Divide 76 m³ of container by 0.084 m³ of carton and you get 905. Ask a forwarder and the answer comes back around 700.
Both numbers are arithmetically correct. They are answering different questions — and the gap between them is roughly the value of one extra container per five, which is why it is worth ten minutes.
This page does the arithmetic explicitly. If you would rather have the number for your own carton, the 3D container loading planner computes it from the same constants.
This is the arithmetic companion to our cost-side piece on container loading, packaging and freight. That article follows the money; this one follows the cartons.
1. The inputs: four dimensions and one payload
| Container | Internal length × width × height | Max payload |
|---|---|---|
| 20GP | 589 × 235 × 239 cm | 28,200 kg |
| 40GP | 1203 × 235 × 239 cm | 28,200 kg |
| 40HQ | 1203 × 235 × 269 cm | 28,600 kg |
| 45HQ | 1355 × 235 × 269 cm | 28,600 kg |
Three things to notice before any division happens:
40GP and 40HQ are the same container except for 30 cm of height. 239 cm versus 269 cm changes internal volume by roughly 12%. On cube-heavy cargo — luggage cartons are cube-heavy — that 30 cm usually earns more than the freight difference costs.
The usable height is not the internal height. The door opening is lower than the interior ceiling, so the top tier has to pass through the door throat. Plan the last tier as if the box were shorter than the table says.
Payload, not volume, is the second ceiling. §4 shows when it binds first.
2. Why cube math overstates: the grid
Dividing volume by volume assumes cartons flow like liquid. They do not. They stack in whole numbers along three axes.
Worked example — illustrative carton 60 × 40 × 35 cm (0.084 m³) in a 40HQ:
Along the length 1203 ÷ 60 = 20.05 → 20
Across the width 235 ÷ 40 = 5.88 → 5
Up the height 269 ÷ 35 = 7.69 → 7
Grid: 20 × 5 × 7 = 700 cartons
Cube: 76.05 m³ ÷ 0.084 m³ = 905 cartons905 optimistic, 700 real. The grid fills 58.8 m³ of 76.05 m³ — about 77%.
Every fractional remainder above is capacity donated to air: 0.05 of a row, 0.88 of a column, 0.69 of a tier. Three small remainders compound into a 23% overstatement.
This single mechanic explains most of the difference between a spreadsheet plan and a forwarder's confirmation — and it explains why "just add a 10% buffer" is the wrong correction. The loss is structural, not random.
3. Five millimetres costs a full row
Cartons are nominal. Real cartons bulge, and a few millimetres per dimension is normal production variation, not a defect.
Same example, with the carton length actually measuring 60.5 cm:
Planned: 1203 ÷ 60.0 = 20.05 → 20 rows
Actual: 1203 ÷ 60.5 = 19.88 → 19 rowsOne row of twenty — 5% of the load, from five millimetres. On 700 cartons that is 35 cartons per container, repeated on every container in the programme.
Two consequences worth acting on:
- Measure the largest carton from the first production run, not the average and not the spec sheet. Tolerance is a range; the grid only cares about the maximum.
- Design the carton to the grid, not to a round number. If a 59.5 cm carton divides the container length cleanly and a 60 cm carton does not, the "rounder" number is the more expensive one.
4. Which ceiling arrives first: the density crossover
Payload is a hard limit. Which limit you hit first depends on how dense your packed carton is. The crossover is computable:
Density at which weight binds first = max payload ÷ internal volume
20GP 28,200 ÷ 33.08 m³ = 852 kg/m³
40GP 28,200 ÷ 67.57 m³ = 417 kg/m³
40HQ 28,600 ÷ 76.05 m³ = 376 kg/m³
45HQ 28,600 ÷ 85.65 m³ = 334 kg/m³Read it as a decision rule:
| Your packed carton density | Which ceiling binds | What to optimise |
|---|---|---|
| Below the number for that container | Volume — you run out of space first | carton geometry, nesting, grid fit |
| Above the number | Weight — you run out of kilograms first | packing density; a bigger box buys nothing |
Compute your own as carton gross weight ÷ carton outer volume. Luggage cartons frequently land in the volume-bound range on a 40HQ, which is exactly why the high-cube earns its premium — but that is a conclusion about your carton, and it has to be checked rather than assumed.
