Decision Theory PPT (Mba3) 2011

2008 Prentice Hall, Inc.
A 1
Operations
Research

Decision-Making
Theory and Tools
2008 Prentice Hall, Inc. A 2
Outline
The Decision Process in Operations
Fundamentals of Decision Making
Decision Tables
Decision Making under Uncertainty
Decision Making Under Risk
Decision Making under Certainty
Expected Value of Perfect Information (EVPI)
Decision Trees
A More Complex Decision Tree
A
-
2
Decision analysis
Decision analysis allows us to select a decision from a set of
possible decision alternatives when uncertainties regarding the
future exist.
The goal is to optimize the resulting payoff in terms of a decision
criterion.
For evaluating and choosing among alternatives

Considers all the possible alternatives and possible outcomes

6.1 Introduction to Decision
Analysis
Maximizing expected profit is a
common criterion when
probabilities can be assessed.
4
Maximizing the decision
makers utility function is the
mechanism used when risk is
factored into the decision
making process.
Decision Making.
Decisions are processes by which a manager seeks to achieve some
desired state. They are means rather than ends. Making a
decision involves making a choice between the alternatives.
Decisions could be a) engineering or scientific or b)
management
Decision making is the sequential process of thought and
deliberation that results in a decision.
The process of decision making is same in both the types of
decisions and involves a) defining the problem b) gathering
facts related to the problem c) comparing these with right or
wrong criteria based on knowledge and experience and then
taking the best course of action
Management decision making a more of an art than a science.
Management decisions are tough because management
problems are wider in scope and they are related to human
behavior which is most unpredictable.
Management decisions could be either a) programmed or b)
Non programmed.
While Programmed decisions are repetitive and routine in
nature and provide solutions to structured problems the Non
programmed decisions are of non routine or unique in nature
and attempt to provide solutions to complex and unstructured
problems
Top Broad, unstructured, infrequent, uncertainty
Middle Both structured and unstructured
Lower Frequent, structured, repetitive,
routine, certainty
Programmed decisions
Un Programmed decisions
M
a
n
a
g
e
m
e
n
t

L
e
v
e
l

A
-
7
Decision theory is an area of study
of discrete mathematics, related to and of
interest to practitioners in all branches
of science, engineeringand in all human
social activities.

It is concerned with how real or
ideal decision-makers make or should
make decisions, and how
optimal decisions can be reached.
The decision process
a
Identify & define
The problem
Develop
alternatives
Evaluate
alternatives
Select
alternatives
Evaluate
& control
Implement
decision
Certainty Risk
Uncertainty
Gather
Information
R
e
v
i
s
e

The Decision-Making Process
PowerPoint presentation to
accompany Operations
Management, 6E (Heizer &
Render)
A
-
9
Problem Decision
Quantitative Analysis
Logic
Historical Data
Marketing Research
Scientific Analysis
Modeling
Qualitative Analysis
Emotions
Intuition
Personal Experience
and Motivation
Rumors
The Decision Process in
Operations
1. Clearly define the problems and the
factors that influence it
2. Develop specific and measurable
objectives
3. Develop a model
4. Evaluate each alternative solution
5. Select the best alternative
6. Implement the decision and set a
timetable for completion
Advantages
Are less expensive and disruptive than
experimenting with the real world system
Allow operations managers to ask What if types
of questions
Are built for management problems and
encourage management input
Force a consistent and systematic approach to the
analysis of problems
Require managers to be specific about constraints
and goals relating to a problem
Help reduce the time needed in decision making
A
-
1
1
Limitations of Models
They
may be expensive and time-consuming to develop
and test
are often misused and misunderstood (and feared)
because of their mathematical and logical
complexity
tend to downplay the role and value of non
quantifiable information
often have assumptions that oversimplify the
variables of the real world
A
-
1
2
Factors influencing decision making
Individual differences influence the decision making process.
The four individual differences which have a significant
impact of the decision making process are
1. Values: Values are the guidelines that a person uses when confronted
with a situation in which a choice has to be made. Values are acquired
early in life and are a basic part of individuals thought. Value judgment
is involved at every stage in the process of decision making. They are
reflected in the decision makers behavior before making the decision,
in making the decision and in putting the decision into effect.

2. Personality : Decision makers are influenced by many psychological
forces both conscious and subconscious. These are strongly reflected in
decision making under uncertainty. Personality traits of the decision
maker combine with situational and interact ional variables influence
the decision making process..

3. Propensity for risk : (Risk taking capacity ) This is a specific aspect
of personality which strongly influences the process of making decision.
4. Potential for dissonance : Traditionally researchers have focused
much of their attention on the forces and influences on the decision
maker before a decision is made.
Utility of the alternatives is the criterion for decision making. Value of
the decision is dependent on the utility.
Recently Behavioral scientists have focused their attention on post
decision anxiety or cognitive dissonance experienced by the decision
maker. Such anxiety is related to lack of consistency or harmony among
individuals various cognitions (attitudes, beliefs and so on) Individuals
are likely to use one or more of the following to reduce their dissonance
a. Seek information that supports their decision.
b. selectively perceive information that supports the decision
c. adopt a less favorable view of the foregone alternatives.
d. Exaggerate the importance of positive aspects of the decision
Fundamentals of
Decision Making
1. Terms:
a. Alternative a course of action or
strategy that may be chosen by the
decision maker
b. State of nature an occurrence or
a situation over which the decision
maker has little or no control
Decision Table
A-16
States of Nature
Alternatives State 1 State 2
Alternative 1 Outcome 1 Outcome 2
Alternative 2 Outcome 3 Outcome 4
6.2 Payoff Table Analysis
Payoff Tables

