A vehicle’s expected fuel efficiency is the km/liter (or liters/100 km) that model should deliver under your fleet’s actual operating conditions. It is a number that looks simple, yet almost nobody has it properly set up. The consequence is direct: without a reliable expected fuel efficiency, you cannot tell whether a vehicle is consuming more than normal, you cannot detect fraud in fill-ups and you cannot compare drivers on a sound basis.
Setting up expected fuel efficiency by model (not by fleet, not by vehicle) takes less than a day. The return is measured over the following months: every deviation caught in time is money that does not leak away. Every driver identified as inefficient is a conversation that saves between 8% and 15% of consumption.
Expected fuel efficiency is the fuel consumption a vehicle model normally delivers when operating under your fleet’s typical conditions: type of route, load, climate, fuel quality, average driving style.
It has to be set by model, not by fleet, for two reasons:
Technical differences between models. A 4×4 diesel pickup performs differently from a gasoline sedan. A truck performs differently from a van. Averaging everything into a single number produces false detections at the extremes.
Differences within the same type. Two mid-size pickups from different brands can differ by 3 km/liter in the same operation. If you configure a single expected fuel efficiency for pickups, one of the two will always look bad.
Fuel efficiency is set in the Model Master: once per model, and it is inherited by every vehicle in that family.
The efficiency listed on the manufacturer’s spec sheet is measured under controlled conditions that do not reflect real operation. Real-world efficiency is almost always between 10% and 25% lower than the factory figure. Calculating real expected fuel efficiency requires your own data, not the figure from the manual.
The recommended process has 4 steps:
Step 1: Take 60 to 90 days of actual fill-ups. For each model in your fleet, export all fill-ups recorded over the last 2-3 months. You need at least 30-50 fill-ups per model for the average to be statistically useful.
Step 2: Calculate the actual efficiency of each fill-up. For each fill-up: kilometers driven since the previous fill-up divided by liters pumped. That gives you the actual efficiency (km/liter) of each event.
Step 3: Remove outliers. Drop the top 10% and the bottom 10%. They are usually odometer errors, partial fill-ups or extreme cases. What remains is the valid population for calculating the average.
Step 4: Calculate the average and define the acceptable range. The average is your expected fuel efficiency. Define an acceptable range of ±10% to ±15% as the normal zone. Below the lower limit indicates a problem (high consumption); above the upper limit is a sign of a data-entry error (false liters or wrong kilometers).
Repeat the process for each model. You will end up with between 5 and 15 expected fuel efficiency figures, depending on how diverse your fleet is.
Once expected fuel efficiency is set up, every new fill-up is automatically compared against that benchmark. The system calculates the fill-up’s actual efficiency and places it in one of three zones:
Red fill-ups are flagged as invalid fill-ups and enter the review workflow. Yellow fill-ups accumulate: if the same vehicle or driver lands in yellow several times in a row, that also triggers a review.
When a vehicle consistently delivers fewer km/liter than expected, it is almost always one of these four causes:
1. Undetected mechanical problem. Dirty air filter, faulty injectors, low tire pressure, dragging brakes. These are the most common causes, and they are caught quickly when the system cross-checks the fuel deviation against maintenance history.
2. Inefficient driving style. Aggressive acceleration, harsh braking, sustained speeding, prolonged idling. It is detected by cross-checking fuel with telematics (if available) or with a driver ranking within the same model.
3. Operation different from what was planned. The vehicle is being used on a more demanding route, with more load, or under different conditions (terrain, climate) than those assumed when expected fuel efficiency was set. The solution is to adjust the expected fuel efficiency or reassign the vehicle.
4. Incorrectly recorded fill-up. The driver pumped fewer liters than recorded, or the odometer reading is wrong. It is the most common fraud case and is detected with the Fuel module’s automatic validations.
Expected fuel efficiency is useless if nobody looks at the deviations. A minimal routine that works:
The routine takes less than 4 hours a month and saves between 8% and 15% of total fuel cost when followed with discipline.
To see the expected fuel efficiency calculation in concrete terms, look at how the 4-step process was applied to three representative models in a mid-sized LATAM fleet:
Model A — Compact gasoline sedan (urban operation, 18 units)
Fill-ups analyzed over 90 days: 412. After removing the top 10% (over 17 km/liter, usually partial fill-ups) and the bottom 10% (under 8 km/liter, usually odometer errors), 330 valid fill-ups remained. Average: 11.8 km/liter. Configured acceptable range: 10.2 to 13.4 km/liter (±15%). The manufacturer’s stated efficiency was 14.5 km/liter: real-world operation is 19% below that, consistent with intensive urban use.
Model B — Mid-size diesel utility van (urban delivery, 22 units)
Fill-ups analyzed over 90 days: 528. After cleanup, 422 valid fill-ups. Average: 9.4 km/liter. Acceptable range: 8.0 to 10.8 km/liter. Manufacturer’s figure: 11.2 km/liter. Real vs. manual deviation: -16%.
Model C — Light diesel truck (intercity delivery, 12 units)
Fill-ups analyzed over 90 days: 198. After cleanup, 158 valid fill-ups. Average: 6.2 km/liter. Acceptable range: 5.3 to 7.1 km/liter. Manufacturer’s figure: 7.5 km/liter. Real vs. manual deviation: -17%.
With the three efficiency figures set up in the Model Master, every new fill-up was automatically classified as green, yellow or red according to its distance from the corresponding model’s expected fuel efficiency. In the first three weeks, the system flagged around 12% of fill-ups as REVIEW. After the first calibration round (adjusting some ranges where the operation was more demanding), the percentage dropped to a stable 5-6%, which is the healthy range.
The operational difference between having expected fuel efficiency by model and having a single average for the whole fleet showed up in the first week: model C (truck) fill-ups would always have been flagged red if compared against the fleet average, generating useless noise and eroding the team’s trust in the system.
VEC Fleet implements expected fuel efficiency natively in the Fuel module:
Want to know how much fuel efficiency each model in your fleet should deliver?
With VEC Fleet you can set expected fuel efficiency by model and compare every fill-up against that benchmark, no spreadsheets involved.
