Operational Decision Making Tools: Decision Analysis

BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU

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.

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.

# 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.

#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.

Page1 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU

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.

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.

Decision Making Under Risk

Decision making under risk is a probabilistic decision situation. Several possible states of nature may occur, each
with a given probability.

There are three types of methods or criteria available, which could be of help to the decision maker.

1. Expected Monetary Value: EMV is the weighted sum of possible payoffs for each alternative.

i.e. EMV (alternative i ) = (Payoff of first state of nature) x ( Probability of first state of nature)
+(Payoff of second state of nature)x(Probability of second state of nature)
+ …. + (Payoff of last state of nature)x(Probability of last state of nature).

Example: 1 Mc Douglas a national chain fast food restaurant, has been offering a traditional selection of
hamburgers, French fries, soft drinks etc. The company want to introduce breakfast items to the
menu.

Breakfast items are relatively easy to prepare and would not require a large capital outlay for
additional cooking equipment. Most important such items would be sold in the morning when the
demand for the company’s traditional products has been very week. However, because
a. Many people are known to skip breakfast and
b. The company does not know how competitors may react, the demand for the new
products is uncertain.

So, they consider three levels of customer demand- strong, average and weak.
There are two alternative acts available to Mc Douglas

A1 : Introduce breakfast items.

A2 : Do not introduce breakfast items.
And three possible states of nature

S1 : Strong demand
S2 : Average demand
S3 : Weak demand

The management developed a set of payoffs for each act / state combination. The payoff
considered such items as capital outlay, depreciation policies, training costs, additional advertising

Page2 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU
expenditures and so on. Act A 2 , do not introduce breakfast items, has zero payoffs for all states
since three would be no incremental revenue or cost associated with this decision.

Solution: The payoff table according to the data is

State Act
(demand) A1 A2 (Not
(Introduced) Introduced)
Strong, S1: 30 0
Average, S2: 5 0
Weak, S3: -15 0

Status Quo, means do not introduced anything.

Now the management assigns the subjective probability distribution based on the beliefs.

State
(demand) Probability
Strong, S1: 0.2
Average, S2: 0.4
Weak, S3: 0.4

Hence the payoff matrix

S Act EMV (A1) = 30 (.2) + 5 (.4) – 15 (.4)

A1 A2 P (S) = 6 + 2 – 6 = $2
S1: 30 0 0.2
S2: 5 0 0.4 EMV (A2) = 0(.2) + 0 (.4) + 0(.4) = 0
S3: -15 0 0.4

A1 is the optimal act. So, introduced breakfast items.

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 0.25
13 0.25
14 0.30
Cost of a copy is 30 paisa, sale price is 50 paisa. He cannot return unsold copies. How many copies should he
order?

Solution: Sales magnitude are 10,11,12,13,14 . There is no reason to buy less than 10 or more than 14.
Now from any possible combination of supply and demand. The conditional profit table is

Page3 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU

1. Stocking of 10 copies each day will always result in a profit of 200 paisa irrespective of demand. Even if
the demand on some day is 13 copies, he can sell only 10 and hence his conditional profit is 200 paisa.
2. When he stocks 11 copies his profit will be 220 paisa on days when buyers request 11, 12, 13 or 14
copies. But on days when he has 11 copies on stock and buyers buy only 10 copies, his profit
decreases to (200 – 30) = 170 paisa.

Thus the conditional profit in paisa is given by Payoff = 20 x copies sold – 30 x copies unsold.