Payoff table analysis can be applied when:
There is a finite set of discrete decision alternatives.
The outcome of a decision is a function of a single future event.
In a Payoff table -
The rows correspond to the possible decision alternatives.
The columns correspond to the possible future events.
Events (states of nature) are mutually exclusive and collectively
exhaustive.
The table entries are the payoffs.
17
Fundamentals of
Decision Making
2. Symbols used in a decision tree:
a. decision node from which one
of several alternatives may be
selected
b. a state-of-nature node out of
which one state of nature will occur

Decisions under uncertainty
3. Uncertainty : The decision maker (dm) has
absolutely no knowledge of the probability of
outcome of each alternative.
When no information exists the personality
characteristics of the decision maker become
more important for determining which decision
is made. The following five characteristics
describe what most of the dms do.
a. Optimistic Decisions
b. Pessimistic Decisions
c. Realistic decisions.
d. Regret minimizing Decisions.
e. Insufficient Reasoner
2.Decision making under risk:-The
decision maker has some probabilistic
estimate of the outcome of each
decision. Condition of risk occurs
when the decision maker has enough
information to allow the use of
probability in evaluating the
alternatives. Probability of occurrence
of an is event is the expectancy of
event happening.
Several states of nature may occur
Each has a probability of occurring
Decisions with risk
Probability can be assigned based on
a. Logic or deduction: This is Objective probability.
This reflects the historical evidence. Ex. Getting
head/tail for a tossed coin. Or getting a number
on rolling dice etc.
b. Past experience is with empirical evidence.
c. Subjective estimate due to intelligence or
intuition.
When the decision maker has access to probability
information, the criterion for decision making is
to maximize s the expected value of the decision.

Decision Making Under Certainty
The consequence of every alternative is known
Usually there is only one outcome for each
alternative
State of nature is known, The decision maker
has a complete knowledge of the
outcome of each alternative.
This seldom occurs in reality
Decision Making Under
Uncertainty - The Maximin
Criterion
23
Decision Making Under Uncertainty

Probabilities of the possible outcomes are not
known
Decision making methods:
1. Maximax
2. Maximin
3. Criterion of realism/The hurwiczcriterion
4. Equally likely/Criterion of rationality./the laplace
criterion
5. Minimax regret/The savage criterion
Uncertainty
Maximax - Choose the alternative that maximizes
the maximum outcome for every alternative
(Optimistic criterion)
Maximin - Choose the alternative that maximizes
the minimum outcome for every alternative
(Pessimistic criterion)
Equally likely or Criterion of rationality/laplace
criterion - chose the alternative with the highest
average outcome.
A
-
2
5
Uncertainty
4.The hurwitz criterion/Criterion of realism
:this criterion represents that the decision making should
not be done by completely optimistic or completely
pessimistic and it is a mixture of both.It introduced
the cofficient of optimism to measure the degree of
optimisim.It is represented by aplha()which lies b/w
0 and 1.This coefficient ranges from complete
pessimistic (0) to complete optimistic(1) attitude
about future.
Uncertainty
5.The savage criterion/Minimax regret
:-It is based on the opportunity loss concept, which
selects the course of actions that minimizes the
maximum regret. It is also known as minimax regret
because after taking wrong alternative decision, which
results loss,the decision maker feels regret.
Criteria of Decision making
a. Optimistic Decisions The DM think optimistically about the
event that influence decisions. They choose the alternative that
maximizes the outcome
b. Pessimistic Decisions They believe that worst possible outcome
will occur no matter what they do. They estimate the worst outcomes
associated with each alternative and select the best of these worst
outcomes.
c. Realistic Decisions. They take the middle path neither optimistic
nor pessimistic.
d. Regret minimizing Decisions. They want to minimize the
dissonance they experience after the fact.
e. Insufficient Reason Decisions. These are also called eqi-
probable decision maker. They assume that all the possible outcomes
have equal chance of occurring.

Example - Decision Making Under
Uncertainty
A firm has two options for expanding production of a
product: (1) construct a large plant; or (2) construct a small
plant. Whether or not the firm expands, the future market for
the product will be either favorable or unfavorable.
If a large plant is constructed and the market is favorable,
then the result is a profit of $200,000. If a large plant is
constructed and the market is unfavorable, then the result is
a loss of $180,000.
If a small plant is constructed and the market is favorable,
then the result is a profit of $100,000. If a small plant is
constructed and the market is unfavorable, then the result is
a loss of $20,000. Of course, the firm may also choose to do
nothing, which produces no profit or loss.

