--- Jonathan Kalbfeld <[EMAIL PROTECTED]>
wrote:
"So we've noticed that wherever we go it seems as if
the lines seem to grow exponentially:  book store,
post office, bank.  Is there some economic precept to
describe this growth?"

There may well be; however, the problem has also been
addressed from the perspective of the traffic
engineer.  You may benefit by searching the internet. 
The traffic forum, www.trafficforum.de, may be a
start.  It can tend to be highly technical, e.g.
physics papers on granular flows and stuff like that,
but it may point the way to better lay explanations. 
I note this because I'll probably botch the
explanation as I understand it.

I think it stems from the facts that arrivals are not
evenly distributed and that people can be serviced
only so quickly.  Traffic engineers estimate the
service time of a stop sign to be about five seconds
per car.  Therefore a stop sign can service about
60*60/5 = 720 cars per hour, or 12 cars per minute. 
If cars arrived in 5 second intervals there would
always be a car at the sign, but no line waiting to
offically stop.  But cars don't arrive evenly spaced. 
(Arrival rates are described by a Poisson
distribution, if I recall correctly.)  If there are a
couple minutes in a row where more than 12 cars show
up, let's say 16 & 20, then at the end of the second
minute there are 12 cars left waiting to be serviced. 
That's a full minute to clear the waiting cars.  Now
the cars could pile even faster if you have a couple
of greater-than-twelve minutes, or it could clear
(slowly) with a some of less-than-twelve minutes.

I think it's clear that this can quickly get out of
hand with a little bad luck (from the perspective of
the poor slob at the end of the line...).  Of course,
you can say that my explanation doesn't account for
the frequency with which things get out of hand. 
That's fair.  I would respond by suggesting that you
keep track for a while.  This is a memorable
experience and humans commit the "availability
hueristic," that is we estimate probabilities based on
how memorable an event is.  Just for kicks I went to a
shark-fishing java applet here:
http://www.math.csusb.edu/faculty/stanton/m262/poisson_distribution/Poisson.html
and ran the experiment for three sets of five trials
each.  The results were:

1:  14 13 15 16 10
2:  10 16 17 21 14
3:  10  6 11 10 11

Trial 1 gives a line of 8 cars after 5 minutes; trial
2 gives 18; trial 3 gives zero cars in line.  So maybe
my explanation is plausable.

Also interesting is that your perspective is from one
who's *not* in line!  By definition your odds of being
in a line are greater since there's more people in
lines--otherwise there'd be no line.  We should go
drinking together...I hate waiting in line for a beer
at the bar.

I hope that made sense, was accurate, and not too
confused,
-jsh


=====
"...for no one admits that he incurs an obligation to another merely because that 
other has done him no wrong."
-Machiavelli, Discourses on Livy, Discourse 16.

__________________________________________________
Do You Yahoo!?
Yahoo! Finance - Get real-time stock quotes
http://finance.yahoo.com

Reply via email to