Conditional profit table

Possible Demand Proba Possible Stock action

(no. of copies ) bility 10 Copies 11 Copies 12 Copies 13 Copies 14 Copies
10 0.10 200 170 140 110 80
11 0.15 200 220 190 160 130
12 0.20 200 220 240 210 180
13 0.25 200 220 240 260 230
14 0.30 200 220 240 260 280

Expected Monetary Value:

EMV (10) = .10 (200) + .15 (200) + .20 (200) + .25 (200) + .30 (200) = 20 + 30 + 40 + 50 + 60 = 200
EMV (11) = .10 (170) + .15 (220) + .20 (220) + .25 (220) + .30 (220) = 17 + 33 + 44 + 55 + 66 = 215
EMV (12) = .10 (140) + .15 (190) + .20 (240) + .25 (240) + .30 (240) = 14 + 28.5 + 48 + 60 + 72 = 222.5
EMV (13) = .10 (110) + .15 (160) + .20 (210) + .25 (260) + .30 (260) = 11 + 24 + 42 + 65 + 78 = 220
EMV (14) = .10 (80) + .15 (130) + .20 (180) + .25 (230) + .30 (280) = 8 + 19.5 + 36 + 57.5 + 84 = 205

The news boy must, therefore order 12 copies to earn the highest possible average daily profit of 222.5 paisa.

2. Expected Opportunity Loss (EOL): It is an approach alternative to the EMV approach.

Opportunity loss, sometimes called regret, refers to the difference between the optimal profit or
payoff and the actual payoff received. In other words, EOL is the cost of not picking the best
solution.

The minimum expected opportunity loss is found by constructing and opportunity loss table and
computing EOL for each alternative. The steps are:

i. The first step is to create the opportunity loss table. This is done by determining the opportunity
loss for not choosing the best alternative for each state of nature.

Define Lij = as the opportunity loss under state Si for act Aj and Li j = M ij  M i*
Where Mi* =The best pay
off under state Si.

ii. The second step is to compute EOL by multiplying the probability of each state of nature times the
appropriate opportunity loss value.

Page4 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU

Example: 3 Mc Dougla’s payoff matrix

S Act
A1 A2 P (S)
S1: 30 0 0.2
S2: 5 0 0.4
S3: -15 0 0.4

Now, Li j = M ij  M i* i.e. L11 = M 11  M 1*  30  30  0 , L12 =

M 12  M 1*  0  30  30 ,

L21 = M 21  M 2*  5  5  0 , L22 =
M 22  M *
2  0 5 5 ,

L31 = M 31  M 3*   15  0 15 , L32 =

M 32  M *
3  0 0 0 ,
Hence the opportunity loss table on the basis of the original matrix is

S Act EOL (A1) = 0 (.2) + 0(.4) + 15 (.4)

A1 A2 P (S) = $6
S1: 0 30 0.2 EOL (A2) = 30(.2) + 5 (.4) + 0(.4) = 8
S2: 0 5 0.4
S3: 15 0 0.4
Hence A1 is the optimal act as it minimize EOL.
Example:4

The Conditional Profit Table of the news paper boy is given

Possible Demand Proba Possible Stock action

(no. of copies ) bility 10 Copies 11 Copies 12 Copies 13 Copies 14 Copies
10 0.10 200 170 140 110 80
11 0.15 200 220 190 160 130
12 0.20 200 220 240 210 180
13 0.25 200 220 240 260 230
14 0.30 200 220 240 260 280

The Opportunity Loss Table / Conditional Loss table (Paisa)

Possible Demand Proba Possible Stock action

(no. of copies ) bility 10 Copies 11 Copies 12 Copies 13 Copies 14 Copies
10 0.10 0 30 60 90 120
11 0.15 20 0 30 60 90
12 0.20 40 20 0 30 60
13 0.25 60 40 20 0 30
14 0.30 80 60 40 20 0

Hence EOL (10) = .10 (0) + .15 (20) + .20 (40) + .25 (60) + .30 (80) = 0 + 3 + 8 + 15 + 24 = 50 (Paisa)
EOL (11) = .10 (30) + .15 (0) + .20 (20) + .25 (40) + .30 (60) = 3 + 0 + 4 + 10 + 18 = 35
EOL (12) = .10 (60) + .15 (30) + .20 (0) + .25 (20) + .30 (40) = 6 + 4.5 + 0 + 5 + 12 = 27.5