Example - Maximax
A
-
3
0
States of Nature
Alternatives
Favorable
Market
Unfavorable
Market
Construct
large plant
$200,000 -$180,000
Construct
small plant
$100,000 -$20,000
$0 $0
Do
nothing
Maximax decision is to construct large plant.
Example - Maximin
A
-
3
1
Minimum
in Row
-$180,000
-$20,000
$0
Maximin decision is to do nothing.
(Maximum of minimums for each alternative)
States of Nature
Alternatives
Favorable
Market
Unfavorable
Market
Construct
large plant
$200,000 -$180,000
Construct
small plant
$100,000 -$20,000
$0 $0
Do
nothing
Example - Decision Making
Under Uncertainty
A
-
3
2
States of Nature
Alternatives
Favorable
Market
Unfavorable
Market
Maximum
in Row
Minimum
in Row
Row
Average
Construct
large plant
$200,000 -$180,000 $200,000 -$180,000 $10,000
Construct
small plant
$100,000 -$20,000 $100,000 -$20,000 $40,000
Do nothing $0 $0 $0 $0 $0

Maximax Maximin Equally
likely
Decision Making Under Uncertainty
States of Nature
Favorable Unfavorable Maximum Minimum Row
Alternatives Market Market in Row in Row Average
Construct
large plant $200,000 -$180,000 $200,000 -$180,000 $10,000
Construct
small plant $100,000 -$20,000 $100,000 -$20,000 $40,000
Do nothing $0 $0 $0 $0 $0
1. Maximax choice is to construct a large plant
2. Maximin choice is to do nothing
3. Equally likely choice is to construct a small plant
Maximax Maximin
Equally
likely
Probability unknown unknown
Criterion of Realism Uses the coefficient of realism () to estimate the
decision makers optimism
0 < < 1

x (max payoff for alternative)
+ (1- ) x (min payoff for alternative)
= Realism payoff for alternative
Criterion of Realism

Alternativ
es
Outcomes (Demand) Realism
Payoff
High Moderat
e
Low x (max
payoff )
+ (1- ) x
(min payoff
Large
plant
200,000 100,000 -120,000 .45x2ooooo
)+(1-.45)x(-
12oooo)=2
4000
Small
plant
90,000 50,000 -20,000 29500
No plant 0 0 0 0
Suppose = 0.45

Choose small plant

Alternatives
Realism
Payoff
Large plant 24,000
Small plant 29,500
No plant 0
Minimax Regret Criterion
Regret or opportunity loss measures much
better we could have done
Regret = (best payoff ) (actual payoff )

Alternatives
Outcomes (Demand)
High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0
The best payoff for each outcome is
highlighted
Alterna
tives
Outcomes (Demand)
High(Best
payoff-
actual
payoff)
200,000
Moderate
(Best payoff-
actual
payoff)1000,
00
Low
(Best payoff-
actual
payoff)
0
Large
plant
(200000-
200000)=0
(100000-
100000)=0
(0-
(1200000)=1
20,000
Small
plant
(200000-
90000)=110
,000
(100000-
50000)=50,0
00
(0-
(200000)=20
,000
No
plant
(200000-
0)=200,000
(100000-
100,000)=o
(0-0)=0
Regret Values
Max
Regre
t

120,0
00

110,0
00
200,0
00

We want to minimize the amount of regret
we might experience, so chose small plant
TOM BROWN INVESTMENT
DECISION
Tom Brown has inherited $1000.
He has to decide how to invest the money for
one year.
A broker has suggested five potential
investments.
Gold
Junk Bond
Growth Stock
Certificate of Deposit
Stock Option Hedge
40
TOM BROWN
The return on each investment depends on the
(uncertain) market behavior during the year.
Tom would build a payoff table to help make the
investment decision
41
TOM BROWN - Solution
Select a decision making criterion, and apply it
to the payoff table.
42
S1 S2 S3 S4
D1 p
11
p
12
p
13
p
14

