Loading...

Truck Fuel Consumption: When the Fleet Average Misleads

See why fleet fuel consumption can rise as individual journey types improve. Compare distance, fuel and the work mix with a clear numerical example.

A truck fleet's average fuel consumption can rise even when fuel use per kilometre falls within every type of journey. That can happen when the balance of work changes. If more of the distance comes from operations with higher consumption, the fleet-wide average can increase. Understanding the difference requires looking at fuel, distance, and how those kilometres are distributed, rather than relying on the final average alone.

Imagine comparing two months using GPS information and fuel records. The second month shows a worse figure, prompting questions about drivers or vehicles. Before assigning a cause, check whether the fleet performed the same kind of work. A new urban operation can change the average even if both urban and highway journeys improved separately. The hypothetical example below makes that difference visible through calculations you can verify. It does not describe a customer result or a fuel-saving claim for Plaspy.

The fleet average contains the balance of work

A rate expressed in litres per 100 kilometres relates fuel to distance. For the fleet as a whole, divide the included fuel consumption by the corresponding distance and multiply by one hundred. That result describes the particular set of activity you included. Changing which journeys contribute, or how much distance each contributes, can change the average without every operation becoming less efficient.

Adding individual vehicle rates and dividing by the number of vehicles does not necessarily answer the fleet-wide question. A truck that travelled a short distance would receive the same weight as one that travelled much farther, even though they contributed very different amounts to the total. Distance matters when describing combined consumption. Total litres divided by total kilometres preserves that relationship, provided both refer to the same scope and period.

“How much fuel did we use?” and “How much did we use per 100 kilometres?” are also different questions. The first describes a volume; the second relates that volume to distance. A fleet that travels farther may consume more litres without increasing its rate. A fleet that keeps its total distance unchanged may show another rate if its work is distributed differently. The appropriate figure depends on what you are trying to explain.

Plaspy Reports lets you examine information by dates, devices, and groups, and download the result in Excel. That selection helps define the activity being compared. The analysis of work distribution developed here is performed on data whose scope and origin have been checked. It is not presented as an automatic adjusted-comparison calculation built into the platform.

The same total distance can produce different stories

Suppose a fleet performs urban and highway journeys. These are simplified categories for the example, rather than a universal classification of transport work. In the first period, A, the fleet covers 4000 urban kilometres and 6000 highway kilometres. The corresponding fuel consumption is 1600 and 1800 litres. Those figures produce rates of 40 and 30 litres per 100 kilometres respectively. Neither rate is offered as a normal value or a target for your trucks.

In period B, the fleet undertakes more urban work: 8000 urban kilometres and 2000 highway kilometres. It consumes 3040 litres for the former and 580 for the latter. The respective rates are now 38 and 29 litres per 100 kilometres. Both are lower than before. To follow the comparison, assume that each category's fuel and distance are complete, comparable, and correctly attributed. All figures remain illustrative.

Hypothetical example data
PeriodJourneyDistance (km)Fuel (L)
AUrban40001600
AHighway60001800
BUrban80003040
BHighway2000580

Period A totals 3400 litres over 10000 kilometres, or 34 litres per 100 kilometres. Period B totals 3620 litres over the same 10000 kilometres, or 36.2 litres per 100 kilometres. The fleet average has increased by 2.2 litres per 100 kilometres, even though the urban rate fell from 40 to 38 and the highway rate fell from 30 to 29.

There is no mathematical contradiction. Urban journeys represented 40% of the distance in A and 80% in B. The type of journey with the higher rate in this example gained weight in the total. Improvement within each category and an increase in the overall average can coexist because they describe different comparisons. The aggregate alone cannot show which of those changes produced the final number.

Nor should the increase in litres be dismissed as a reporting error. Period B used 220 more litres in our example, and that consumption belongs to the operation described. What changes is the explanation. The fleet average by itself does not establish that every type of journey deteriorated, and it cannot attribute the result to a particular person's behaviour.

Holding the work mix constant answers another question

We can ask what rate would result if B's rates were applied to the same distance distribution as A. This is a hypothetical comparison with fixed weights. Retain 40% urban and 60% highway kilometres, but use rates of 38 and 29 litres per 100 kilometres. The calculation is 0.40 × 38 + 0.60 × 29, which gives 32.6 litres per 100 kilometres.

