How do you investigate fuel loss in a fleet?
Start with one vehicle and one review period. Record the fuel signal source, opening quantity, additions and closing quantity. Separate measured consumption from estimates, then compare the difference with refuelling receipts and the vehicle’s trips and stops.
Check whether a sudden change is explained by slope, movement, maintenance, calibration or a delayed reading. Preserve the original values and timestamps. Mark the result as explained, unresolved or escalated with supporting evidence; a fuel-drop alert is a reason to review the event, not proof of theft.
Establish the measurement before the alert
Confirm whether fuel data comes from the vehicle, an installed level sensor or manual records. Record units, sampling intervals, calibration and known limitations. A GPS tracker without a fuel signal cannot directly measure a tank-level drop.
Use an authorized installer for sensor and vehicle-interface work. Keep the calibration record and note any maintenance or tank change. Compare readings around known refuelling events so that the team understands what a normal update looks like before setting investigation thresholds.
Reconcile purchases, use and remaining fuel
Build a review record containing opening quantity, documented additions, measured or estimated consumption, and closing quantity. Keep measured values separate from estimates. Match fuel receipts and authorization records to the relevant vehicle and time period.
A difference in the reconciliation can come from incomplete records, measurement timing or the consumption estimate. Keep an uncertainty note instead of presenting a precise loss that the evidence cannot support. Use the same definition and interval when comparing vehicles.
Review the context of an unusual change
Check whether the reading coincides with a known refuel, slope, movement, sensor fault, disconnection or delayed transmission. Review the vehicle’s trip and stop record alongside it. Where the available data cannot distinguish between explanations, record the event as unresolved.
Escalate through the fleet’s existing investigation process and preserve original records. Speak with the responsible people without presenting an automated flag as a conclusion. Do not direct anyone to confront a suspected person or interfere with a vehicle during operation.
Close the operational gap
Look for process weaknesses that can be improved regardless of the investigation outcome: missing vehicle IDs on receipts, unclear refuelling permissions, inconsistent reconciliations or poor sensor maintenance. Assign an owner and due date to the chosen control.
Use a comparable review period to check whether unexplained differences recur. Document changes in vehicle use, calibration and record quality. Fewer alerts may reflect a changed threshold rather than fewer losses; record those configuration changes alongside the result.
Define the deployment you need
For SafarTrak, confirm fuel-signal availability and the supported review or alert workflow for your vehicles, devices and sensors. The objective is a traceable investigation with journey context, not a promise that every fuel loss will be detected.
Before rollout, ask for a sample record containing the signal source, timestamp, vehicle and related trip context. Agree on who reviews an exception and how it is resolved. Keep your manual fuel records where they provide information the connected setup does not capture.
Put the criteria side by side
Swipe or scroll sideways to view every column.
| Question | Evidence to record | Next check |
|---|---|---|
| Can the fuel reading be trusted? | Signal source, units, calibration and measurement time | Check sensor condition and compare with known refuelling events. |
| Do purchases and quantities reconcile? | Opening quantity, additions, closing reading and receipts | Separate measured consumption from estimates and note uncertainty. |
| Does the journey explain the change? | Vehicle, trip, stop and refuelling context | Check movement, slope, maintenance and delayed transmission. |
| What can the evidence support? | Original records, alternative explanations and unresolved gaps | Assign an owner and document the outcome before escalation. |
WORKED EXAMPLE
Reconciling fuel readings and records
Opening fuel of 120 L plus recorded additions of 80 L, minus estimated consumption of 150 L, gives an expected 50 L. A closing reading of 42 L leaves an 8 L difference to investigate. Because consumption is estimated and measurement has limitations, this arithmetic is not proof that 8 L was stolen.
Common questions
Can a GPS tracker show whether fuel was stolen?
GPS provides journey and stop context. Direct tank-level review needs a supported fuel signal and a validated installation. Neither a location record nor a sudden fuel drop alone proves theft.
Can fuel theft be prevented by software alone?
Software can help organize supported readings and context. Installation quality, fuel records, permissions and the investigation process remain necessary parts of the control.
What drop should trigger a theft alert?
Define an investigation threshold using the actual sensor behavior, sampling, tank and operating conditions. A universal litre threshold would be misleading.
Sources and further reading
- SafarTrak: device specifications and interfaces
Product reference for discussing signal and peripheral requirements; confirm the proposed fuel setup.
Written and reviewed by Yash Verma, Founder & CEO, SafarTrak. Sources checked on . How we write · Suggest a correction
Download the review sheet (CSV)
