Who goes first?# SLOP2710 · The Economics of Waiting
Do not introduce yourself yet. Put the question up, leave it for a beat, then
go straight to the coffee shop photo. The introduction can wait ten minutes;
the question cannot be un-asked once they have started answering it.
There is a coffee shop downstairs# Between 9:00 and 9:20 it serves about thirty people, on two staff.
For most of that twenty minutes, the line is out the door.
Why is the line that shape ?# Not “why is there a line” — that one is easy. Push them past it. Expect
someone to say “not enough staff” within thirty seconds; that is the answer the
whole course is designed to complicate, so write it on the board and come back
to it in week 4.
Not one decision. Dozens.#
along the window, not doubling back
one queue, not two
the person taking orders is not the person making the drink
a filter coffee waits behind four oat lattes
ρ —
L —
W —
W90 —
In line now — What to watch: Same floor, same two staff, same customers arriving at the same rate — only the arrangement changes. One line puts its tail out of the door. Two lines halve it and let a till stand idle beside a full one. Order-here-collect-there uses the same two people and is visibly shorter. Along the foot of the frame is that arrangement's rule written out as an instruction, and the figure in pink is whoever it has already chosen.Run all three at the default load without commentary. Then ask which one they
would build, take a vote, and only then show the numbers: 4.7 minutes against
0.8, on identical staff.
If there is time, push the load past two a minute and let them watch the third
arrangement break first. Each of these rooms is optimal for a different
morning, and nobody in that building chose which morning.
Some of it was designed by somebody thinking about throughput.
At least one of it was a shopfitter in 2019 thinking about power outlets.
This is the line that lands. Pause after it.
One of those decisions is costing about four minutes per customer at peak and
nobody in the building could tell you which one.
Four knobs# Every queue, everywhere:
arrivals — when people turn upservice — how long each takesservers — how many at oncediscipline — who goes next Almost everything in this course# is a consequence of turning one knob# And it is the fourth one. Nobody has strong feelings about inter-arrival
variance. Everybody has strong feelings about queue-jumping.
The claim#
A queue is not a thing that happens to people.
It is a rule, applied to people, by someone, usually invisibly.
Which means# Deciding who goes first is deciding whose time is spent .
Time is the one thing everybody in the line is paying with, whether they agreed
to or not.
So the rule is moving cost from one group of people to another —
and almost never saying so.
There is no neutral queue# Thesis. It will come back in week 4 with numbers attached, in week 8 with a
price attached, and in week 12 with a transplant list attached. Say it now so
they recognise it later.
This is not a queueing theory course# You will have the core of the mathematics by week 3.
It takes about forty minutes.
The mathematics answers how long .
It does not answer who .
The rule we will hold you to# Name the rule before you judge it.
Most complaint about waiting turns out to be complaint about a rule the
complainer has never articulated.
Once you can state it, you can ask the useful question: what would it cost to
change?
Next week# Two systems. Same throughput, same staff, same speed.
One has no queue. The other has eleven people in it.
Do not resolve this. Let them think it is a trick.
Also flag the aircraft here — one sentence, no more — so that when it comes
back in week 10 they remember it was promised in week 1.
Before Thursday# Go and stand in front of a queue for ninety seconds.
Write down the rule it is running, as an instruction beginning
“when a server becomes free, serve…”
Bring it to the lab.
Warn them: nearly everyone writes “first come first served” and is wrong on a
second look. That discovery is the entire point of lab 1, so do not spoil it
here.