With that distribution held constant, A's reference rate was 34 and B's would be 32.6. The difference is 1.4 litres per 100 kilometres. This comparison shows lower category rates within a common distribution of distance. It does not change B's actual combined consumption, which remains 36.2 under the work mix described in the example. The hypothetical comparison and the actual total retain different purposes.

The two figures support different conversations. The actual average describes fuel used for the activity performed. The fixed-mix average compares rates without allowing the change in work distribution to dominate that question. Showing only the second could conceal a real fuel requirement. Showing only the first could conceal improvement within the categories. Neither replaces the other simply because it produces a more reassuring result.

There is another useful check. Giving 38 and 29 equal weight produces 33.5. That is neither B's actual distribution nor A's reference distribution. It describes a scenario in which half the kilometres belong to each category. The arithmetic is valid; the problem would be presenting that different scenario without identifying it. Every average carries a distribution, even when the distribution has not been stated explicitly.

The chosen weights should therefore be visible. We used A's distribution because it is the starting reference for this comparison. That need not be the reference for every purpose, but silently changing weights makes apparently comparable results rest on different bases. Across repeated reviews, a declared reference helps explain what changed and prevents choosing whichever combination produces the most convenient number afterward.

This adjustment does not explain why the category rates fell. Vehicles, loads, traffic, and other conditions may have changed within either category. Holding the urban and highway mix constant addresses only that difference in composition. It does not turn two months of activity into a controlled test of GPS, driver training, or a maintenance intervention.

Categories should represent the work you want to compare

“Urban” and “highway” explain the example but may be too broad for your operation. Two urban journeys can involve different loads, gradients, and frequencies of stopping. Treating them as equivalent could allow a changing mix within the category to hide another difference. Classification needs enough detail to answer the question without creating so many groups that each contains very little activity.

The relevance of the duty cycle also appears in Southwest Research Institute's heavy-duty fuel economy evaluations , which distinguish procedures and cycles suited to different operations. That reference supports considering the work performed. It does not mean our illustrative comparison follows a certified test, or that Plaspy has measured the results shown here.

A useful category has a definition another person can apply without knowing the eventual fuel consumption. It might correspond to a service type identified in company records. If journeys are separated afterward according to which consumed more fuel, the classification already contains the result being explained. It becomes difficult to tell whether you discovered an operational distinction or merely rearranged the figures.

Inclusion rules matter as well. A journey with insufficient data does not become an efficient operation because it disappears from the calculation. If a category loses records between periods, the comparison may represent a different portion of the activity. Describe what was included and what is missing. Avoid presenting a partial result as though it covers every kilometre travelled by the fleet.

There is a specific limit when a type of work disappears. Imagine a variation in which B includes no highway kilometres at all. There would be no observed highway consumption rate for that period. Entering zero and applying the reference's retained 60% highway weight would invent a rate that was never observed. You could instead compare only categories with suitable data in both periods, but that requires redefining and declaring the common scope. If only urban work remains, the result describes urban work. It is no longer an adjusted comparison covering the whole original fleet, however neatly the remaining figures fit the calculation.

A comparison can remain useful even when the operations are not perfectly matched. The point is to recognize which differences are known and what conclusions they permit. A broad category may indicate where a closer examination would help. It need not immediately become a driver or truck rating. It can instead provide the starting point for a more clearly defined enquiry.

An estimated rate cannot prove its own improvement

Before interpreting litres, establish how they were obtained. The Plaspy Sensors guide explains that when a device has no fuel sensor, an approximate consumption value can be used to estimate fuel from distance. Such a calculation can support a projection, but its result remains an estimate. Displaying it alongside the journey does not change the source of the number.

If you change the configured rate and the estimated calculation then falls, you have not demonstrated a measured reduction in the vehicle. You have changed an input. Investigating an actual improvement requires a consumption source appropriate to the question, with its limitations understood. The platform does not turn a configured assumption into an independent measurement simply by presenting the calculated result in a report.