D2 p
21
p
22
p
23
P
24

D3 p
31
p
32
p
33
p
34

S1 S2 S3 S4
D1 p
11
p
12
p
13
p
14

D2 p
21
p
22
p
23
P
24

D3 p
31
p
32
p
33
p
34

Criterion
P1
P2
P3
Construct a payoff table.
Identify the optimal decision.
Evaluate the
solution.
The Payoff Table
43
Decision States of Nature
Alternatives Large Rise Small Rise No Change Small Fall Large Fall
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D account 60 60 60 60 60
Stock option 200 150 150 -200 -150
The states of nature are mutually
exclusive and collectively
exhaustive.
Define the states of nature.
DJ A is down more
than 800 points
DJ A is down
[-300, -800]
DJ A
moves
within
[-
300,+300]
DJ A is up
[+300,+1000]
DJ A is up more
than1000 points
The Payoff Table
44
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D account 60 60 60 60 60
Stock option 200 150 150 -200 -150
Determine
the set of
possible
decision
alternatives.
The Payoff Table
45
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D account 60 60 60 60 60
Stock option 200 150 150 -200 -150
The stock option alternative is
dominated by the
bond alternative
250 200 150 -100 -150
-150
TOM BROWN - The Maximin Criterion
To find an optimal decision
Record the minimum payoff across all states of
nature for each decision.
Identify the decision with the maximum minimum
payoff.
46
The Maximin Criterion Minimum
Decisions Large Rise Small rise No Change Small Fall Large Fall Payoff
Gold -100 100 200 300 0 -100
Bond 250 200 150 -100 -150 -150
Stock 500 250 100 -200 -600 -600
C/D account 60 60 60 60 60 60
The Maximin Criterion Minimum
Decisions Large Rise Small rise No Change Small Fall Large Fall Payoff
Gold -100 100 200 300 0 -100
Bond 250 200 150 -100 -150 -150
Stock 500 250 100 -200 -600 -600
C/D account 60 60 60 60 60 60
The Maximin Criterion - spreadsheet
47
=MAX(H4:H
7)
* FALSE is the range lookup
argument in the VLOOKUP
function in cell B11 since the
values in column H are not in
ascending order
=VLOOKUP(MAX(H4:H7),H4:I7,
2,FALSE)
=MIN(B4:F
4)
Drag to H7
The Maximin Criterion - spreadsheet
48
To enable the spreadsheet to correctly
identify the optimal maximin decision in cell
B11, the labels for cells A4 through A7 are
copied into cells I4 through I7 (note that
column I in the spreadsheet is hidden).
I4
Cell I4
(hidden)=A4
Drag to I7
Uncertainty - The Minimax
Regret Criterion
The Minimax Regret Criterion
This criterion fits both a pessimistic and a
conservative decision maker approach.
The payoff table is based on lost
opportunity, or regret.
The decision maker incurs regret by failing
to choose the best decision.
49
Regret Criterion
The Minimax Regret Criterion
To find an optimal decision, for each state of nature:
Determine the best payoff over all decisions.
Calculate the regret for each decision alternative as the
difference between its payoff value and this best payoff value.
For each decision find the maximum regret over all
states of nature.
Select the decision alternative that has the minimum of
these maximum regrets.
50
TOM BROWN Regret Table
51
The Payoff Table
Decision Large rise Small rise No change Small fall Large fall
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
Let us build the Regret Table
The Regret Table
Gold 600 150 0 0 60
Bond 250 50 50 400 210
Stock 0 0 100 500 660
C/D 440 190 140 240 0
Investing in Stock generates no
regret when the market exhibits
a large rise
52
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
The Regret Table Maximum
Decision Large rise Small rise No change Small fall Large fall Regret
Gold 600 150 0 0 60 600
Bond 250 50 50 400 210 400
Stock 0 0 100 500 660 660
C/D 440 190 140 240 0 440
Gold 600 150 0 0 60 600
Bond 250 50 50 400 210 400
Stock 0 0 100 500 660 660
C/D 440 190 140 240 0 440
Investing in gold generates a
regret of 600 when the market
exhibits
a large rise
500 (-100) = 600
The Minimax Regret -
spreadsheet
53
=MAX(B$4:B$7
)-B4
Drag to F16
=VLOOKUP(MIN(H13:H16),H13:I16,2,
FALSE)
=MIN(H13:H
16)
=MAX(B14:F14
)
Drag to H18
Cell I13 (hidden)
=A13
Drag to I16
Risk
Each possible state of nature has an
assumed probability
States of nature are mutually exclusive
Probabilities must sum to 1
Determine the expected monetary value
(EMV) for each alternative
Regret Criterion
55
56
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
Let us build the Regret Table
The Regret Table
Gold 600 150 0 0 60
Bond 250 50 50 400 210
Stock 0 0 100 500 660
C/D 440 190 140 240 0
Investing in Stock generates no
regret when the market exhibits
a large rise
57
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
The Payoff Table
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
Gold 600 150 0 0 60 600
Bond 250 50 50 400 210 400
Stock 0 0 100 500 660 660
C/D 440 190 140 240 0 440
Gold 600 150 0 0 60 600
Bond 250 50 50 400 210 400
Stock 0 0 100 500 660 660
C/D 440 190 140 240 0 440
Investing in gold generates a
regret of 600 when the market
exhibits
a large rise
500 (-100) = 600
The Minimax Regret -
spreadsheet
58
=MAX(B$4:B$7
)-B4
Drag to F16
=VLOOKUP(MIN(H13:H16),H13:I16,2,
FALSE)
=MIN(H13:H
16)
=MAX(B14:F14
)
Drag to H18
Cell I13 (hidden)
=A13
Drag to I16
Uncertainty - The Maximax
Criterion
This criterion is based on the best possible scenario.
It fits both an optimistic and an aggressive decision
maker.

An optimistic decision maker believes that the best
possible outcome will always take place regardless of
the decision made.

An aggressive decision maker looks for the decision
with the highest payoff (when payoff is profit).
59
TOM BROWN - The Maximax Criterion
60
The Maximax Criterion Maximum
Decision Large rise Small rise No change Small fall Large fall Payoff
Gold -100 100 200 300 0 300
Bond 250 200 150 -100 -150 200
Stock 500 250 100 -200 -600 500
C/D 60 60 60 60 60 60
TOM BROWN - Insufficient Reason
Sum of Payoffs
Gold 600 Dollars
Bond 350 Dollars
Stock 50 Dollars
C/D 300 Dollars
Based on this criterion the optimal decision alternative
is to invest in gold.
61
62
Uncertainty Spreadsheet
template
Payoff Table
Large Rise Small Rise No Change Small Fall Large Fall
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D Account 60 60 60 60 60
d5
d6
d7
d8
Probability 0.2 0.3 0.3 0.1 0.1
Criteria Decision Payoff
Maximin C/D Account 60
Minimax Regret Bond 400
Maximax Stock 500
Insufficient Reason Gold 100
EV Bond 130
EVPI 141
RESULTS
63
The probability estimate for the
occurrence of
each state of nature (if available) can
be incorporated in the search for the
optimal decision.
For each decision calculate its
expected payoff.

