Decision Analysis: Chapter 3

Decision Analysis: Chapter 3

Q: What do you mean by Decision Analysis and what are the types of probability?
Decision Analysis: Decision analysis is a set of quantitative decision making
techniques for decision situations in which uncertainty exists.

Now, uncertainty can be classified into two ways/ types:

1. Subjective Probability: Subjective probability is the degree of belief

to occurrence of the event.
2. Objective Probability: Objective probability is the probability which
can be derived either based on historical occurrences or based on experimentation.
Alternatively can be derived from statistical formula.
Decision Analysis: Chapter 3
Q: What is consistancy requirement?

Consistency requirement: If the probability of an event A is 0.65, then the

probability of event B must be 0.35.

i.e. P(A) + P(B) = 1

Mathematically, if A, B Є E
Then
A, B  E
A B = φ

P(A) + P(B) = 1, which is called Consistency requirements.

Decision Analysis: Chapter 3
Q: Discuss about the elements of Decision Problem.

• # Elements of Decision Problems: A decision problem is usually viewed as

having four common elements
1. The alternative course of action: The alternative course of action involves two
or more options or alternative course of action. One and only one of these
alternatives must be selected.
2. The states of nature: The state of nature are factors that affect the outcome of
a decision but are beyond control of the decision maker, such as rain, inflation,
political development etc.
3. Payoff table: A payoff table is the combination for each possible combination of
alternative course of action and state of nature.
4. Uncertainty: The decision maker is uncertain about what state of nature will
occur. However choose the criterion that results in the largest payoff.
Decision Analysis: Chapter 3
Q: What are the different types of decision making environment ?
• #Types of Decision –Making Environment:
The types of decisions people make depend on how much knowledge or
information they have about the situation. Three decision making environments
are defined and explained as follows:

Type 1: Decision Making Under Certainty: In the environment of decision

making under certainty, decision makers know with certainty the consequence of
every alternative that will maximize their- well – being or will result in the best
outcome. Let’s say that you have $ 1000 to invest for a one year period. One
alternative is to open a savings account paying 6% interest and another is to invest
in a government treasury bond paying 10% interest. Both investment are secure
and guaranteed, but as treasury bond will pay a higher return, you may choose that
one.
Decision Analysis: Chapter 3
• #Types of Decision –Making Environment:
• Type 2: Decision Making Under Risk: In decision making under risk, the decision maker
knows the probability of occurrence of each outcome. For example, that the probability
of being dealt a club is 0.25. The probability of rolling a 5 on a die is 1/6. In decision
making under risk, the decision maker attempts to maximize his or her expected well-
being. Decision theory models for business problems in this environment typically
employ two equivalent criteria: maximization of expected monetary value and
minimization of expected loss.

Three Criteria are there under Risk:

1. EMV ( Expected Monetary Value)
2. EOL ( Expected Opportunity Loss)
3. EVPI (Expected Value With Perfect Information)
Decision Analysis: Chapter 3
• #Types of Decision –Making Environment:
Type 3: Decision Making Under Uncertainty: In decision making under uncertainty the
decision maker does not know the probabilities of the various outcomes. As an example,
the probability that a BNP personnel will be president of Bangladesh 25 years from now is
not known. Sometimes it is impossible to assess the probability of success of a new
undertaking or product

Five Criteria are there under Risk:

1. Maximax or Optimistic Criterion: Under this the decision maker finds the
maximum possible payoff for each alternative and then chooses the alternative with maximum
payoff within this group.
2. Maximin or Pessimistic Criterion: To use this criterion the decision maker finds
the minimum possible payoff for each alternative and then chooses the alternative with maximum
payoff within this group.
Decision Analysis: Chapter 3
• #Types of Decision –Making Environment:
3. Minimax or Regret Criterion : The decision maker tries to minimize the regret
before actually selecting a particular alternative. For this he determines the maximum regret
amount for each alternative and then choose the alternative with the minimum of the above
maximum regrets.
4. Hurwicz Criterion: Also called the weighted average criterion. It is a compromise
between the maximax and maximin decision criteria. It takes both of them into account by assigning
them weights in accordance with the degree of optimism or pessimism.
Select α = Index of optimism, If α = 0 pessimistic, then α = 1 optimistic.
Hence α is specified (0,1) range.
Also α = 0.5 implies neither optimistic nor pessimistic.
5. Laplace or Equally likely Probability: It is based on what is known as the principle
of insufficient reason. Because of the probability distribution of the states of nature is not known,
the criterion assigns equal probabilities toa ll the events of each alternative and select the
alternative associated with the maximum expected payoff.
 Decision Making Under Risk
EMV (Expected Monetary Value ( calculation)
Example : 1
Alternative Course of Action