The planner applies both ceilings and tells you which one bound the result. A pure cube calculation applies neither.
5. Palletisation changes the plan, not the arithmetic
Everything above is a floor-loaded plan. Palletised goods change three inputs at once:
- Pallet height (typically 14–15 cm) comes out of usable height per stack
- Each pallet footprint loses its overhang margin
- Void space appears between pallets where cartons could have sat
Floor-loaded versus palletised is usually a two-digit percentage difference in cartons per container. The arithmetic is identical — you simply run the grid against a smaller effective volume. What is not optional is deciding which configuration you are planning, before freight is booked on the other one.
6. Mixed SKUs: volume fits, weight does not
A container with three SKUs breaks the single-grid model, because each carton type occupies its own grid pattern and they do not tessellate into one another.
Practical sequence for a mixed load:
- Run the grid once per carton type to get cartons-per-container for each
- Allocate by commercial priority, not arithmetically — the SKU that must ship first gets the space that fits cleanest
- Then total the weight and check it against payload, because volume can fit while weight does not
Step 3 is the one teams skip. A container can be visually full and still be legal — or look fine by volume and exceed payload once the densest SKU is loaded last.
7. What no planner can tell you
A model is only as good as its inputs. These sit outside any loading calculation, ours included:
| Outside the model | Why it decides the outcome |
|---|---|
| Actual carton dimensions after production | ±5 mm changes row counts (§3) |
| Carton strength under stacking | the bottom tier carries everything above it for weeks |
| Pallet configuration at origin | palletised and floor-loaded are different plans (§5) |
| Stack-height or orientation restrictions | some markets and carriers constrain both |
| Real stuffing practice at the warehouse | labour, dunnage, and whether the crew follows the plan |
Carton strength is the row people skip. A carton that survives a single drop is not automatically a carton that survives being the bottom tier at sea — those are different tests on different objects. The product-side tests are covered on the testing lab page; compression is the packaging-side counterpart.
Which is also why the loading plan is the last gate and not the first: the carton you are planning is the carton the product was tested inside. If production switched carton specification after testing, the plan is optimising a box that has never been validated — and the compliance documents naming the product no longer describe what is inside it.
8. Four steps, in this order
- Fix the carton — measure the largest one from the first run
- Run the grid for the candidate container at that size
- Apply the ceilings — weight crossover (§4) and, if relevant, palletisation (§5)
- Round to MOQ-compatible lots, then set the order quantity
Step 4 is where the plan and the PO have to agree. If the plan says 700 and the PO says 1,000, the difference is 300 cartons of warehouse rent plus the freight on a container that did not need to sail.
The planner shows the order-quantity arithmetic next to the fit result so both numbers come from one carton specification.
Frequently asked questions
Why is 905 wrong if the arithmetic is right? The arithmetic answered "how much volume is there" rather than "how many whole cartons fit". Integer packing reaches roughly 70–80% of theoretical cube on luggage cartons.
Is a 40HQ always better than a 40GP? On cube-heavy cargo, usually — same footprint, 30 cm more height, about 12% more internal volume. On dense cargo that hits the payload ceiling first, the extra height buys nothing.
How much clearance should I allow for carton tolerance? Do not use a blanket percentage. Measure the largest carton and re-run the grid; a single lost row is worth more than a generic cushion.
Can I mix SKUs in one 40HQ? Yes, but plan per carton type and then verify total weight. Volume fitting is not the same as being within payload.
What is the density where a 45HQ stops being volume-bound? 334 kg/m³ at the stated 28,600 kg payload and 85.65 m³ internal volume. Above that, weight binds first and the extra cube is wasted.
Does the planner replace a forwarder's confirmation? No. It gives you a defensible starting number to negotiate from. The forwarder's figure is the one that has to survive contact with a warehouse.
Written by CLK Luggage (Jiangxi Chengleke Leather Co., Ltd). Container internal dimensions and payloads quoted here are the constants the loading planner works from. The 60 × 40 × 35 cm carton and the 60.5 cm tolerance example are illustrative arithmetic, not a specification of any CLK carton.