Where probabilities of outcomes are available

Expected Monetary Value (EMV) uses the probabilities
to calculate the average payoff for each alternative

EMV (for alternative i) =
(probability of outcome) x (payoff of outcome)
Expected Monetary value(EMV):It is a value of
average pay off as follows

Here,the decision maker selects the course of action
which yields the optinal EMV.
Probability of payoff
EMV A V P V
V P V V P V V P V
i i
i
i
N N
( ( )
( ) ( ) ( )
) =
N
=

*
=
*
+
*
+ +
*
1
1 1 2 2
Number of states of
nature
Value of Payoff
Alternative i
...
Expected Opportunity Loss
EOL is the cost of not picking the best solution
EOL = Expected Regret
A
-
6
6
Expected Opportunity Loss (EOL)
How much regret do we expect based on the
probabilities?

EOL (for alternative i) =
(probability of outcome) x (regret of outcome)

B.Expected Opportunity loss(EOL):-This approach
is to minimize the expected opportunity loss. It is
the difference b/w the highest pay-off and actual
profit obtained for course of action taken.

Probability of payoff
EOL Pi
i
Lij
)
=
N
=

*
1
Number of states of
nature
Opportunity loss due to
state of nature and
course of action
EMV Example
1. EMV(A
1
) = (.5)($200,000) + (.5)(-$180,000) = $10,000
2. EMV(A
2
) = (.5)($100,000) + (.5)(-$20,000) = $40,000
3. EMV(A
3
) = (.5)($0) + (.5)($0) = $0
States of Nature
Favorable Unfavorable
Alternatives Market Market
Construct large plant (A1) $200,000 -$180,000
Construct small plant (A2) $100,000 -$20,000
Do nothing (A3) $0 $0
Probabilities .50 .50
Table A.3
EMV Example
1. EMV(A
1
) = (.5)($200,000) + (.5)(-$180,000) = $10,000
2. EMV(A
2
) = (.5)($100,000) + (.5)(-$20,000) = $40,000
3. EMV(A
3
) = (.5)($0) + (.5)($0) = $0
States of Nature
Best Option
Table A.3
Example - Expected Value
Render)
2001 by Prentice Hall, Inc., Upper Saddle River,
N.J . 07458
A
-
7
1
Suppose: Probability of favorable market = 0.5
Probability of unfavorable market = 0.5
States of Nature
Alternatives
Favorable
Market
Unfavorable
Market
Construct
large plant
$200,000 -$180,000
Construct
small plant
$100,000 -$20,000
$0 $0
Do
nothing
Expected
Value
$10,000
$40,000
$0
Decision is to Construct small plant.
1. EMV(A
1
) = (0.3)($200,000) + (0.7)(-$180,000) = -$66,000
2. EMV(A
2
) = (0.3)($100,000) + (0.7)(-$90,000) = -$33,000
3. EMV(A
3
) = (0.3)($0) + (0.7)($0) = $0
States of Nature
Probabilities 0.3 0.7
From Table A.3
If A3 is excluded, The preferable option is A2
Decision Making Under Risk (2)

In some cases the states of nature expected are certain, however the
values of each states are uncertain.
States of Demand
Seasonal Ticket Occasional
Alternatives Prob. Market Prob. Market
Sell 100 tickets early (A1) 0.7 $200,000 0.3 $50,000
Sell 100 tickets later (A2) 0.7 $150,000 0.3 $300,000
1. EMV(A
1
) = (0.7)($200,000) + (0.3)($50,000) = $155,000
2. EMV(A
2
) = (0.7)($150,000) + (0.3)($300,000) = $195,000
3. EMV(A
3
) = (0)($0) + (0)($0) = $0
The preferable option is A2
Case: Selling 100 tickets each time early or later in markets.
Expected Monetary Value
EMV (Alternative i) = (Payoff of 1
st
state of
nature) x (Probability of 1
st

state of nature)
+ (Payoff of 2
nd
state of
nature) x (Probability of 2
nd

state of nature)
++ (Payoff of last state of
nature) x (Probability of
last state of nature)
TOM BROWN - The Expected Value Criterion
75
The Expected Value Criterion Expected
Decision Large rise Small rise No change Small fall Large fall Value
Gold -100 100 200 300 0 100
Bond 250 200 150 -100 -150 130
Stock 500 250 100 -200 -600 125
C/D 60 60 60 60 60 60
Prior Prob. 0.2 0.3 0.3 0.1 0.1
EV = (0.2)(250) + (0.3)(200) + (0.3)(150) + (0.1)(-100) + (0.1)(-150) = 130
EOL Example
States of Nature


Table A.3
Computing EOL - The
Opportunity Loss Table
State of Nature
Alternative Favorable Market
($)
Unfavorable
Market ($)
Large Plant 200,000 - 200,000 0 - (-180,000)
Small Plant 200,000 - 100,000 0 -(-20,000)
Do Nothing 200,000 - 0 0-0
accompany Operations Management,
6E (Heizer & Render)
2001 by Prentice Hall, Inc., Upper
Saddle River, N.J . 07458
A-77
The Opportunity Loss Table -
continued
State of Nature
Alternative Favorable Market
($)
Unfavorable
Market ($)
Large Plant 0 180,000
Small Plant 100,000 20,000
Do Nothing 200,000 0
A-78
The Opportunity Loss Table -
continued
Alternative EOL
Large Plant (0.50)*$0 +
(0.50)*($180,000)
$90,000
Small Plant (0.50)*($100,000)
+ (0.50)(*$20,000)
$60,000
Do Nothing (0.50)*($200,000)
+ (0.50)*($0)
$100,000
A-79
Certainty
Is the cost of perfect information
worth it?
Determine the expected value of
perfect information (EVPI)

