How Many Dock Doors Does a Warehouse Need? A Door-Hours Method

The new distribution centre opened with the number of doors the developer's brochure suggested for its floor area. By the second Monday, eight trailers and three ocean containers were lined up on the access road at 07:30, every door was occupied by a container being hand-unloaded, and the outbound team could not start preloading for the afternoon carriers. Nothing was wrong with the daily volume. The doors were sized for an average hour that never happens.

The door-hours method replaces ratios with arithmetic: count how long each kind of vehicle actually holds a door, add up that time for the busiest window of the day, and divide by the door time you can realistically use. Done separately for receiving and shipping, it gives a defensible count, a yard queue estimate and a cost for every door added or removed.

• What this guide covers: the door-hours equation, measuring occupancy time by load type, the peak window, a utilisation ceiling with a queueing check, the inbound and outbound split, and door types.
• Worked numbers: a US regional DC and an Indian 3PL site sized step by step, with door costs and the cost of a short dock in dollars and rupees.

Why floor-area ratios and daily truck totals undercount dock doors

Dock doors are a time resource, not a space resource, so any rule based on square footage or daily truck counts ignores the two things that decide the answer: how long each vehicle sits at a door and when the vehicles arrive. Two buildings of the same size can need very different door counts.

• Floor-area ratios: a cross-dock and a slow-moving spare parts store of equal area handle completely different truck volumes.
• Daily truck totals: 60 trucks spread over 16 hours and 60 trucks bunched into four morning hours need different docks.
• Single average service time: a palletised trailer may hold a door for under an hour while a floor-loaded container holds it for three or four.
• Ignoring held doors: a dropped outbound trailer being preloaded occupies its door for the whole preload period, not only the loading minutes.

The planning question this article answers comes before scheduling. Our guide to dock appointment scheduling deals with booking the doors a building already has; here the aim is to decide how many to build or lease. The warehouse design process article shows a quick door check within a whole-building design, and this page expands that step into a full method.

Door-hours: the equation behind every dock door count

A door count is total door-hours demanded in a window divided by usable door-hours per door in that window, rounded up. Each vehicle type gets its own occupancy time, and the calculation runs separately for receiving and shipping before any doors are shared.

Door count for one flow (inbound or outbound) in one planning window:

  Doors = ceil( SUM over vehicle types k of ( n_k * t_k ) / ( W * U ) )

  n_k  = vehicles of type k expected in the window (peak window, not daily average)
  t_k  = door occupancy hours per vehicle of type k
       = t_spot + t_handle + t_check + t_release
         t_spot    : back-in, restraint or chocks, leveller deployed
         t_handle  : unload or load = units / crew rate (pallets/h or cartons/h)
         t_check   : count, seal, paperwork, damage notes
         t_release : leveller stowed, restraint off, vehicle pulled
         (for drop trailers held at the door, t_k = placement to pull time)
  W    = window length in hours
  U    = target utilisation, 0 < U < 1 (planning choice, see queue check)

Little's Law cross-check (Little, 1961):
  average doors busy = arrival rate (vehicles/h) * average t_k
  average vehicles waiting = arrival rate * average wait in the yard

Example: 22.4 vehicles in a 4 h window, door-hours 35.0, U = 0.80
  Doors = ceil( 35.0 / (4 * 0.80) ) = ceil( 10.94 ) = 11

Little's Law, published by John Little in 1961, states that the average number of items in a system equals their arrival rate multiplied by the average time each spends there. Applied to a dock, the average number of busy doors equals vehicles per hour times average door occupancy, which gives a quick sanity check on any proposed count.

Dock door sizing by the door-hours method divides peak-window door demand, vehicles multiplied by measured occupancy hours from dock door timestamps, by window length times a utilisation target below 1.0; Little's Law confirms the average busy-door figure, and a target near 0.80 keeps average yard waits under 20 minutes in the worked example.

Measuring door occupancy time for each vehicle and load type

Occupancy time must be measured from door events, not estimated from labour standards alone. The clock starts when the vehicle is placed at the door and stops when the door is free for the next one, so it includes spotting, restraint, checks and release as well as handling.

