Door counts are set by trucks, time at the door and the loading window, not by floor area, and the loading window moves the answer as much as the volume does.
Size dock doors from the trucks, not the floor area: pallets a day divided by pallets per truck gives trucks, trucks times minutes at the door gives door-hours, and door-hours times a peak factor divided by the hours each door is usable gives doors. For an illustrative 30,000 m² warehouse moving 2,000 pallets in and 2,400 out a day, that is 8 inbound and 18 outbound doors, 26 in all, 0.87 per 1,000 m². Compress the outbound window from ten hours to six and the building needs 37.
Worked in full in Logistics and Industrial Real Estate by Julian R. Sterling, with every figure reproduced in a free workbook.See the book on Amazon →
Dock doors are one of the few specification items that cannot be added cheaply once a building is up: each one needs a pit or a leveller, a frame in the cladding and yard depth in front of it. A warehouse with too few doors for the occupier market it will be re-let into loses tenants to the building next door, whatever its clear height. Underwriting the door count means knowing what the likely occupiers need, and that is a throughput calculation.
| Input | Inbound | Outbound |
|---|---|---|
| Pallets a day | 2,000 | 2,400 |
| Pallets per trailer, full / average fill | 33 / 85% | 33 / 85% |
| Minutes per truck at the door, including positioning and paperwork | 60 | 75 |
| Hours a day the flow runs | 16 | 10 |
| Peak-day factor over the average day | 1.30 | 1.30 |
| Target door utilisation within the window | 80% | 80% |
Thirty-three is the number of euro pallets that fit a standard 13.6 m trailer floor; real loads average less. Outbound takes longer per truck because loads are mixed and checked, and it runs in a shorter evening window to meet next-morning deliveries. The figures are illustrative; the method is what transfers.
Pallets per truck = 33 × 85% = 28.05
Inbound: 2,000 / 28.05 = 71.3 trucks a day; × 60 min = 71.3 door-hours; × 1.30 peak = 92.7
Inbound doors = 92.7 / (16 h × 80%) = 92.7 / 12.8 = 7.24, round up to 8
Outbound: 2,400 / 28.05 = 85.6 trucks; × 75 min = 107.0 door-hours; × 1.30 = 139.0
Outbound doors = 139.0 / (10 h × 80%) = 139.0 / 8.0 = 17.38, round up to 18
Total = 26 doors, 0.87 per 1,000 m², one per 1,154 m²
Excel, per flow: =ROUNDUP(Pallets/(33*Fill)*Minutes/60*Peak/(Window*Util),0)
A specification shorthand of one door per 1,000 m² would give 30 for this building. That is not far off here, but only by coincidence: the same building used by a different occupier can need anywhere from 20 to 37 doors, as the next tables show.
| Outbound window, hours | Exact | Outbound doors | Total with 8 inbound |
|---|---|---|---|
| 6 | 28.97 | 29 | 37 |
| 8 | 21.72 | 22 | 30 |
| 10 | 17.38 | 18 | 26 |
| 14 | 12.41 | 13 | 21 |
| Peak factor | Inbound | Outbound | Total | Per 1,000 m² |
|---|---|---|---|---|
| 1.0 | 6 | 14 | 20 | 0.67 |
| 1.3 | 8 | 18 | 26 | 0.87 |
| 1.6 | 9 | 22 | 31 | 1.03 |
The window matters as much as the volume: doors rise in proportion to pallets a day and in inverse proportion to the hours available to move them. A retailer replenishing stores overnight in six hours needs 37 doors in a building where a manufacturer shipping across a fourteen-hour day needs 21. The peak factor matters next: a seasonal occupier with a peak 60 per cent above the average day needs 31 doors, 1.03 per 1,000 m².
If inbound and outbound genuinely run in separate windows, say goods in during the day and loading in the evening, the same doors can serve both, and the building needs the larger of the two counts, 18, not 26. Shared doors save capital and yard but tie the occupier to that schedule; an investor should price the building on the count a typical occupier needs, not on one occupier's rota.
Sizing on the average day at full utilisation. Without the peak factor and with every door busy every minute of the window, the same arithmetic returns 4.46 inbound and 10.70 outbound, 16 doors after rounding up. A building with 16 doors would queue trucks on every peak day, and queues in the yard are paid for by the occupier in driver time and missed slots, which is why occupiers walk away from such buildings at lease expiry. The second mistake is to count doors without the yard: each door needs a trailer's length of apron and somewhere for the next truck to wait.
At an illustrative 25,000 a door, the 26 doors cost 650,000, a small line in a warehouse of this size. The marginal door is cheap at construction and very expensive afterwards, which is why speculative schemes tend to over-provide; how the full specification adds up is in how much it costs to build a warehouse per square metre, and how capacity translates into rent in how many pallet positions fit in a warehouse.
The book's specification checklist covers the dock and yard alongside height, floor loading and power; it is in the free workbook for this case.
There is no single ratio: it follows from throughput, time per truck and the loading window. In the illustrative 30,000 m² case the answer is 0.87 doors per 1,000 m² with a peak factor of 1.30, 0.67 on an average day and 1.03 for a seasonal occupier with a 1.6 peak. A one-per-1,000 m² rule would give 30.
It depends on the load and the handling, but the calculation should use the whole time a truck occupies the door, including positioning, checks and paperwork. The worked case uses 60 minutes inbound and 75 outbound for an average of 28.05 pallets per trailer; mixed or hand-stacked loads take longer and need more doors.
Yes, if the two flows run in separate windows. In the worked case goods in need 8 doors over sixteen hours and loading 18 over ten; if the windows do not overlap, 18 shared doors serve both instead of 26 dedicated ones, at the cost of tying the operation to that timetable.
This article is one calculation from Logistics and Industrial Real Estate. The book takes the same case from first principles to the decision, chapter by chapter, and every figure it prints is a live formula in the free companion workbooks.
Get the book on Amazon →Free companion files
Also on Amazon UK · Amazon Germany · Amazon France · Amazon Canada
Reading guide: real estate investing, finance and fund management → · All 1340 articles →
If this book helped, or didn’t, a few lines on Amazon are worth more than they look: they are what the next reader goes on. Write a review. The workbook stays free either way.