Perfect Information
Perfect Information would tell us with certainty which
outcome is going to occur
Having perfect information before making a decision
would allow choosing the best payoff for the outcome
Expected Value of
Perfect Information
EVPI is the difference between the payoff
under certainty and the payoff under risk
EVPI =
Expected value
with perfect
information
Maximum
EMV
Expected value with
perfect information
(EVwPI)
= (Best outcome or consequence for 1
st
state
of nature) x (Probability of 1
st
state of nature)
+ Best outcome for 2
nd
state of nature)
x (Probability of 2
nd
state of nature)
+ + Best outcome for last state of nature)
x (Probability of last state of nature)
6.4 Expected Value of Perfect
Information
The gain in expected return obtained from knowing
with certainty the future state of nature is called:
Expected Value of Perfect
Information (EVPI)

83
Information (EVPI)
EVPI places an upper bound on what one would pay
for additional information

EVPI is the expected value with perfect information
minus the maximum EMV
Render)
N.J . 07458
A
-
8
4
Expected Value With
Perfect Information (EVwPI)
The expected payoff of having perfect information
before making a decision

EVwPI = (probability of outcome)
x ( best payoff of outcome)
Expected Value of
Perfect Information (EVPI)
The amount by which perfect information would
increase our expected payoff
Provides an upper bound on what to pay for additional
information

EVPI = EVwPI EMV
EVwPI = Expected value with perfect information
EMV = the best EMV without perfect information

Alternatives
Demand
High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0
Probability 0.3 0.5 0.2
Max payoff for state of nature:-
1.High -----200000
2.Moderate-----100000
3.low----------0
Alternatives
Demand
High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0
Probability
0.3 0.5 0.2
Max payoff
for state of
nature
200000 100000 0
EVWPI 200000*o.3+100000*0.5+0*0.2=11
0000.
Altern
atives
Demand
High Moderate Low EMV
Large
plant
200,000*0
.3=60000
100,000*0.
5=50000
-120,000
*0.2=-24000
(60000+5
0000+(-
24000)=8
6000
Small
plant
90,000*0.
3
50,000*0.5 -20,000*0.2
48000
No
plant
0*0.3 0*0.5 0*0.2 o
Probability 0.3 0.5 0.2
Expected Value of Perfect Information
EVPI = EVwPI EMV
= $110,000 - $86,000 = $24,000

The perfect information increases the expected value
by $24,000
Would it be worth $30,000 to obtain this perfect
information for demand?

Example - EVUC
Render)
N.J . 07458
A
-
9
1
Best outcome for Favorable Market = $200,000
Best outcome for Unfavorable Market = $0
States of Nature
Alternatives
Favorable
Market
Unfavorable
Market
Construct
large plant
$200,000 -$180,000
Construct
small plant
$100,000 -$20,000
$0 $0
Do
nothing
Information
Render)
N.J . 07458
A
-
9
2
State of Nature
Alternative
Probabilities
Construct a
large plant
Construct a
small plant
Do nothing
200,000 -$180,000
$0
Favorable
Market ($)
Unfavorabl
e Market ($)
0.50 0.50
EMV
$40,000
$100,0
00
$20,00
0
$0 $0
$20,000
Information
EVPI = EVUC - max(EV)
= ($200,000*0.50 + 0*0.50) - $40,000
= $60,000

Thus, you should be willing to pay up to $60,000 to
learn whether the market will be favorable or not.
Render)
N.J . 07458
A
-
9
3
Information
= ($200,000*0.70 + 0*0.30) - $86,000
= $54,000

Now, you should be willing to pay up to $54,000 to
Render)
N.J . 07458
A
-
9
4
Now suppose: Probability of favorable market = 0.7
Information
EVPI = expected value with perfect information -
max(EMV)

= $200,000*0.50 + 0*0.50 - $40,000

= $60,000
Render)
N.J . 07458
A
-
9
5
Information
Render)
N.J . 07458
A
-
9
6
State of Nature
Alternative
Probabilities
Construct a
large plant
Construct a
small plant
Do nothing
200,000 -$180,000
$0
Favorable
Market ($)
Unfavorable
Market ($)
0.50 0.50
EMV
$40,000
$100,0
00
$20,00
0
$0 $0
$20,000
Information
= ($200,000*0.50 + 0*0.50) - $40,000
= $60,000

Thus, you should be willing to pay up to $60,000 to
Render)
N.J . 07458
A
-
9
7
Information
= ($200,000*0.70 + 0*0.30) - $86,000
= $54,000

Now, you should be willing to pay up to $54,000 to
Render)
N.J . 07458
A
-
9
8
Now suppose: Probability of favorable market = 0.7
Information
EVPI = expected value with perfect information -
max(EMV)

= $200,000*0.50 + 0*0.50 - $40,000

= $60,000
Render)
N.J . 07458
A
-
9
9
TOM BROWN - EVPI
100
The Expected Value of Perfect Information
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
Probab. 0.2 0.3 0.3 0.1 0.1
If it were known with certainty that there will be a Large Rise in the market
Large rise
... the optimal decision would be to invest in...
-100
250