Most sites already hold the timestamps: a yard system records door assignment and departure, a WMS records receipt start and close, and a dock door sensor or RFID dock door portal can log the first and last read. Take at least four weeks of events, split by vehicle and load type, and use a high percentile as well as the mean.

The values below are illustrative occupancy times for common load types, built from the four components in the equation. Replace them with your own door events; crew size and the product's handling difficulty move them far more than the vehicle does.

From peak day to peak hour: shaping the arrival profile

Size each flow for its busiest window, not its daily total. Take the design-year peak day, then use the hourly arrival profile to find the two- to four-hour block with the most door-hours demanded. Receiving usually peaks in the morning; shipping peaks before carrier departure times.

• Peak day: the 90th to 95th percentile day of the design year, or the day used for labour planning, not the average day.
• Arrival profile: the share of each vehicle type arriving in each hour, from gate timestamps; appointments shift it only if they are enforced.
• Window length: the shortest block in which the peak would have to be cleared to protect downstream work, such as putaway before the day's picking starts.
• Carry-over: long services that start late in the window, such as containers, run past its end, so count their door-hours inside the window that holds most of their duration.

The arithmetic difference is large. The US example later in this article needs 8 receiving doors if its 100 door-hours are spread evenly across a 16-hour shift at 80% utilisation, but 11 doors once 35% of arrivals are placed in the 06:00 to 10:00 block, where they actually come.

Receiving door counts sized from yard gate timestamps on the peak window, rather than the daily average, rose from 8 to 11 doors in the worked US example, because 35% of inbound vehicles and 35 door-hours fell between 06:00 and 10:00, against 100 door-hours spread across a 16-hour receiving shift.

Utilisation ceiling and the yard queue near full door use

Never size doors to 100% utilisation. Arrivals are uneven and service times vary, so waiting time grows sharply as busy doors approach the total. A target utilisation is a choice about how long trucks will wait, and a queueing estimate makes that choice visible.

The table uses the Erlang C formula for random arrivals, a standard steady-state approximation, for the US peak: 22.4 vehicles over four hours, average occupancy 1.56 hours, an offered load of 8.75 busy doors. Real peaks are not steady states, so treat the waits as indicative and check them against your yard history.

The jump from 11 doors to 9 is not a modest slowdown; the queue becomes unstable and carries into the afternoon. Average trucks queued is also the yard parking the site must hold for waiting vehicles, by Little's Law, so the door count and the number of yard spots have to be read together.

1. 06:50, gate log G-1104: four containers and three trailers check in; nine receiving doors are free, all are assigned within ten minutes.
2. 07:30, door board: eight more trailers and three containers arrive. Four doors are held by containers that will take another three hours of hand unloading, so arriving trailers queue on the access road.
3. 09:15, WMS receipt queue: five trailers have waited more than 90 minutes; two carriers start detention clocks under their rate confirmations, and putaway labour stands idle waiting for receipts.
4. 11:40, outbound plan: two doors normally used to preload afternoon routes are still occupied by inbound trailers, so preloading for the 16:30 linehaul starts late.
5. Controlled path: with 11 receiving doors and containers routed to two dedicated doors, the queue at 07:30 would have averaged under two trucks and the outbound preload doors would have stayed free.

Inbound, outbound and swing doors: deciding the split

Calculate receiving and shipping separately, then decide which doors can swing between them. Sharing saves doors only when the two peaks are far apart in the day and the staging behind each door can serve both directions without blocking.

• Dedicated doors: right when both peaks overlap, or when inbound and outbound need different equipment, such as container-ready levellers inbound and parcel conveyors outbound.
• Swing doors: useful when receiving peaks at 06:00 to 10:00 and shipping at 16:00 to 21:00; size the swing pool as the smaller of the two flows' extra doors at their peaks.
• Staging behind the door: each outbound door needs floor space for at least one trailer's worth of staged pallets, and each inbound door space for one received load awaiting putaway.
• Drop versus live: dropped trailers held at doors for preloading raise outbound occupancy to the full preload period; our drop-and-hook versus live-load guide covers that trade-off.

Cross-dock flows need special care. A cross-docked receipt holds an inbound door and an outbound door at the same time while freight moves across, so model it as one service occupying two doors, not two independent visits.

