LECTURE 13

Contents of the Lecture

• Introduction.
• Types of decision making environment.
• Decision making under uncertainty.
• Decision making under risk.

2
 Introduction
• Decision theory is both descriptive and prescriptive
business modelling approach to classify the degree of
knowledge and compared expected outcomes due to
several courses of action.

• The degree of knowledge is divided into four

categories: certainty, ignorance, risk, and uncertainty.

Ignorance Uncertainty Risk Certainty

3
 Continued…

• Decision alternatives: There is a finite number of

decision alternatives available to the decision maker
at each point in time when a decision is made.

• States of nature: An event or scenario that is not

under the control of decision makers. Example: socio-
economic-political factors.

• Payoff: A numerical value obtained due to

application of each possible combination of decision
alternatives and states of nature.
4
 Steps of Decision Making

St • Identify and define the problem.

ep
1
St • List all possible future events.
ep
2
St • Identify all courses of action.
ep
3
St • Express the payoff matrix.
ep
4
St • Apply appropriate decision theory model.
ep
5
5
 Example
A firm manufactures three types of products.
Fixed Cost Variable Cost per unit
Product A 25000 12
Product B 35000 9
Product C 53000 7

The likely demand (units) of the products is given below:

Demand
Poor (D1) 3000
Moderate (D2) 7000
High (D3) 11000

If the sale price of each type of product is Rs. 25,

then prepare the payoff matrix.
6
 Solution

Product Alternative Demands (in ‘000)

Type
D1 D2 D3

A (3*25) – (25) – (7*25) – (25) – (11*25) – (25) –

(3*12) = 14 (7*12) = 66 (11*12) = 118
B 13 77 141

C 1 73 145

7
Types of Decision Making Environment
1. Decision Making under Certainty: here the DM
has complete knowledge of the outcome due to each
decision alternatives.

2. Decision Making under Risk: DM do not have

perfect knowledge about possible outcomes. Makes
use of probability of occurrences.

3. Decision Making under Uncertainty: here the DM

is unable to specify the probability for occurrence of
particular decision alternative.
8
Optimis Pessimi
m sm

DM
under
Uncerta
inty

Hurwic
z

9
 Optimism Criterion

• Also called maximax or minimin criterion.

• The DM ensures that he achieve largest possible
profit or lowest possible cost.

• Working:
(i) Locate the max (min) payoff value corresponding to
each decision alternative.
(ii) Select the decision alternative with best payoff
value.

10
 Pessimism Criterion

• Also called maximin or minimax criterion.

• DM ensures that he would earn no less or pay no
more than some specified amount.

• Working:
(i) Locate the min (max) payoff value in case of loss
(profit) corresponding to each decision alternative.
(ii) Select the decision alternative with best payoff
value.

11
 Laplace Criterion
• It assumes that all states of nature occur with equal
probability i.e. each state of nature is assigned an
equal probability.

• Working:
(i) Assign equal probability value (1/n) to each state of
nature.
(ii) Compute expected payoff for each decision
alternative.
(iii) Select the best expected payoff value.

12
 Hurwicz Criterion
• It assumes that the DM should neither be completely
optimistic nor pessimistic.
• Defined a coefficient of optimism () that lies between 0
& 1, where 0 is complete pessimistic and 1 is complete
optimistic.
• Calculate criterion for realism:

• Working:
(i) Decide and then calculate H.
(ii) Select the alternative with best weighted payoff value.
13
 Regret Criterion
• Also called savage criterion.
• Here the DM regrets for choosing wrong decision
alternative resulting in an opportunity loss of payoff.