500
60
Stock
Similarly,
TOM BROWN - EVPI
101
The Expected Value of Perfect Information
Gold -100 100 200 300 0
Bond 250 200 150 -100 -150
Stock 500 250 100 -200 -600
C/D 60 60 60 60 60
Probab. 0.2 0.3 0.3 0.1 0.1
-100
250

500
60
Expected Return with Perfect information =
ERPI = 0.2(500)+0.3(250)+0.3(200)+0.1(300)+0.1(60) = $271
Expected Return without additional information =
Expected Return of the EV criterion = $130
EVPI = ERPI - EREV = $271 - $130 = $141
Decision Trees
Graphical display of decision process
Used for solving problems
With 1 set of alternatives and states of nature,
decision tables can be used also
With several sets of alternatives and states of nature
(sequential decisions), decision tables cannot be
used
EMV is criterion most often used
Render)
N.J . 07458
A
-
1
0
2
Characteristics of a decision
tree
A Decision Tree is a chronological representation
of the decision process.
The tree is composed of nodes and branches.
103
A branch emanating from a
state of nature (chance) node
corresponds to a particular
state of nature, and includes
the probability of this state of
nature.
Decision
node
Chance
node
P(S
2
)
P(S
2
)
A branch emanating from
a decision node
corresponds to a
decision alternative. It
includes a cost or benefit
value.
Decision Trees
1. Define the problem
2. Structure or draw the decision tree
3. Assign probabilities to the states of
nature
4. Estimate payoffs for each possible
combination of decision alternatives and
states of nature
5. Solve the problem by working backward
through the tree computing the EMV for
each state-of-nature node
A
-
1
0
5
Process
Decision Theory
Terms:
Alternative: Course of action or choice.
State of nature: An occurrence over which the
decision maker has no control.

Symbols used in decision tree:
A decision node from which one of several
alternatives may be selected.
A state of nature node out of which one state of
nature will occur.
A
-
1
0
6
Decision Tree
A-107
1
2
State 1
State 2
State 1
State 2
Decision
Node
Outcome 1
Outcome 2
Outcome 3
Outcome 4
State of Nature Node
Decision Trees
Information in decision tables can be
displayed as decision trees
A decision tree is a graphic display of the
decision process that indicates decision
alternatives, states of nature and their
respective probabilities, and payoffs for
each combination of decision alternative
and state of nature
Appropriate for showing sequential
decisions
Decision Tree Example
Favorable market
Unfavorable market
Favorable market
Unfavorable market
Construct
small plant
A decision node A state of nature node
Figure A.1
A-110
A firm can build a large plant or small plant initially (for
a new product). Demand for the new product will be
high or low initially. The probability of high demand is
0.6. (The probability of low demand is 0.4.)
If they build small and demand is low, the payoff is
$40 million. If they build small and demand is high,
they can do nothing and payoff is $45 million, or they
can expand. If they expand, there is a 30% chance the
demand drops off and the payoff will be $35 million,
and a 70% chance the demand grows and the payoff is
$48 million.
If they build large and demand is high, the payoff is
$60 million. If they build large and demand is low,
they can do nothing and payoff is -$10 million, or they
can reduce prices and payoff is $20 million. Determine
the best decision(s) using a decision tree.

A-111
Three decisions:
1. Build Large or Small plant initially.
2. If build Small and demand is High, then Expand or
Do nothing.
3. If build Large and demand is Low, then decide to
Reduce prices or Do nothing.

Two states of nature:
1. Demand is High (0.6) or Low (0.4) initially.
2. If build Small, demand is High, and decision is
Expand, then demand Grows (0.7) or demand
Drops (0.3).
Decision Tree
A-112
Do nothing
Reduce prices
Demand grows (0.7)
Demand drops (0.3)
$48
$35
$45
$40
$60
$20
-$10
1
3
2
Decision Tree
A-113
Do nothing
Reduce prices
Demand grows (0.7)
Demand drops (0.3)
$48
$35
$45
$40
$60
$20
-$10
1
3
2
Decision Tree Solution
A-114
Work right to left (from end back to
beginning).

Start with Decision 3:
Reduce prices or Do nothing.

Choose Reduce prices (20 > -10).
Decision Tree
A-115
Do nothing
Reduce prices
Demand grows (0.7)
Demand drops (0.3)
$48
$35
$45
$40
$60
$20
-$10
1
3
2
$20
Decision Tree Solution
A-116
Consider Decision 2: Expand or Do
nothing.