The same issue arises when part of the fleet uses estimates and another part supplies data of a different nature. Adding them without identifying their origins can produce a seemingly precise total with mixed meanings. Before adjusting for types of work, check that the quantities are comparable. Keep separate anything that cannot yet be treated as equivalent, rather than allowing a shared unit label to imply a shared measurement method.

Downloading a report makes it easier to examine information outside the platform, but exporting does not correct its origin or complete missing records. A spreadsheet can apply a formula correctly to inputs that do not answer the same question. Checking the data therefore precedes comparing averages. Accurate arithmetic cannot replace a clear definition of what is being counted and which activity those figures describe.

The cover shows a truck, GPS and LTE references, and an illustrated connection. It represents information use in transport, rather than proof of continuous coverage, universal compatibility, or fuel measurement in every installation. Real data depends on the tracker, its capabilities, the installation, configuration, and the information the equipment actually supplies. The illustration does not establish those conditions for a particular truck.

Use the result to guide an operational decision

Return to the two periods. If the company took on more urban journeys, the conversation can begin with that change in work. The higher fleet average is insufficient grounds to ask every driver to explain a presumed deterioration. First recognize that the distribution changed and that, in this example, both category rates fell. That reading supports questions consistent with what the information actually shows.

Work composition also matters when planning a future operation. Applying the old average of 34 to a different mix could provide an unsuitable reference. A projection can use relevant categories and explicit assumptions, while remaining a projection. It does not guarantee what an individual truck will consume, nor does it replace later observation of the work actually performed. Its value is in making the expected mix understandable before treating its total as certain.

If one category rises while another falls, the combined average can conceal that difference too. A general conversation about the whole fleet would leave the location of the change unexplored. Separating categories directs the enquiry toward an identifiable part of the work, without yet deciding whether the cause was load, route, vehicle, driving, or data quality. The categories narrow the question; they do not supply its answer automatically.

To review what information you can obtain with Plaspy , describe the comparison you want to make: vehicles, periods, types of work, and the source of the fuel figures. That helps establish which data is available and what analysis needs your own records. The commercial question becomes whether you can build a comparison you understand, rather than whether GPS promises a particular saving percentage.

An average summarizes activity and inevitably loses some of the detail behind it. Restoring the distribution of kilometres shows when a number changes with the work performed and when the category rates deserve investigation. With both views, you can explain fleet fuel consumption without asking the final figure to provide a conclusion it cannot support on its own.

Home

Ready to get started with Plaspy?

Buy your GPS tracker or subscribe to our satellite tracking platform.
We are currently offline, but don’t worry.
Leave us your contact details and we’ll get back to you as soon as possible via phone or WhatsApp to +57 (601) 580-3299.


How to check when a truck left the yard after ignition

Separate ignition, movement inside the yard and the truck’s departure. Give dispatch a time it can confirm before updating the receiver.

How long did a truck stop when its GPS reports have gaps?

Separate observed timestamps from the start and end of a truck stop. Use the Plaspy timeline to work out what the available reports can establish.

Truck movement alerts: how to confirm an authorized yard move

Match the unit, yard instruction and Plaspy event to verify an authorized truck move with a Ruptela Eco5 Lite and confirm where the vehicle ended up.

How to check whether a truck used a detour in GPS history

Look at the reports before, within and after a detour to see what Plaspy can establish about the route and where another trip record may help.

How to hand over a shift while a truck is still on the road

Build a dispatch handover around the truck’s last reported location, its timestamp and the action the next shift needs to take before updating a customer.

Boxtrack 4G: Understanding Its Installation Demonstration

Understand the Boxtrack 4G demonstration: its tester, relay and different panic inputs. Prepare a clear discussion about your installation and its checks.

Billing GPS Customers Across Different Platforms

Bring together billing for Plaspy devices, external trackers and phone lines. Understand each item and which actions remain with other providers.

GPS Payments: What to Expect from a Reminder

Start organizing GPS customer reminders with clear expectations. Distinguish sending notices, receiving payments and improving payment timeliness.

Installing a GS1000: What a Handover Check Establishes

A map position does not verify an entire GS1000 installation. Understand each observed response and what to agree when your motorcycle is handed back.

GS1000 Cost: A Monthly Average Is Not a Monthly Payment

Budget three years of GS1000 use without counting the included first year twice. Separate monthly averages, actual payments, renewal and installation.