• Working:
(i) Find the best payoff corresponding to each state of
nature and subtract each element in the row from it.
Form a regret matrix.
(ii) Identify the worst (maximum regret) payoff value.
(iii) Select the decision alternative resulting in smallest
opportunity loss value.
14
 Example 1
A food products’ company is contemplating the
introduction of a revolutionary new product with new
packaging or replacing the existing product at much
higher price (S1). It may even make a moderate change
in the composition of the existing product, with a new
packaging at a small increase in price (S2), or may mall
a small change in the composition of existing product,
backing it with a negligible increase (S3). The 3
possible states of nature are high increase in sales (N1),
no change in sales (N2), and decrease in sales (N3). The
marketing department of the company worked out the
payoffs in terms of yearly net profits of each strategies,
as provided below:
15
 Continued…

Which strategy should the concerned executive choose

on the basis of: (a) Optimism, (b) Pessimism, (c)
Laplace, (d) Hurwicz, and (e) Regret.

Strategies States of Nature

N1 N2 N3
S1 700000 300000 150000
S2 500000 450000 0
S3 300000 300000 300000

16
 Solution
• Optimism Criterion

States of Strategies
nature S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000
N3 150000 0 300000
Column 700000 500000 300000
(max)

17
 Continued…
• Pessimism Criterion

States of Strategies
nature S1 S2 S3

N1 700000 500000 300000

N2 300000 450000 300000
N3 150000 0 300000
Column 150000 0 300000
(min)

18
 Continued…
• Laplace Criterion

Strategy Expected Return

S1 (700000+300000+150000)/3 = 383333.33

S2 (500000+450000+0)/3 = 316666.66

S3 (300000+300000+300000)/3 = 300000

19
 Continued…
• Regret Criterion
States of Strategies
nature
S1 S2 S3

N1 700000- 700000- 700000-

700000= 0 500000= 300000=
200000 400000
N2 450000- 450000- 450000-
300000= 450000= 0 300000=
150000 150000
N3 150000 300000 0
Column (max) 150000 300000 400000
20
 Continued…
• Hurwicz Criterion (let = 0.60)

Strategy Col Max Col Min H

S1 700000 150000 (0.6*700000)+(0.4*

150000) = 480000

S2 500000 0 (0.6*500000) + 0 =
300000

S3 300000 300000 300000

21
 Example 2
A manufacturer manufactures a product of which the
principal ingredient is chemical X. At the moment, the
manufacturer spends Rs 1000 per year on supply of X,
but there is a possibility that the price may soon
increase to four times the present figure. There is
another chemical Y, which the manufacturer can use
along with chemical Z, to get same effect as X. together
they cost Rs 3000 per year, but their price is unlikely to
rise. What action should the manufacturer take? Apply
the minimax and maximin criteria for decision making.
If the coefficient of optimism is 0.4, then find the course
of action that minimizes cost.
22
 Solution
• The trade-off matrix can represent “profit” earned
under each strategy.

States of Nature Courses of Action

Use Y & Z (S1) Use X (S2)
Price of X increases -3000 -4000
(N1)
Price of X does not -3000 -1000
increase (N2)

Do yourself !!!
23
 Example 3
An investor is given the following investment
alternatives and percentage rates of return. Over the past
300 days, 150 days have been medium market
conditions and 60 days have had high market increases.
On the basis of these data, state the optimum investment
strategy for the investment.

Market Conditions
Low Medium High
Regular shares 7% 10% 15%
Risky shares -10% 12% 25%
Property -12% 18% 30%
24
 Solution

Market probability Regular Risky Property

conditions shares shares

Low 0.30 0.07*0.30 0.1*0.3 0.15*0.3

Medium 0.50 -0.1*0.5 0.12*0.5 0.25*0.5

High 0.20 -0.12*0.2 0.18*0.2 0.30*0.2

Expected return 0.136 0.126 0.230

25
 Expected
Monetary
Value (EMV)

Expected
Value of RISK Expected
Perfect Opportunity
Information Loss (EOL)
(EVPI)

26
 Expected Monetary Value (EMV)

• Construct a payoff matrix listing all possible states of

nature and decision alternatives.

• Enter the conditional payoff values along with the

associated probability of occurrence.

• Calculate EMV for each alternative by multiplying

conditional payoffs with associated probabilities in
order to get weighted values for alternatives.

• Select the course of action that yields optimal EMV.

