WEBVTT

00:00:00.000 --> 00:00:01.936 align:middle line:90%
[SQUEAKING]

00:00:01.936 --> 00:00:04.356 align:middle line:90%
[RUSTLING]

00:00:04.356 --> 00:00:07.744 align:middle line:90%
[CLICKING]

00:00:07.744 --> 00:00:13.870 align:middle line:90%


00:00:13.870 --> 00:00:16.420 align:middle line:84%
RICHARD DE NEUFVILLE: So just
imagine that instead of-- this

00:00:16.420 --> 00:00:19.300 align:middle line:84%
is for two choices, two
outcomes, two stages.

00:00:19.300 --> 00:00:22.060 align:middle line:84%
Just imagine extension
of three of each,

00:00:22.060 --> 00:00:27.280 align:middle line:84%
and you're now talking about
a really, really complicated,

00:00:27.280 --> 00:00:28.300 align:middle line:90%
just to set up.

00:00:28.300 --> 00:00:31.660 align:middle line:84%
And deliberately I haven't
tried to show you the way

00:00:31.660 --> 00:00:34.370 align:middle line:84%
to calculate the revisions
of the prior probability

00:00:34.370 --> 00:00:36.610 align:middle line:90%
as we learn about the future.

00:00:36.610 --> 00:00:41.320 align:middle line:84%
Bottom line here is, there's
a good chance of a messy bush.

00:00:41.320 --> 00:00:47.390 align:middle line:84%
So it's something
to warn you about.

00:00:47.390 --> 00:00:54.220 align:middle line:84%
And there's a reason, however,
to show you this, one of which

00:00:54.220 --> 00:00:59.080 align:middle line:84%
is because other people use it,
as Chevron's decision analysis

00:00:59.080 --> 00:00:59.980 align:middle line:90%
manual.

00:00:59.980 --> 00:01:03.910 align:middle line:84%
But it also allows you at
least to have some insights

00:01:03.910 --> 00:01:05.530 align:middle line:84%
into something
important, which is

00:01:05.530 --> 00:01:07.300 align:middle line:84%
the value of
information, which would

00:01:07.300 --> 00:01:10.850 align:middle line:90%
be the topic of next time.

00:01:10.850 --> 00:01:15.880 align:middle line:84%
So let's now step back
a little bit from it,

00:01:15.880 --> 00:01:19.300 align:middle line:84%
but to realize
that with the end,

00:01:19.300 --> 00:01:20.920 align:middle line:84%
as a result of
decision analysis,

00:01:20.920 --> 00:01:22.610 align:middle line:90%
we don't have a simple plan.

00:01:22.610 --> 00:01:28.420 align:middle line:84%
So instead of saying, all
right, our best approach

00:01:28.420 --> 00:01:30.400 align:middle line:84%
is to listen to the
weather forecast

00:01:30.400 --> 00:01:33.280 align:middle line:90%
and then to take the raincoat--

00:01:33.280 --> 00:01:35.890 align:middle line:84%
no, it's to listen to
the weather forecast.

00:01:35.890 --> 00:01:38.560 align:middle line:84%
If one thing happens, then
maybe we take the raincoat.

00:01:38.560 --> 00:01:41.890 align:middle line:84%
If another thing happens, then
we don't take the raincoat.

00:01:41.890 --> 00:01:46.530 align:middle line:84%
So it is not, OK,
step one, phase one,

00:01:46.530 --> 00:01:48.810 align:middle line:84%
phase two, phase three,
here is the right choice.

00:01:48.810 --> 00:01:51.630 align:middle line:84%
It is, OK, here is
the right start,

00:01:51.630 --> 00:01:56.890 align:middle line:84%
and then we should do
something else later on.

00:01:56.890 --> 00:02:00.510 align:middle line:84%
So for example, you
might say the first stage

00:02:00.510 --> 00:02:02.070 align:middle line:90%
is to build your plant.

00:02:02.070 --> 00:02:07.550 align:middle line:84%
In the second phase, if
there is growth, expand it.

00:02:07.550 --> 00:02:12.200 align:middle line:84%
If there's no particular
growth, don't do anything.

00:02:12.200 --> 00:02:16.310 align:middle line:84%
If it's failing, then
close it, for example.

00:02:16.310 --> 00:02:19.130 align:middle line:90%
So you don't have one plan.

00:02:19.130 --> 00:02:24.870 align:middle line:84%
You have a strategy, a dynamic
plan that lasts through time.

00:02:24.870 --> 00:02:29.630 align:middle line:84%
Now, coming back to the
decision analysis consequences,

00:02:29.630 --> 00:02:31.640 align:middle line:84%
is it going through
this kind of process

00:02:31.640 --> 00:02:34.130 align:middle line:84%
of showing the
complexity of it, is

00:02:34.130 --> 00:02:38.420 align:middle line:84%
an education for the client or
the person you're trying to do.

00:02:38.420 --> 00:02:42.830 align:middle line:84%
And it also shows a
strategy, a dynamic plan,

00:02:42.830 --> 00:02:46.760 align:middle line:84%
is better than having a step
one, step two, step three,

00:02:46.760 --> 00:02:47.510 align:middle line:90%
a phase.

00:02:47.510 --> 00:02:50.360 align:middle line:84%
This is what we're going
to do through time.

00:02:50.360 --> 00:02:54.570 align:middle line:84%
Just generally, they don't
work out that way anyway.

00:02:54.570 --> 00:02:57.520 align:middle line:84%
Moreover, often it's a
second best strategy that

00:02:57.520 --> 00:02:59.730 align:middle line:90%
is optimal for any one outcome.

