How to Prevent Fuel Theft in Your Fleet

Fuel is the largest variable cost in road transport, and the only one that passes through other people's hands every day. That is why losses rarely happen in large amounts that get noticed right away, but in small quantities that repeat for months. Preventing such losses starts with making them visible. That is exactly why they are hard to spot by looking at receipts.

The ways fuel disappears

In practice there are four patterns, and each calls for a different way of detecting it.

1. Siphoning from the tank

The classic case: the vehicle is parked, the fuel level suddenly drops, and mileage does not change. It happens at night, in parking lots, during breaks. This is the only pattern that shows up as an instant change, so it is the easiest to detect, provided the fuel level is measured at all.

2. Fuel that ends up in another vehicle

Fuel is paid for with the company card, the receipt exists and is in order, but it was pumped into a private vehicle or one that is not part of the business. The receipt does not help here because everything is formally correct. It is detected by comparing the time and place of payment with the vehicle's location at that moment.

3. Billing for fuel that was never pumped

An arrangement with the gas station so the receipt shows more liters than were actually pumped, with the difference split between them. It is detected by comparing the quantity on the receipt with the actual rise in the tank level: the receipt says 80 liters, the probe records a rise of 60.

4. Overfilling and reselling

More is pumped than is needed, and the surplus is later siphoned off and sold. It is recognized by the pattern: unusually frequent or unusually large refuels relative to the distance driven.

Why a pump receipt is not proof

A receipt confirms that fuel was paid for. It does not confirm that it was pumped into that vehicle, or that it stayed there. The same point applies to fuel cards: they give an excellent record of the expense, but no record at all of what happened after payment.

The factory fuel gauge does not solve the problem either. Its resolution is coarse, the display is non-linear and meant for the driver, not for measurement. A drop of a few liters does not show up on it.

The conclusion is simple: until the fuel level is measured inside the tank itself, or consumption is measured at the engine, there is no data to compare the receipt against.

Three measurement methods and when each is used

In practice these are most often combined. Fuel level sensor detects sudden drops in level and most directly answers the question of whether fuel was siphoned. Fuel flow meter gives a more accurate picture of actual consumption and is especially useful for heavy machinery that burns fuel while stationary, where mileage tells you nothing. CAN bus telemetry is the fastest to roll out because it requires no installation inside the tank, but it is often unavailable on older vehicles.

Smart cap and camera: when a probe is not enough

A probe sees that the level dropped. It does not see who opened the tank. For vehicles that spend nights away from base, or where the drop keeps recurring, the next step is a smart fuel cap, which logs every opening and closing, with the time. A drop with no recorded opening points to a leak or a measurement error. A drop with an opening at three in the morning in a parking lot points to something else.

When the area around the tank is also covered by a camera, three sources are checked against each other: the fuel level from the probe, the cap opening and the footage from that moment. That is the highest level of control used in practice, and it makes sense on vehicles where the problem keeps recurring rather than across the whole fleet.

What to look for in a report

Data on its own does not solve anything. A report is useful if it answers specific questions:

• Every refuel, with quantity, time and location, comparable against the receipt.
• Every sudden drop in level, with time, location and whether the vehicle was moving.
• Consumption per 100 km by vehicle and by driver, so deviations show up in comparison.
• Engine idling time, separate from driving.
• An automatic alert the moment the level drops, not at the end of the month.

That last item matters more than it seems. If a drop in level only shows up in the monthly report, reconstructing an event from three weeks ago rarely leads anywhere. We described how alerts are set up on the Alarms and notifications page, and the structure of reports on the Reports page.

What to do when the system reports a drop in level

The first rule is that one alert is not proof. The fuel level can change suddenly for entirely innocent reasons too: the vehicle was parked on a slope, fuel redistributed between connected tanks, the temperature changed suddenly. So:

1. Check whether the vehicle was stationary at that moment and on what kind of terrain.
2. Compare the drop with that same vehicle's history - does the pattern repeat.
3. Compare it with other vehicles on the same route and under the same conditions.
4. Only once the pattern repeats several times, have a conversation with the driver.

A system used to make accusations based on a single reading quickly loses the company's trust. A system used to talk about patterns changes behavior without a single sanction, which is usually the actual goal.

How much this costs

The cost consists of the device and a monthly subscription, plus sensor installation where needed. Going by the math at the start of this article, where the loss is estimated at around 705,000 dinars a year on twenty vehicles, the system usually pays for itself within the first few months. Current prices and what is included are on the Price of GPS vehicle tracking page.

It is also worth noting that fuel prices in Serbia are not fixed: the highest retail prices change once a week, usually on Fridays. If you are doing this math for your own company, use an average over several months, not a single day's price.

Frequently asked questions