27
 Example 1

Mr X flies quite often fro town A to B. he can use

airport bus that costs Rs 25 but if he takes it, there is
o.08 chance that he will miss the flight. The stay in the
hotel costs Rs 270 with a 0.96 chance of being on time
for flight. For Rs 350 he can use a taxi which will make
99 % chance of being on time for flight. If X catches the
plane on time, he will conclude a business transaction
that will produce a profit of Rs 10000, otherwise, he
will lose it. Use EMV to select the mode of transport
that Mr X must use.

28
 Solution
States Bus Hotel Taxi
of
Nature Cost Prob. Cost Prob. Cost Prob.

Catches 10000-25 0.92 10000-270 0.96 10000-350 0.99

flight = 9975 = 9730 = 9650
Misses -25 0.08 -270 0.04 -350 0.01
flight
States of Expected Value
Nature Bus Hotel Taxi

Catches flight 9177 9340.80 9553.50

Misses flight -2.0 -10.80 -3.50
EMV 9175 9330 9550
29
 Example 2
The manager of a flower shop promises its customers
delivery within four hours on all flower orders. All
flowers are purchased on the previous day and delivered
by 8 am the next morning. The daily demand for roses
is provided below. Also, the manager purchases roses
for Rs 10 per dozen and sells them for Rs 30. all unsold
roses are donated to the hospital. How many dozens of
roses should be ordered so as to gain profit?

Dozen of roses 70 80 90 100

Prob. 0.1 0.2 0.4 0.3

30
 Solution

• Let the quantity of roses to be purchased each day be

denoted as the courses of action (alternatives) and the
daily demand be States of Nature.

Where D is the number of roses sold per day and S is

the number of roses stocked.

31
 Continued…
States of Prob. Conditional Profit (Rs)
nature 70 80 90 100

70 0.1 140 130 120 110

80 0.2 140 160 150 140
90 0.4 140 160 180 170
100 0.3 140 160 180 200

States of Expected Payoff (Rs)

nature 70 80 90 100

70 14 13 12 11
80 28 32 30 28
90 56 64 72 68
100 42 48 54 60
EMV 140 157 168 167
32
 Expected Opportunity Loss (EOL)
• Prepare a conditional payoff matrix for each
combination of states of nature and course of action
with associated probabilities.
• For each state of nature, calculate conditional
opportunity loss (COL) values by subtracting each
payoff from the max.
• Calculate EOL for each alternative by multiplying
probabilities with COL.
• Select the course of action for which EOL is min.

33
 Example 1
A company manufactures goods for a market in which
the technology of the product is changing rapidly. The
research and development department has produced a
new product that appears to have potential for
commercial exploitation. A further Rs 60000 is required
for development testing. The company has 100
customers and each customer might purchase at most
one unit. Market research suggests that SP of Rs 6000
for each unit with total variable cost of manufacturing
and selling estimate as Rs 2000 for each unit.
PTO…

34
 Continued…

From previous experience it has been possible to derive

a probability distribution relating to proportion of
customers who will buy the product. Using EOL,
determine whether or not the company should develop
the product.

Proportion of 0.04 0.08 0.12 0.16 0.20

customers
Prob. 0.10 0.10 0.20 0.40 0.20

35
 Solution
• Let p be the proportion of customers who will buy the
product. Then the conditional profit is given as:

States of nature Conditional profit

Develop Do not develop
0.04 -44000 0
0.08 -28000 0
0.12 -12000 0
0.16 4000 0
0.20 20000 0
36
 Continued…

States of Prob. Opportunity Loss

nature
Develop Do not
develop

0.04 0.10 44000 0

0.08 0.10 28000 0
0.12 0.20 12000 0
0.16 0.40 0 4000
0.20 0.20 0 20000
EOL 9600 5600

37
 Expected Value of Perfect Information
(EVPI)
• The EMV or EOL criterion helps the DM to select a
course of action that optimizes expected payoff
without any additional information.

• EVPI represents the max amount of money required

to pay for getting additional information about the
occurrence of various states of nature.

