI < 6 months
• ✓ Automated executive reports every Monday