WEBVTT

00:00:00.000 --> 00:00:02.435 align:middle line:90%
[SQUEAKING]

00:00:02.435 --> 00:00:04.383 align:middle line:90%
[RUSTLING]

00:00:04.383 --> 00:00:05.844 align:middle line:90%
[CLICKING]

00:00:05.844 --> 00:00:15.912 align:middle line:90%


00:00:15.912 --> 00:00:17.370 align:middle line:84%
RICHARD DE NEUFVILLE:
I'm now going

00:00:17.370 --> 00:00:21.960 align:middle line:84%
to give you the logic of the
analysis for the garage case

00:00:21.960 --> 00:00:24.530 align:middle line:90%
as that prototype example.

00:00:24.530 --> 00:00:26.920 align:middle line:84%
The first step is going to
be to consider what I'll

00:00:26.920 --> 00:00:28.510 align:middle line:90%
call the engineering base case.

00:00:28.510 --> 00:00:32.170 align:middle line:84%
That is the way the engineering
profession typically addresses

00:00:32.170 --> 00:00:38.800 align:middle line:84%
such a question and show how it
leads to a particular solution.

00:00:38.800 --> 00:00:43.570 align:middle line:84%
And we'll be showing then how
an analysis, understanding

00:00:43.570 --> 00:00:45.100 align:middle line:84%
the uncertainties
and everything,

00:00:45.100 --> 00:00:47.360 align:middle line:84%
will give you a
different solution.

00:00:47.360 --> 00:00:51.580 align:middle line:84%
So the second step, having
set up the base case,

00:00:51.580 --> 00:00:54.220 align:middle line:84%
is to look at how
that system would

00:00:54.220 --> 00:00:56.590 align:middle line:90%
perform under uncertainty.

00:00:56.590 --> 00:00:59.780 align:middle line:84%
And that tells us that
there is, in fact,

00:00:59.780 --> 00:01:03.610 align:middle line:84%
some considerable
downside, and we would want

00:01:03.610 --> 00:01:05.600 align:middle line:90%
to do something about that.

00:01:05.600 --> 00:01:07.990 align:middle line:84%
And the third one
then is to say,

00:01:07.990 --> 00:01:11.350 align:middle line:84%
let's look at how we might
have a flexible design, how

00:01:11.350 --> 00:01:15.280 align:middle line:84%
we could then profit
from it by not building

00:01:15.280 --> 00:01:17.710 align:middle line:84%
so aggressively at the
beginning because one has

00:01:17.710 --> 00:01:21.190 align:middle line:84%
the option, the possibility
of expanding later

00:01:21.190 --> 00:01:25.060 align:middle line:84%
on as needed, which has the
advantage of deferring the cost

00:01:25.060 --> 00:01:28.850 align:middle line:84%
and maybe saving them completely
if they're not needed.

00:01:28.850 --> 00:01:33.130 align:middle line:84%
So we're going to have three
steps of the total analysis.

00:01:33.130 --> 00:01:36.690 align:middle line:84%
We'll start with the
engineering base case.

00:01:36.690 --> 00:01:39.660 align:middle line:84%
So the deterministic design
is for a major garage

00:01:39.660 --> 00:01:41.790 align:middle line:90%
serving a mega-mall.

00:01:41.790 --> 00:01:47.980 align:middle line:84%
In this case, the Bluewater
case in England, for example.

00:01:47.980 --> 00:01:51.540 align:middle line:84%
The actual demand over time
is, of course, very uncertain.

00:01:51.540 --> 00:01:52.380 align:middle line:90%
Why is that?

00:01:52.380 --> 00:01:55.020 align:middle line:84%
Because we don't know
the population growth.

00:01:55.020 --> 00:01:57.750 align:middle line:84%
We don't know the demographics
of the area over the next 20

00:01:57.750 --> 00:01:58.390 align:middle line:90%
years.

00:01:58.390 --> 00:02:01.080 align:middle line:84%
We don't know whether
the mall will succeed.

00:02:01.080 --> 00:02:03.300 align:middle line:84%
Maybe they have the
wrong stores in it.

00:02:03.300 --> 00:02:04.440 align:middle line:90%
Maybe people don't like it.

00:02:04.440 --> 00:02:05.567 align:middle line:90%
Maybe it'll be wonderful.

00:02:05.567 --> 00:02:06.150 align:middle line:90%
We don't know.

00:02:06.150 --> 00:02:08.669 align:middle line:84%
We just cannot
know that for sure.

00:02:08.669 --> 00:02:12.630 align:middle line:84%
There may be competition also
from other parking facilities.

00:02:12.630 --> 00:02:16.292 align:middle line:84%
If somebody else opens
a garage and gives you

00:02:16.292 --> 00:02:18.000 align:middle line:84%
some kind of access
to the shopping mall,

00:02:18.000 --> 00:02:19.780 align:middle line:90%
maybe you'd prefer that.

00:02:19.780 --> 00:02:25.860 align:middle line:84%
So despite those realities that
the future is very uncertain,

00:02:25.860 --> 00:02:29.040 align:middle line:84%
must be uncertain, the
engineering process

00:02:29.040 --> 00:02:31.020 align:middle line:90%
does it to a fixed forecast.

