Short answer. Stowing a bag in an overhead bin makes a passenger stop in the aisle and can delay people behind them. In this simulator, heavy carry-on nearly doubled the boarding time of a random queue on a nearly full A320 and more than doubled it for back-to-front zones. The Steffen method was affected least because it lets many passengers stow their bags at the same time.
Why a bag costs more than the time it takes to stow
Lifting a small bag into a bin takes a few seconds; a large case that needs rearranging takes considerably longer. Model assumption: In this model, a small bag takes 2 to 4 seconds to stow, a typical cabin bag 6 to 12 seconds, and two bags 10 to 16 seconds. A passenger lifting a bag into the bin blocks the aisle while doing so.
The cost extends beyond the time spent stowing the bag. People queued behind the passenger may also be delayed. If ten passengers are queued behind someone stowing a bag for ten seconds, that bag could add a hundred seconds to their combined waiting time. This is why boarding order and the amount of luggage interact so strongly.
Boarding time by luggage level
Simulator result: The simulator results below are for an easyJet A320 with 175 passengers boarding through the forward door only. Each time is the median of eight seeded runs (seeds 20260801 to 20260808). "None" means personal items only; "heavy" means most passengers bring a full-size cabin bag and some bring two.
| Luggage | Random queue | Back-to-front | WILMA | Steffen method |
|---|---|---|---|---|
| None | 10:10 | 11:18 | 7:39 | 6:52 |
| Light | 11:03 | 13:23 | 8:57 | 7:12 |
| Typical | 14:50 | 19:15 | 12:31 | 8:21 |
| Heavy | 19:08 | 27:14 | 17:24 | 9:31 |
Simulator result: From no bags to heavy carry-on, the random queue took about 88% longer, back-to-front about 141% longer, WILMA about 127% longer, and the Steffen method about 39% longer.
Why some methods cope better
Back-to-front can concentrate passengers with bags in the same part of the cabin. When people in a zone arrive at nearby rows, a passenger stowing a bag may have a queue behind them.
WILMA reduces the need to squeeze past seated neighbours, but window passengers may still queue behind others stowing bags. In these runs, it remained faster than random boarding at every luggage level shown, although the gap narrowed with heavy bags.
The Steffen method places consecutive passengers two rows apart. This gives passengers in a wave more room to reach their rows and stow bags at the same time. More bags make each wave slower, but spreading passengers out can limit queues behind them.
When the bins run out
Model assumption: With the overhead-bin capacity option on, each row has a fixed amount of bin space. A passenger who finds the bins near their row full looks a few rows either side, which costs extra time. If there is still no space, the crew gate-checks the bag. Simulator result: With heavy luggage and bin capacity on, the random A320 boarding went from 19:08 to 26:03, and back-to-front from 27:14 to 38:19, while Steffen went from 9:31 to 11:05.
Frequently asked questions
Do carry-on bags really slow down boarding?
In the simulator, luggage made a substantial difference to boarding time. With a random queue on a nearly full A320, heavy carry-on took 19:08, compared with 10:10 with personal items only.
Which boarding method copes well with lots of carry-on luggage?
Simulator result: The Steffen method was fastest among the methods shown: 9:31 with heavy luggage on the reference flight, compared with 17:24 for WILMA and 27:14 for back-to-front. Methods that spread passengers along the cabin can let bags be stowed in parallel.
Would checking more bags make boarding faster?
In the model, removing bags that would otherwise be stowed in the cabin reduces aisle stops. Whether checking them saves time overall depends on the time the airport and airline need to handle the extra checked bags, which this simulator does not include.
Updated 2026-09-25. Figures labelled simulator result are outputs of this site's model at the settings stated, not airline measurements. What the model leaves out.