Flight Line
Paper aeroplane production · four stations · twenty aircraft · one clock that never stops
Read this out to the four students — it is how their phones find your line. Any short word will do. Use a different one for each group running at the same time.
Twenty is the standard run.
This sets takt time — the drumbeat the line must keep to. Twenty aeroplanes in thirty minutes is one every 90 seconds. Leave it empty and the charts simply show no target.
Write these on the board, send them to the class, or open one on each phone yourself before you hand them out. Each page remembers its line code, so a student only types it once.
Press when a station finishes an aeroplane — that same moment is the next station's start. Keyboard: 1 2 3 4
Average per plane is how long that one person takes to do their own job on a single aeroplane. Time working is all of that added up. Time waiting is time they sat there with nothing to do, because the aeroplane had not reached them yet — that is not their fault, and it is the waste this whole activity is about. How much it varied tells you whether they were steady or all over the place.
Write it down before you run round two, and be specific. "Work harder" is not an improvement and will not show up in the numbers. "Move the labelling step onto Station C" is, and will.
Why did speeding up the fast station change nothing?
The line delivers one aeroplane every time the bottleneck finishes one. Every other station is capable of more than that, so their spare speed has nowhere to go. This is the Theory of Constraints in one sentence: an hour saved at the bottleneck is an hour gained for the whole line, and an hour saved anywhere else is an illusion.
Who looked busiest, and who actually was?
The station with the biggest pile in front of it looks overwhelmed and usually is. The station with an empty desk looks idle and is being starved. Neither is a comment on the student sitting there.
Would a bigger batch have helped?
Try passing five aeroplanes at a time instead of one. Throughput barely moves, work in progress jumps, and the average aeroplane takes far longer to get through. That is the argument for one-piece flow, and it is much more convincing when a class has just measured it themselves.
What would happen if Quality Check moved to the front?
It cannot — you cannot inspect an aeroplane that has not been made. But each station could check its own work before passing it on. Compare the effort thrown away on rejects with the few seconds that self-checking would have cost.
One row per aeroplane. Under each station letter is the time on the clock when that station finished it, and in brackets how long they actually spent on it.
Then, for the whole aeroplane: Total time is how long it took from being picked up at Marking to leaving Quality Check. Being worked on is the part where somebody was actually touching it. Sat waiting is the rest — time it spent in a pile. Those two always add up to the total, so you can check the maths yourself on any row.
Click any time to correct it if somebody tapped late or missed one — a single wrong tap otherwise spoils two averages. Corrections are saved as corrections; nothing is ever quietly overwritten.
The Average row at the bottom shows, under each station letter, how long that station takes on one aeroplane. Only the bracketed figure can be averaged — the average of a set of clock readings would just tell you where the middle of the round fell.
1 · Set the room up
Four students in a row, in this order: Marking, Folding, Labelling, Quality Check. A stack of A4 beside Marking. Each student needs their own phone with their own station page open. Put this screen where all four can see it.
2 · One rule, and it is the whole activity
Tap your button the instant you finish an aeroplane — not when you pass it on, not when you pick the next one up. That single moment is three things at once: you finished this one, you start your next one, and the station after you starts on this one.
3 · Let it go wrong
Do not tell them to balance the line. The pile in front of the slow station and the boredom at the end of it are the lesson. Let the first round be as bad as it wants to be.
4 · End the round and read it together
Press End round and go to Analysis. Ask them to guess the bottleneck before you show it. Ask the student who waited most how it felt, then show them their waiting time in minutes.
5 · Change one thing, then run it again
Write the plan on the Improve tab, press New round, and run twenty more. One change only — two changes and nobody can say which one worked. The Compare tab does the before-and-after.
If the wifi is bad
Use solo mode. One person taps all four buttons on this screen, keys 1–4. Every chart and every number is identical — the only thing you lose is the students holding their own phones.
If a tap gets missed
Carry on. Go to the Data tab afterwards and click the empty cell's neighbours to correct them, or click a wrong time to fix it. Readings that cannot be true are left out of the averages rather than allowed to distort them, and the Analysis tab says which ones.
Plane
The time is already recorded. Take as long as you like.
Which station caused the fault?