At a glance
- 21.8% of kilometres driven by heavy goods vehicles in Europe are driven empty — in France, that figure reaches 17.5%.
- Two mathematical problems underlie every loading decision: the knapsack problem (what to load?) and bin packing (how many trucks?).
- The average fill rate for trucks over 20 tonnes is only 69% — a third of capacity is wasted.
- Solutions exist, but US-based tools raise questions about logistics data sovereignty.
- Octave Engine solves both problems in seconds, with data hosted in France and Europe.
Truck loading: a daily puzzle
Every morning, thousands of logistics managers face the same question: how do we fill today’s trucks?
The answer seems straightforward. In practice, it’s a puzzle. Goods come in different sizes, weights and values. A box of envelopes takes up far less space than a refrigerator. A pallet of chilled products doesn’t carry the same priority as a batch of office supplies. And every vehicle has hard limits — on weight and on volume.
The result? Trucks leaving half-empty. Too many vehicles put into service. Transport costs spiralling. And a dock manager running calculations in Excel and hoping they haven’t made a mistake.
What most carriers don’t realise is that this daily puzzle has a name — or rather, two.
Two problems, billions of euros at stake
Behind every loading decision lie two optimisation problems that mathematicians have been studying for decades. They have technical names, but their on-the-ground reality is very concrete.
The knapsack problem — what to load first?
Picture a 20-tonne truck. You have 25 tonnes of goods to ship, each with a different commercial value. Not everything fits. Which items should you load to maximise the value carried?
That’s the knapsack problem. Just like a hiker deciding what to pack with a limited weight allowance, the carrier must make trade-offs between shipments. A poor choice means leaving revenue by the roadside.
This problem arises in any situation where demand exceeds capacity: selecting orders for a sea container, loading air freight, picking items for e-commerce fulfilment. Even Amazon uses knapsack-inspired approaches to optimise its warehouses.
Bin packing — how many trucks do you need?
The other side of the problem: all goods must go out. The question is no longer what to load, but how many vehicles to deploy to move everything.
That’s bin packing — literally filling bins. You have 50 parcels of very different sizes: envelopes, shoeboxes, flat cartons, a washing machine, a fridge. How do you distribute them across the fewest possible trucks?
The figures reveal the scale of the problem:
- In Europe, 21.8% of kilometres driven by heavy goods vehicles are driven empty (Eurostat).
- The average fill rate is only 69% for vehicles over 20 tonnes.
- In France, 17.5% of journeys are made empty.
- The industry uses on average only 65% of available capacity.
- Each unnecessary truck represents 5 to 10% of net margin lost (Trucknet).
One extra truck per round trip, multiplied by 250 working days, adds up to an enormous outlay in fuel, drivers and vehicle wear.
Existing solutions — and their limitations
Three approaches currently dominate the market.
Manual planning remains the norm for the majority of small and mid-sized carriers. The dock manager relies on experience, a spreadsheet and personal knowledge of the goods. It’s slow, error-prone and doesn’t scale as volumes grow.
Specialised software solutions exist — 3DPACK.ING, Optioryx, DeepPack — and they perform well. But they raise a question that rarely gets asked: where does your data go? Most of these tools are hosted in the United States. Yet your loading manifests contain sensitive information: volumes transported, cargo values, customer names. Transferring that data outside the European Union means losing control over strategically sensitive commercial information.
General-purpose ERPs (SAP, Oracle, Dynamics 365) offer logistics modules, but the implementation complexity is disproportionate for a loading optimisation need. The integration cost, learning curve and dependency on a heavy ecosystem put off most carriers.
Octave Engine — the simple, sovereign alternative
Simple and intuitive
You don’t need to be a data scientist to optimise a load. With Octave Engine, the process comes down to three steps:
- Import your data — a CSV or Excel file with your parcels, weights and volumes.
- Run the calculation — the algorithm solves the knapsack and bin packing problems simultaneously.
- Get your loading plan — in seconds, even for hundreds of items.
No three-day training course. No six-month IT integration project. A tool built to be used from the very first minute.
Sovereign and confidential data
That’s the fundamental difference. Octave Engine is hosted in France and Europe, on a sovereign infrastructure. Your data never leaves the European Union.
Why does that matter? Because your logistics data is sensitive data. The volumes you carry, the value of your goods, the identity of your customers — that is strategically sensitive commercial information. With Octave Engine, it stays confidential and GDPR-compliant by design.
Fast and effective
Octave Engine’s algorithms are specifically built for logistics knapsack and bin packing. Computation time is measured in seconds, not minutes. The result: fewer trucks on the road, less fuel consumed, less CO₂ emitted.
Real-world use cases
Load optimisation applies across every sector of the transport industry:
- Retail and grocery: maximise fill rates on multi-stop deliveries to supermarkets and distribution centres, factoring in both weight and volume constraints.
- E-commerce: move from 65% utilisation to over 85% with an algorithm tailored to parcels of wildly varying sizes.
- Construction and waste collection: minimise skip and truck movements on job sites, where every extra journey costs money in time and fuel.
- Temperature-controlled transport: manage weight, volume and product compatibility constraints simultaneously — a multi-dimensional knapsack problem.
Every truck counts
A truck leaving half-empty is money burned. An unnecessary truck on the road is wasted fuel and CO₂ emitted for nothing. And a dock manager spending two hours in Excel is time that could go to higher-value work.
Bin packing and the knapsack problem are not mathematical curiosities. They are the two concrete problems that every carrier solves — well or poorly — every day. Solving them effectively means cutting costs, optimising your fleet and reducing your carbon footprint.
Octave Engine puts these algorithms at your fingertips, with the guarantee that your data stays in France and Europe.