00:02:31.020 --> 00:02:34.080 align:middle line:84%
In this particular case,
some traffic engineers,

00:02:34.080 --> 00:02:36.630 align:middle line:84%
some urban consultants
went through

00:02:36.630 --> 00:02:41.550 align:middle line:84%
and predicted a forecast for
the situation against which

00:02:41.550 --> 00:02:45.390 align:middle line:84%
the engineering, the structural
engineers that were doing it

00:02:45.390 --> 00:02:49.170 align:middle line:84%
said, OK, what is
the best way to meet

00:02:49.170 --> 00:02:53.010 align:middle line:84%
that particular forecast
given that forecast?

00:02:53.010 --> 00:02:56.340 align:middle line:84%
Even though that
forecast is a fairytale

00:02:56.340 --> 00:02:59.100 align:middle line:84%
because the world will
actually be different than what

00:02:59.100 --> 00:03:01.150 align:middle line:90%
that forecast will be.

00:03:01.150 --> 00:03:03.430 align:middle line:90%
So they will optimize it.

00:03:03.430 --> 00:03:05.650 align:middle line:84%
They'll find the
highest value design.

00:03:05.650 --> 00:03:07.550 align:middle line:90%
That's what optimization means.

00:03:07.550 --> 00:03:11.080 align:middle line:84%
And in this case, you can do
it fairly straightforwardly

00:03:11.080 --> 00:03:13.630 align:middle line:84%
because the design
variable, given

00:03:13.630 --> 00:03:17.860 align:middle line:84%
that you have a fixed area, is
how many floors do you have?

00:03:17.860 --> 00:03:21.175 align:middle line:84%
One, two, three, four,
six, seven, eight.

00:03:21.175 --> 00:03:23.410 align:middle line:84%
So you can look at those
cases and figure out

00:03:23.410 --> 00:03:27.610 align:middle line:84%
which is the best one, which
is what the process leads to.

00:03:27.610 --> 00:03:31.240 align:middle line:84%
And typically in this
kind of situation,

00:03:31.240 --> 00:03:34.480 align:middle line:84%
in the design situation you
have some kind of sweet spot.

00:03:34.480 --> 00:03:36.160 align:middle line:84%
That is if you
build it too small,

00:03:36.160 --> 00:03:38.710 align:middle line:84%
you don't have enough
revenues to justify the cost

00:03:38.710 --> 00:03:40.060 align:middle line:90%
of the leases and everything.

00:03:40.060 --> 00:03:42.490 align:middle line:84%
If it's too large, you
may have a white elephant.

00:03:42.490 --> 00:03:44.770 align:middle line:84%
Somewhere in the
middle typically

00:03:44.770 --> 00:03:46.780 align:middle line:90%
there is a sweet spot.

00:03:46.780 --> 00:03:50.920 align:middle line:84%
Think for a moment what kind
of graph that would have.

00:03:50.920 --> 00:03:53.600 align:middle line:90%
What would happen in that case?

00:03:53.600 --> 00:03:55.810 align:middle line:90%
What would the graph look like?

00:03:55.810 --> 00:03:58.980 align:middle line:84%
Here is the response
that you have.

00:03:58.980 --> 00:04:01.840 align:middle line:84%
On the one side,
if it's too small,

00:04:01.840 --> 00:04:06.120 align:middle line:84%
this is one or two floors, you
have that you're in losses.

00:04:06.120 --> 00:04:08.363 align:middle line:84%
If it's too large,
you can have losses

00:04:08.363 --> 00:04:09.780 align:middle line:84%
because you have
a white elephant.

00:04:09.780 --> 00:04:12.480 align:middle line:84%
Somewhere in here you
have a sweet spot.

00:04:12.480 --> 00:04:15.720 align:middle line:84%
And in this particular design
with the particular values

00:04:15.720 --> 00:04:18.240 align:middle line:84%
that we use, which were
representative but they weren't

00:04:18.240 --> 00:04:22.260 align:middle line:84%
exactly the ones for the actual
facility, which we didn't have

00:04:22.260 --> 00:04:24.210 align:middle line:84%
access to, but they
were representative,

00:04:24.210 --> 00:04:25.560 align:middle line:90%
there is a sweet spot.

00:04:25.560 --> 00:04:31.210 align:middle line:84%
The optimal design, the one
that maximizes the expected

00:04:31.210 --> 00:04:34.060 align:middle line:90%
present value is six stories.

00:04:34.060 --> 00:04:35.430 align:middle line:90%
How did we calculate that?

00:04:35.430 --> 00:04:38.980 align:middle line:84%
We used a spreadsheet
where on the one hand

00:04:38.980 --> 00:04:42.150 align:middle line:84%
we had all the various costs
and revenues down here.

00:04:42.150 --> 00:04:45.300 align:middle line:84%
We had what might happen
in different years

00:04:45.300 --> 00:04:48.870 align:middle line:84%
as the demand grew and as
the performance went out.

00:04:48.870 --> 00:04:51.570 align:middle line:84%
And that leads to a
simple calculation

00:04:51.570 --> 00:04:55.980 align:middle line:84%
of the net present
value as shown here.

00:04:55.980 --> 00:04:58.440 align:middle line:84%
The optimum design
is six stories.

00:04:58.440 --> 00:05:01.860 align:middle line:84%
That is the result of
the typical engineering

00:05:01.860 --> 00:05:05.060 align:middle line:90%
process applied in this case.

00:05:05.060 --> 00:05:10.149 align:middle line:90%