Demand Alternative A 1 Alternative A 2 Probability

(Introduce Break Fast) (Not Introduce Break Fast)
State of Nature

Strong (S1) 15 0 0.20

Medium (S2) 5 0 0.40
Poor (S3) -10 0 0.40
EMV 1.00

EMV (A1) = 0.20 (15) + 0.40 (5) + 0.40 (-10) = 3 + 2 – 4 = $1

EMV (A2) = 0.20 (0) + 0. 40 (0) + 0.40 (0) = $0
Since EMV (A1) is greater than than EMV (A2). We should go for introduce break fast.
 Decision Making Under Risk
EMV (Expected Monetary Value ( calculation)
Example : 2 A newspaper boy has the following probabilities of selling a magazine.
No. of copies Sold Probabilities
10 0.10
11 0.15
12 0.20
13 0.25
14 0.30
Cost of a copy is 30 Tk, sale price is 50 Tk. He cannot return unsold copies. How
many copies should he order?

Solution: Given, Selling Price = 50 Tk/ unit

Cost = 30 Tk / unit
Profit = 20 Tk/ unit
Unsold news paper cannot be return back.
 Given, Selling Price = 50 Tk/ unit
Cost = 30 Tk / unit
Profit = 20 Tk/ unit (Unsold news paper cannot be return back. )
Alternative Course of Action
Demand Probability 10 11 12 13 14
(No. of Copies
Sold) Conditional Pay off:
10 0.10 200 170 140 110 80
20.S If D ≥ S
11 0.15 200 220 190 160 130
State of Nature

12 0.20 200 220 240 210 180 50. D – 30.S If D < S

13 0.25 200 220 240 260 230
14 0.30 200 220 240 260 280
Since the maximum EMV is
EMV 1.00 200 215 222. 220 205 222.5. News paper boy
5 should stock 12 nos of
newspaper daily

EMV (10 ) = .10 (200) + .15 (200 ) + .20 (200) + .25 (200) + .30 (200) = 200
EMV (11) = .10 (170) + .15 (220 ) + .20 (220) + .25 (220) + .30 (220) = 215
EMV (12) = .10 (140) + .15 (190 ) + .20 (240) + .25 (240) + .30 (240) = 222.5
EMV (13) = 220
Exercise 1: Page 33

An Ice cream retailer buy ice cream at a cost of 5 Tk. Per cup and sales 8 Tk. Per cup.
Any unsold at the end of the day can be disposed at a salvage value of 2 Tk. Per cup. Past
sales have ranged between 15 and 18 cup per day. Find EMV, EOL and EVPI if the sales
history has the following probabilities.

Market Size: 15 16 17 18
Probabilities: 0.10 0.20 0.40 0.30

• Solution:
• Given Selling price = Tk. 8 /cup
Cost = Tk. 5/ cup
Profit = Tk. 3/ cup
Salvage Value = Tk. 2/ cup
 Table : 1 : Expected Monetary Value
Given
Alternative Course of Action
Market Probability Possible Stock Action Conditional Payoff:
Size
(Demand)
3 (S) D≥S
15 16 17 18 8 D – 5 S + 2 ( S –D) D<S
15 0.10 45 42 39 36
State of Nature

16 0.20 45 48 45 42
17 0.40 45 48 51 48 Example:
For Stock:18, Demand (Sell): 16
18 0.30 45 48 51 54
8 (16) – 5 (18) +2 (18-16) = 42
EMV 1.00 45 47.4 48.6 47.4
 Alternative Course of Action
Market Probability Possible Stock Action
Size
15 16 17 18 Example:
For Stock:18, Demand (Sell): 16
15 0.10 45 42 39 36
State of Nature

16 0.20 45 48 45 42 8 (15) – 5 (16) +2 (16-15) = 42

17 0.40 45 48 51 48
8 (15) – 5 (17) +2 (17-15) = 39
18 0.30 45 48 51 54
EMV 1.00 45 47.4 48.6 47.4
 Table : 1 : Expected Monetary Value
Given
Alternative Course of Action
Market Probability Possible Stock Action Selling price = Tk. 8 /cup
Size Cost = Tk. 5/ cup
15 16 17 18 Profit = Tk. 3/ cup
15 0.10 45 42 39 36
Salvage Value = Tk. 2/ cup
16 0.20 45 48 45 42
State of Nature

17 0.40 45 48 51 48 Conditional Payoff:

18 0.30 45 48 51 54
3 .S if D≥S
EMV 1.00 45 47.4 48.6 47.4 8D – 5S + 2 (S – D) if D<S

Since highest EMV is 48.6, So, the ice cream retailer should stock 17 cups of ice cream.
 Expected Opportunity Loss
means , loss of not choosing the
Table : 2 : Expected Opportunity Loss best Alternative.
Alternative Course of Action
Formula For determine EOL
Market Probability Possible Stock Action Table:
Size
15 16 17 18 i
*