Worked door counts for a US regional DC and an Indian 3PL site

Both examples use illustrative volumes and costs, chosen to show how the peak window and vehicle mix change the answer. The US site's receiving peak is driven by containers; the Indian site's shipping peak is driven by an early-morning dispatch band for small vehicles.

• US receiving, daily: 40 trailers at 1.0 h, 16 containers at 3.5 h and 8 parcel drops at 0.5 h give 100 door-hours; over 16 hours at 0.80, 100 ÷ 12.8 = 7.8, so 8 doors.
• US receiving, peak: 35% of each type between 06:00 and 10:00 gives 14 + 19.6 + 1.4 = 35 door-hours; 35 ÷ (4 × 0.80) = 10.9, so 11 doors.
• US shipping, peak: 60% of 48 trailers depart 16:00 to 21:00; 28.8 × 1.25 h = 36 door-hours; 36 ÷ (5 × 0.85) = 8.5, so 9 doors.
• India receiving, peak: 40% of 30 hand-unloaded 32-ft trucks at 2.5 h and 20 small trucks at 0.75 h fall 08:00 to 12:00; 30 + 6 = 36 door-hours; 36 ÷ 3.2 = 11.3, so 12 doors.
• India shipping, peak: 70% of 90 light commercial vehicles leave 04:00 to 08:00; 63 × 0.6 = 37.8 door-hours; 37.8 ÷ (4 × 0.85) = 11.1, so 12 doors, against 4 doors if all 54 daily door-hours were spread over an 18-hour day at 0.85.

The Indian result illustrates a common pattern. Shipping volume per vehicle is small, so daily door-hours look modest, but city delivery routes have to leave in a narrow early band, often because local traffic notifications restrict daytime entry of goods vehicles. Check the notification that applies to your delivery area before fixing the window.

Door types, leveller ranges and mixed truck bed heights

A door count is only valid if each door can serve the vehicles assigned to it. Split the count by door type, because a door that cannot meet a vehicle's bed height or take its length is not available capacity for that vehicle, however idle it looks.

• Bed-height range: list the lowest and highest bed heights in your arriving and departing fleet; levellers must cover that range at a grade forklifts can climb.
• Container doors: ocean containers and some trailers need longer lips or different leveller types; reserve and count those doors separately.
• Small vehicles: light commercial vehicles common on Indian city routes often sit well below trailer bed height, so they may need a lower dock, a ramp or a separate loading bay.
• Restraint and chocks: OSHA 29 CFR 1910.26(d) requires wheel chocks, sand shoes or similar measures while employees are on a dockboard placed on a vehicle, and 1910.178(m)(7) requires brakes set and wheel blocks in place during loading.
• Apron and spacing: door centres and apron depth must let the longest vehicle back in without swinging across a neighbouring door; check against your actual vehicle lengths.

In India no single code sets dock dimensions for warehouses, and fleets mix trailers, rigid trucks and light vehicles with different bed heights. Measuring the actual vehicles at the gate for two weeks is usually the fastest way to settle the door-type mix.

Dock door capacity counts only doors whose leveller range, lip length and restraint suit the arriving vehicle; OSHA 29 CFR 1910.26(d) requires chocks or similar measures while staff cross a dockboard, so a container door, a trailer door and a small-vehicle bay belong in separate door-hours totals.

Door count rules a yard system enforces and inputs a planner owns

Once the doors exist, software should protect the assumptions behind the count, such as door type matching and reserved container doors. The inputs to the count itself, like the peak day and the utilisation target, are planning judgements that a named person signs and revisits each year.

For the yard behind the doors, our guide to trailer pool sizing applies the same timestamp logic to trailers rather than doors; the two calculations share the gate and door events and should be run from the same data.

Review checklist before the dock wall is fixed in the drawing

Dock walls are expensive to change after the slab and steel are designed. Work through these checks with operations, engineering and finance before the drawing is frozen, and keep the calculation file with the design record so it can be rerun as volumes change.

Teams sizing a new dock, or testing whether an existing one can absorb a new customer, can send their arrival data and constraints through our request for proposal page. Related warehouse and yard guides are collected on the Logistics and Supply Chain hub.