Fuel theft is detected by fitting a capacitive probe inside the tank that reports the level continuously. Normal consumption produces a gentle downward slope while the engine runs; theft produces a sudden vertical drop while the vehicle is stationary with the ignition off. The system flags that pattern with the litres lost, the timestamp and the exact location.
Why fuel theft stays invisible for years
Most fleets in Pakistan measure fuel the same way: pump receipts in, kilometres out, a rough average per vehicle. That method cannot see theft, because a driver who siphons 40 litres and drives more economically for a week produces exactly the same monthly average as a driver who does neither.
Averages hide events. To see theft you need the shape of the tank level over time, not a number at the end of the month.
What the tank curve actually shows
A capacitive probe cut to the depth of the tank reports the level every time the device transmits. Plotted over 24 hours, three patterns become obvious and distinguishable:
- Normal burn — a gentle, steady decline while the engine is running and the vehicle is moving. The slope varies with load and terrain but never falls off a cliff.
- A refill — a sharp rise, at a fuel station, usually with the engine off for a few minutes. This is the line you compare against the pump receipt.
- Siphoning — a sharp fall over a few minutes, with the ignition off, usually at night, and rarely at a fuel station. There is no legitimate explanation for fuel leaving a tank while the engine is not consuming it.
What a useful alert contains
An alert that says "fuel drop detected" is close to worthless. To act on it — and to have a conversation that holds up — you need four things attached:
- Volume. How many litres left the tank. "38 litres" is actionable; "a drop" is not.
- Timestamp. To the minute, in local time, with the ignition state at that moment.
- Location. A map pin, not a coordinate pair. Roadside on the N-5 near Okara reads very differently from a fuel station forecourt.
- Driver. Which named person was assigned to that vehicle, ideally confirmed by an RFID card rather than a roster.
Look for patterns, not incidents
A single event can have an innocent explanation — a leaking seal, a probe still settling after installation, a genuine maintenance drain. Patterns cannot.
What operators typically find in the first month is a repeating shape: the same stretch of road, the same hour, the same day of the week, sometimes across several vehicles at once. That regularity is the finding. It also tells you whether you are dealing with one driver or an arrangement.
What to check in your first fortnight of data
- Overnight drops on vehicles parked outside a depot
- Refills logged at the tank that are smaller than the receipt
- Unscheduled stops of 15 to 40 minutes on long-haul routes
- Idling hours, which burn fuel with nothing to show for it
Handling it without wrecking your yard
This part matters more than the technology. Fuel monitoring goes wrong when it is introduced as surveillance and lands as an accusation.
- Announce it. Tell drivers the tanks are monitored before the probes go in. A significant share of theft simply stops at that point, which is the cheapest possible outcome.
- Give the first month as a baseline. Calibration settles, and people adjust. Treat early data as learning, not evidence.
- Ask before you conclude. Show the driver the curve and ask what happened. Sometimes there is an answer — a mechanic drained the tank, a seal failed.
- Reward the clean ones. Fleets that pair monitoring with a fuel-efficiency bonus see better results than those that only punish. You want drivers who care about the number, not drivers who resent it.
Fuel is only the largest leak, not the only one
Once the tank curve is honest, the next savings are usually idling and route deviation — both visible on the same platform, both easier to fix than theft because nobody is hiding them. Our fuel monitoring feature page shows what the tooling looks like in practice, and the logistics solution page covers the route side.
A correctly calibrated capacitive probe resolves changes of a few litres, which is far finer than a factory float sender. Calibration is done on site by filling the tank in measured increments so the readings map to real volume rather than a generic curve.
Tampering leaves traces. Disconnecting the probe, cutting power to the device or opening its case each raise their own alert with the last known position, so an attempt to defeat the system becomes an event you are notified about rather than silence you never notice.
Fleets that combine fuel monitoring with idling control and route adherence commonly report reductions of around 15 to 20 percent in the first year. The largest single component is usually stopping siphoning, followed by reduced idling.
Not necessarily. Fuel probes pay back fastest on diesel vehicles with large tanks running long routes. Pool cars and short-radius vans often do not justify the sensor cost, and position plus idling data is enough for them.
Frequently asked
Short, direct answers to what people ask us about this topic.
Ask us something else