15 0.10
16 0.20 ? Where, L = Loss Table
State of Nature

M = Monetary Table
17 0.40 i = row, j= column
18 0.30
Now, Lij = L 34 means position of
loss table, row 3 and column 4

Now i
*

L 34 = M34 – M3*

L 34 = 48– 51* = 3
 Table : 1 : Expected Monetary Value Table : 2 : Expected Opportunity Loss
Alternative Course of Action Alternative Course of Action
Market Probability Possible Stock Action Market Probabilit Possible Stock Action
Size Size y
15 16 17 18 15 16 17 18
15 0.10 45 42 39 36 15 0.10 0 3 6 9
State of Nature

State of Nature
16 0.20 45 48 45 42 16 0.20 3 0 3 6
17 0.40 45 48 51 48 17 0.40 6 3 0 3
18 0.30 45 48 51 54
18 0.30 9 6 3 0
EMV 1.00 45 47.4 48.6 47.4 EOL 1.00 5.7 3.3 2.1 3.3

Now i
*
, L 34 = M34 – M3*, L 34 = 48–51= 3
 Table : 1 : Expected Monetary Value Table : 2 : Expected Opportunity Loss
Alternative Course of Action Alternative Course of Action
Market Probability Possible Stock Action Market Probabilit Possible Stock Action
Size Size y
15 16 17 18 15 16 17 18
15 0.10 45 42 39 36 15 0.10 0 3 6 9
State of Nature

State of Nature
16 0.20 45 48 45 42 16 0.20 3 0 3 6
17 0.40 45 48 51 48 17 0.40 6 3 0 3
18 0.30 45 48 51 54
18 0.30 9 6 3 0
EMV 1.00 45 47.4 48.6 47.4 EOL 5.7 3.3 2.1 3.3

EVPI (Expected Value with Perfect Information)

= EPPI – Max EMV, Where EPPI= Expected profit with perfect Information.Since the minimum loss 2.1 is
= Σ Mi* (Psi) – 48.6 occurred at 17 cups. So, the ice
= 45 (0.10) + 48 (0.20) + 51 (0.40) + 54(0.30) – 48.6 cream retailer should stock 17
= 2.1 cups of ice cream.
 Exercise 2: Page 33

A TV dealer finds that the cost of a TV in stock for a week is Tk. 30. and cost of a
unit shortage is Tk. 70. For one particular model of the probability distribution of
weekly sales is as follows.
Weekly Sales: 0 1 2 3 4 5 6
Probabilities: 0.10 0.10 0.20 0.25 0.15 0.15 0.05
How many units per week should the dealer order? Also determine
EVPI

• Solution:
• Given Holding cost = Tk. 30 /week
Shortage cost = Tk. 70/ unit
 Table : 1 Cost Table
Alternative Course of Action
Weekly Probability Possible Stock Action
Sales

0 1 2 3 4 5 6
0 0.10 0 30 60 90 120 150 180
1 0.10 70 30 60 90 120 150 180
State of Nature

2 0.20 140 100 60 90 120 150 180

3 0.25 210 170 130 90 120 150 180
4 0.15 280 240 200 160 120 150 180
5 0.15 350 310 270 230 190 150 180
6 0.05 420 380 340 300 260 220 180
Expected 1.00 203
Cost
 Table : 2 Cost Table
Alternative Course of Action
Weekly Probability Possible Stock Action
Sales
0 1 2 3 4 5 6
0 0.10 0 30 60 90 120 150 180
1 0.10 70 30 60 90 120 150 180
2 0.20 140 100 60 90 120 150 180
3 0.25 210 170 130 90 120 150 180
State of Nature

4 0.15 280 240 200 160 120 150 180

5 0.15 350 310 270 230 190 150 180
6 0.05 420 380 340 300 260 220 180
Expected 1.00 203 170 144 132 137.50 153.50 180
Cost
Since the minimum expected cost is 132. The TV dealer should stock 3 numbers of TV
weekly.
EVPI = Mini. Expected Cost - ECPI , Where ECPI= Expected cost with perfect Information.
= 132 - {0(0.10) +30 (0.10) +60 (0.20) +90 (0.25) + 120 (0.15) +150 (0.15) +180 (0.05)}
= 132- 87 = 45
 Decision Making Under Uncertainty

Five Criteria are there under Risk:

1. Maximax or Optimistic Criterion:
2. Maximin or Pessimistic Criterion:
3. Minimax or Regret Criterion :
4. Hurwicz Criterion:
5. Laplace or Equally likely Probability:
 Decision Making Under Uncertainty

Example 6: Page 35
The following matrix gives the payoff of different strategies (alternatives) A1,A2, A3
against conditions (events) N1, N2, N3 & N4 .
N1 N2 N3 N4
A1 Rs. 4000 Rs. –100 Rs. 6000 Rs. 18000
A2 20000 5000 400 0
A3 20000 15000 -2000 1000