• EVPI = (Expected profit with perfect information) –

(expected profit without perfect information or
EMV*) 38
 Example 1
A company needs to increase its production beyond its
existing capacity. It has narrowed down on 2 alternatives in
order to increase the production capacity: (a) expansion, at
the cost of Rs 8 million, or (b) modernization, at the cost of
Rs 5 million. Both approaches require same computational
time. Management believes that over the required payback
period, demand will either be high or moderate. Since high
demand is considered to be somewhat less likely than
moderate, the prob. of high demand is set at 0.35. if the
demand is high, expansion would gross an additional Rs 12
million but modernization would only gross an additional Rs
6 million. On the other hand, if the demand is moderate, the
comparable figures would be Rs 7 million for expansion and
Rs 5 for modernization. Calculate the EVPI.
39
 Solution
State of Prob. Conditional Profit
Nature
Expand (S1) Modernize
(S2)
High demand 0.35 12 – 8 = 4 6–5=1
(N1)
Moderate 0.65 7 – 8 = -1 5–5=0
demand (N2)

EMV (S1) = (0.35*4) + (0.65*-1) = 0.75 million

EMV (S2) = (0.35*1) + (0.65*0) = 0.35 million
Therefore, EMV* = 0.75

40
 Continued…
To calculate EVPI, first calculate EPPI by choosing
optimal course of action for each state of nature.
Multiply conditional profit associated with each course
of action by the given probability to get weighted profit
and then add these weights.

State of Prob. Optimal Conditional Weighted

Nature course of profit profit
action
High 0.35 S1 4 4*0.35 =
demand 1.40
Moderate 0.65 S2 0 0*0.65 = 0
demand
41
 Continued…

• Since EMV* = 0.75 million corresponding to S1,

then
EVPI = EPPI – EMV = 1.40 – 0.75 = Rs. 0.65 million.

• In other words, to get perfect information on demand

patterns, company should consider paying up to Rs
0.65 million.

42
 Example 2
A certain piece of equipment has to be purchased for a
construction project at a remote location. This equipment
contains an expensive part that is subject to random failure.
Spares of this part can be purchased at the same time the
equipment is purchased. Their unit cost is Rs 1500 and they
have no scrap value. If the part fails on the job and no spare is
available, the part will have to be manufactured on a special
order basis. If this is required, the total cost including down
time of the equipment is estimated at Rs 9000 for each
occurrence. Based on previous experience, the following
probability estimates are provided. Determine the EVPI value.

Failure 0 1 2
Prob. 0.80 0.15 0.05
43
 Solution
Let N1 (no failure), N2 (1 failure), and N3 (2 failure) be possible
states of nature. Similarly, let S1 (no spare purchased), S2 (1 spare
purchased), and S3 (2 spare purchased) be possible course of action.

State of nature Courses of Purchase cost Emergency cost Total

action conditional
cost
N1 (0) 0 0 0 0
N1 (0) 1 1500 0 1500
N1 (0) 2 3000 0 3000
N2 (1) 0 0 9000 9000
N2 (1) 1 1500 0 1500
N2 (1) 2 3000 0 3000
N3 (2) 0 0 18000 18000
N3 (2) 1 1500 9000 10500
N3 (2) 2 3000 0 3000
44
 Continued…
State of Prob. Conditional Cost
Nature S1 S2 S3

N1 0.80 0 1500 3000

N2 0.15 9000 1500 3000
N3 0.05 18000 10500 3000

State of Weighted Cost

Nature S1 S2 S3

N1 0.80*0 = 0 1200 2400

N2 1350 225 450
N3 900 525 150
EMV 2250 1950 3000 45
 Continued…
States of Prob. Optimal Conditional Weighted
nature course of cost opportunity
action loss
N1 0.80 S1 0 0*0.80 = 0

N2 0.15 S2 1500 0.15*1500 =

225
N3 0.05 S3 3000 0.05*3000 =
150
EPPI 375

46
 Continued…

Since, EVPI = EPPI – EMV*

= (-375) – (-1950)
= Rs. 1575

47