00:02:59.730 --> 00:03:03.040 align:middle line:84%
For example, it's
best to buy insurance.

00:03:03.040 --> 00:03:07.410 align:middle line:84%
So thinking about you
can choose to have a car

00:03:07.410 --> 00:03:10.860 align:middle line:84%
or not, if it's
within your means,

00:03:10.860 --> 00:03:13.350 align:middle line:84%
and then if you
have an accident,

00:03:13.350 --> 00:03:17.370 align:middle line:84%
and you're really
hurt and everything,

00:03:17.370 --> 00:03:22.020 align:middle line:84%
you'll look back and say,
driving was not a good idea.

00:03:22.020 --> 00:03:24.750 align:middle line:84%
And it would have been
better not to drive at all.

00:03:24.750 --> 00:03:28.320 align:middle line:84%
If you have no accident,
you might say, gee,

00:03:28.320 --> 00:03:33.105 align:middle line:84%
my best choice would have
been to not have insurance.

00:03:33.105 --> 00:03:37.150 align:middle line:84%
Insurance was a bad idea to
have because I didn't use it.

00:03:37.150 --> 00:03:42.690 align:middle line:84%
So the optimal choice might be
to drive with insurance, which

00:03:42.690 --> 00:03:44.730 align:middle line:84%
is never the best,
which would not

00:03:44.730 --> 00:03:47.670 align:middle line:84%
be the best choice,
your first choice,

00:03:47.670 --> 00:03:49.350 align:middle line:90%
in terms of the outcomes.

00:03:49.350 --> 00:03:53.640 align:middle line:84%
If the outcome is where you
didn't have any accident,

00:03:53.640 --> 00:03:57.210 align:middle line:84%
you spend all that money
that you didn't need to.

00:03:57.210 --> 00:03:59.040 align:middle line:84%
That would have been
the better choice.

00:03:59.040 --> 00:04:02.040 align:middle line:84%
If you did have a bad
accident, then maybe

00:04:02.040 --> 00:04:05.340 align:middle line:84%
you shouldn't have been doing
that kind of activity anyway.

00:04:05.340 --> 00:04:11.250 align:middle line:84%
So it's the idea that the better
choice, like to buy insurance,

00:04:11.250 --> 00:04:15.390 align:middle line:84%
is not necessarily
the best choice

00:04:15.390 --> 00:04:16.680 align:middle line:90%
for all the possible outcomes.

00:04:16.680 --> 00:04:21.480 align:middle line:90%


00:04:21.480 --> 00:04:25.040 align:middle line:84%
So let me now, step back and
put this decision analysis

00:04:25.040 --> 00:04:28.220 align:middle line:90%
versus flexibility analysis.

00:04:28.220 --> 00:04:30.670 align:middle line:84%
So the decision
analysis assumes you

00:04:30.670 --> 00:04:34.210 align:middle line:84%
can define the choices
to take in any stages.

00:04:34.210 --> 00:04:37.420 align:middle line:84%
That's essential to building
your decision trees,

00:04:37.420 --> 00:04:41.590 align:middle line:84%
to say at stage two, here's
what I can do, at stage three,

00:04:41.590 --> 00:04:44.230 align:middle line:90%
here's what I can do.

00:04:44.230 --> 00:04:46.840 align:middle line:84%
Contrarily,
flexibility encourages

00:04:46.840 --> 00:04:51.590 align:middle line:84%
you to explore what decisions
to take and when to take them.

00:04:51.590 --> 00:04:53.800 align:middle line:84%
It doesn't tell you
what you have to do

00:04:53.800 --> 00:04:55.550 align:middle line:84%
and it doesn't tell
you when to do it.

00:04:55.550 --> 00:04:57.370 align:middle line:84%
So the simulation
analysis allows

00:04:57.370 --> 00:05:00.870 align:middle line:90%
you to look at those easily.

00:05:00.870 --> 00:05:03.920 align:middle line:84%
Secondly, the decision
analysis focuses

00:05:03.920 --> 00:05:05.480 align:middle line:90%
on a few distinct choices.

00:05:05.480 --> 00:05:09.200 align:middle line:84%
We do this, that, or the
other, whereas the flexibility

00:05:09.200 --> 00:05:15.240 align:middle line:84%
analysis allows you, via Excel,
to be as granular as desired.

00:05:15.240 --> 00:05:17.450 align:middle line:84%
You can look through
2,000 simulations

00:05:17.450 --> 00:05:19.730 align:middle line:84%
and it's easy to do,
does it in seconds.

00:05:19.730 --> 00:05:24.960 align:middle line:84%
But this is not really possible
with decision analysis.

00:05:24.960 --> 00:05:31.640 align:middle line:84%
So having said that, well,
it emphasizes the reasons why

00:05:31.640 --> 00:05:34.580 align:middle line:84%
decision analysis,
which was popular,

00:05:34.580 --> 00:05:38.090 align:middle line:84%
and was a breakthrough,
frankly speaking,

00:05:38.090 --> 00:05:42.060 align:middle line:84%
into thinking a long time
ago-- for me a long time ago,

00:05:42.060 --> 00:05:45.590 align:middle line:84%
before you were born,
even, possibly--

00:05:45.590 --> 00:05:49.910 align:middle line:84%
it is really not so relevant
today, although it does exist,

00:05:49.910 --> 00:05:54.350 align:middle line:84%
and is used, and has
some insights, which

00:05:54.350 --> 00:05:58.480 align:middle line:84%
I'll be sharing with you in
terms of value of information.

00:05:58.480 --> 00:06:03.000 align:middle line:90%