Indicate the decision taken under the following approach:

a. Optimistic b. Pessimistic c. Equally likely Probability
d. Regret e. Hurwicq Criterion, degree of optimism being 0.7
Solution: State Of Nature
N1 N2 N3 N4

Alternative Course of
A1 Rs. 4000 Rs. –100 Rs. 6000 Rs. 18000

A2 20000 5000 400 0

Action
A3 20000 15000 -2000 1000

Table 1
Maximum Minimum Equllay Likely Probability
A1 18000 -100 ¼(4000 -100+6000+18000)= 6975

A2 20000 0 ¼(20000+5000+400+0)= 6350

A3 20000 -2000 ¼(20000+15000-2000+1000) = 8500

a. Under Maximax or Optimistic criteria A2 and A3 is the optimal act.

b. Under Maximin or Pessimistic criteria A2 is the optimal act
c. Under laplace or Equally likely Probability A3 is the optimal act.
Solution: State Of Nature
N1 N2 N3 N4

Alternative Course of
A1 Rs. 4000 Rs. –100 Rs. 6000 Rs. 18000

A2 20000 5000 400 0

Action
A3 20000 15000 -2000 1000

Table 2: Regret Table

N1 N2 N3 N4 Maximum
A1 16000 15100 0 0 16000

A2 0 10000 5600 18000 18000

A3 0 0 8000 17000 17000

d. Under Minimax or Regret criteria A1 is the optimal act.

Solution: State Of Nature
N1 N2 N3 N4

Alternative Course of
A1 Rs. 4000 Rs. –100 Rs. 6000 Rs. 18000

A2 20000 5000 400 0

Action
A3 20000 15000 -2000 1000

Table 3 ( α= Degree of realism, given α = 0.7)

Maximum Minimum Pay off = α (Maximum) + (1-α) (Minimum)
A1 18000 -100 0.7 (18000) + 0.3 (-100) =12570

A2 20000 0 0.7 (20000) + 0.3 (0) = 14000

A3 20000 -2000 0.7(20000) + 0.3 (-2000) = 13400

e. Under Hurwica criteria A2 is the optimal act.

Exercise: 9 (page 38)

The Dynamax Company is going to introduce one of three new products: a widget, a hummer,
or a nimnot. The market conditions (favorable, stable, or unfavorable) will determine the
profit or loss the company realizes, as shown in the following payoff table.

Market Conditions
Favorable Stable Unfavorable
Product 0.2 0.5 0.3
Widget $160,000 $90,000 $50,000
Hummer 70,000 40.000 20,000
Nimnot 45,000 35,000 30.000

a. Compute the expected value for each decision and select the best one.
b. Determine how much the firm would be willing to pay to a market research firm to gain better
information about future market conditions.
c. Assume that probabilities cannot be assigned to future market conditions, and determine the
best decision using the maximax, maximin, minimax regret, and equal likelihood criteria
Solution: Market Conditions
Favorable Stable Unfavorable
Product 0.2 0.5 0.3
Widget $160,000 $90,000 $50,000
Hummer 70,000 40.000 20,000
Nimnot 45,000 35,000 30,000

a . Market Probability Widget Product

Hummer Nimnot
Condition

Favorable 0.2 160000 70000 45000

Stable 0.5 90000 40000 35000

Unfavorable 0.3 50000 20000 30000

EMV 1.00 92000 40000 35500

Since the maximum EMV is 92000. Hence alternative Widget is the best selection.
b. The company is willing to pay $92000.
Solution: Market Conditions
Favorable Stable Unfavorable

Widget $160,000 $90,000 $50,000

Hummer 70,000 40.000 20,000
Nimnot 45,000 35,000 30,000

c . Product Maximum Minimum Equally Likely Probability

Widget 160000 50000 (160000 + 90000 +50000)/ 3 = 100000

Hummer 70000 20000 (70000 + 40000 + 20000)/3 = 43333.34

Nimnot 45000 30000 (70000 + 35000 + 30000)/3 = 45000

i. Under Maximax criteria Widget is the optimal act.

ii. Under Maximin criteria Widget is the optimal act.
iii. Under laplace or equally likely probability Widget is the optimal act.
Solution: Market Conditions
Favorable Stable Unfavorable
Widget $160,000 $90,000 $50,000
Hummer 70,000 40,000 20,000
Nimnot 45,000 35,000 30,000
Table: 2 (Regret Table)
Product Favorable Stable Unfavorab Maximum
le loss

Widget 0 0 0 0

Hummer 90000 50000 30000 90000

Nimnot 115000 55000 20000 115000

iv. Under Minimax or regret criteria Widget is the optimal act.

Thank You