To compare outcomes we need expected
value if we Expand: (48*0.7) + (35*0.3) =
44.1

Choose Do nothing (45 > 44.1).
Decision Tree
A-117
Do nothing
Reduce prices
Demand grows (0.7)
Demand drops (0.3)
$48
$35
$45
$40
$60
$20
-$10
1
3
2
$44.1
$45
$20
$45
Decision Tree
A-118
Do nothing
Reduce prices
Demand grows (0.7)
Demand drops (0.3)
$48
$35
$45
$40
$60
$20
-$10
1
3
2
$44.1
$45
$20
$45
$44
$43
Decision Tree Final Solution
A-119
Decisions:

1. Build Large.

2. If demand is Low, then Reduce
prices.

Expected payoff = $44 million.
Render)
N.J . 07458
A
-
1
2
0
Advantages
Render)
N.J . 07458
A
-
1
2
1
Disadvantages
= (.5)($200,000) + (.5)(-$180,000)
EMV for node 1
= $10,000
EMV for node 2
= $40,000
= (.5)($100,000) + (.5)(-$20,000)
Payoffs
$200,000
-$180,000
$100,000
-$20,000
$0
Construct
small plant
Favorable market (.5)
Unfavorable market (.5)
1
2
Figure A.2
Decision Table Example
State of Nature
Alternatives Favorable Market Unfavorable Market
Construct large plant $200,000 $180,000
Construct small plant $100,000 $ 20,000
Do nothing $ 0 $ 0
Table A.1
= (.5)($200,000) + (.5)(-$180,000)
EMV for node 1
= $10,000
EMV for node 2
= $40,000
= (.5)($100,000) + (.5)(-$20,000)
Payoffs
$200,000
-$180,000
$100,000
-$20,000
$0
Construct
small plant
1
2
Figure A.2
In-Class Exercise
Two suppliers deliver a product to us. Supplier 1
charges $ 575 for the part. The part seldom fails
but its probability of failing is 0.1. The amount
that we loose if part fails is $100.
Supplier 2 charges $ 550 for the same part. The
probability of having a good part for this
supplier is 0.8. The amount that we loose if a
part fails is $460.
Find the expected monetary value of defective
part from supplier 1?
Find the expected monetary value of defective
part from supplier 2?
Find the total expected monetary value for each
supplier?
BGD plans to do a commercial development on a
property.
Relevant data
Asking price for the property is 300,000 dollars.
Construction cost is 500,000 dollars.
Selling price is approximated at 950,000 dollars.
Variance application costs 30,000 dollars in fees and expenses
There is only 40% chance that the variance will be approved.
If BGD purchases the property and the variance is denied, the
property can be sold for a net return of 260,000 dollars.
A three month option on the property costs 20,000 dollars,
which will allow BGD to apply for the variance.
126
BILL GALLEN DEVELOPMENT
COMPANY
BILL GALLEN DEVELOPMENT COMPANY
A consultant can be hired for 5000 dollars.
The consultant will provide an opinion about the
approval of the application
P (Consultant predicts approval | approval granted) = 0.70
P (Consultant predicts denial | approval denied) = 0.80
BGD wishes to determine the optimal strategy
Hire/ not hire the consultant now,
Other decisions that follow sequentially.
127
Construction of the Decision Tree
Initially the company faces a decision about hiring the
consultant.

After this decision is made more decisions follow regarding
Application for the variance.
Purchasing the option.
Purchasing the property.
128
BILL GALLEN - Solution
BILL GALLEN - The Decision Tree

129
Buy land
-300,000
Apply for variance
Apply for variance
-30,000
-30,000
0
3

130
12
-300,000 -500,000 950,000
Buy land
Build Sell
-50,000
100,000
-70,000
260,000
Sell
Build Sell
950,000 -500,000
120,000
Buy land and
apply for
variance
-300000 30000 +260000 =
-300000 30000 500000 +950000 =
Purchase option
and
apply for
variance
131
This is where we are at this stage
Let us consider the decision to hire a consultant
BILL GALLEN
The Decision Tree
132
-5000
Apply for variance
Apply for variance
Apply for variance
Apply for variance
-5000
-30,000
-30,000
-30,000
-30,000
Let us consider
the decision to
hire a consultant
Done
Buy land
-300,000
Buy land
-300,000

133
?
?
Build Sell
950,000 -500,000
260,000
Sell
-75,000
115,000

134
?
?
Build Sell
950,000 -500,000
260,000
Sell
-75,000
115,000
The consultant serves as a source for additional
information
about denial or approval of the variance.

135
?
?
Build Sell
950,000 -500,000
260,000
Sell
-75,000
115,000
Therefore, at this point we need to calculate the
posterior probabilities for the approval and denial
of the variance application

136
22
Build Sell
950,000 -500,000
260,000
Sell
-75,000
27
25
115,000
23 24
26
The rest of the Decision Tree is built in a
similar manner.
Posterior Probability of (approval | consultant predicts approval) = 0.70
Posterior Probability of (denial | consultant predicts approval) = 0.30
?
?
.7
.3
The Decision Tree
Determining the Optimal Strategy
Work backward from the end of each branch.

At a state of nature node, calculate the expected
value of the node.

At a decision node, the branch that has the
highest ending node value represents the
optimal decision.
137
138
22
27
25
23 24
26
-75,000
115,000
115,000
-75,000
115,000
-75,000
115,000
-75,000
115,000
-75,000 22
115,000
-75,000
58,000
?
?
0.30
0.70
Build Sell
950,000 -500,000
260,000
Sell
-75,000
115,000
With 58,000 as the chance node value,
we continue backward to evaluate
the previous nodes.
139
$10,000
$58,000
$-5,000
$20,000
$20,000
Buy land; Apply
for variance
Build,
Sell
Sell
land
$-75,000
$115,000

THANK YOU

Decision Theory PPT (Mba3) 2011

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Decision Theory PPT (Mba3) 2011

Uploaded by

Copyright:

Available Formats

2008 Prentice Hall, Inc.

You might also like