Page5 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU
EOL (13) = .10 (90) + .15 (60) + .20 (30) + .25 (0) + .30 (20) = 9 + 9 + 6 + 0+ 6= 30
EOL (14) = .10 (120) + .15 (90) + .20 (60) + .25 (30) + .30 (0) = 12 + 13.5 + 12 + 7.5+ 0= 45
Hence stocking 12 copies each day will minimize expected opportunity loss, which is 27.5 paisa.
3. Expected Value of Perfect Information: (EVPI)

Complete and accurate information about the future demand, referred to as perfect information
would remove all uncertainty form the problem. With this perfect information, the decision maker
would know in advance exactly about the future demand.

EVPI represents the maximum amount he would pay to get the additional information on which may
be based the decision alternative.

EVPI = Expected profit with perfect information – EMV i.e EVPI = EPPI – EMV (max)

Example: 5 Given Mc Douglas payoff matrix

S Act
A1 A2 P (S)
S1: 30 0 0.2
S2: 5 0 0.4
S3: -15 0 0.4

Let Mi* = Maximum payoff or best outcome for first state of nature.

 EPPI  M 8
i . P ( Si )

S Mi* P (Si) M i8 . P ( Si )
S1: 30 0.2 6
S2: 5 0.4 2
S3: 0 0.4 0
 i
M 8
. P ( S )
i = 8

 EPPI  8 Also Max EMV = 2

 EVPI  8  2  $6 * EVPI is sometimes termed the cost of uncertainty.

Exercise: 1 An ice-cream retailer buys ice-cream at a cost of Tk. 5 per cup and sells it for Tk. 8 per cup; any
remaining unsold at the end of the day can be disposed of at a salvage price of Tk. 2 per cup. Past sales have
ranged between 15 and 18 cups per day; there is no reason to believe that sales volume will take on any other
magnitude in future. Find the EMV, EOL and EVPI if the sale history has the following probabilities:
Market size: 15 16 17 18
Probability: 0.10 0.20 0.40 0.30

Page6 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU

Decision Making Under Uncertainty

When a manager cannot assess the outcome probability with confidence or when virtually no probability data are
available, other decision criteria are required. This type of problem has been referred to as decision making under
uncertainty. The criteria or method that we cover in this section include

1. Maximax (optimistic)
2. Maximin (pessimistic)
3. Minimax
4. Hurwicz Criterion (Criterion of realism)
5. Laplace Criterion or Equally likely criterion or Criterion of Rationality.

1. Maximax (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 (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.

3. Minimax 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 Criterion: 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.

Example: 6

The following matrix gives the payoff of different strategies (alternatives) S 1,S2, S3 against conditions (events)
N1, N2, N3 & N 4 .

N1 N2 N3 N4
S1 Rs. 4000 Rs. –100 Rs. 6000 Rs. 18000
S2 20000 5000 400 0
S3 20000 15000 -2000 1000

Indicate the decision taken under the following approach:

Page7 of 8
Operational Decision Making Tools: Decision Analysis
BBA/BRE Program.
Sayma Suraiya, Faculty of Business & Economics. DIU

i. Pessimistic
ii. Optimistic
iii. Equal Probability
iv. Regret
v. Hurwicq Criterion, his degree of optimism being 0.7

Solution:
Pessimistic Optimistic Equal Probability Value

S1 -100 18000 Rs. ¼ ( 4000 – 100 + 6000 + 18000) = 6975

S2 0 20000 Rs. ¼ ( 20000 + 5000 + 400 + 0) = 6350
S3 -2000 20000 Rs. ¼ ( 20000 + 15000 -2000 + 1000) = 8500

i. S2 is the optimal decision.

ii. S2 or S3 is the optimal decision.
iii. S3 is the alternative to be selected.
iv. Under regret criterion

i th regret = (maximum payoff – i th payoff) for the jth event.

N1 N2 N3 N4 Maximum
regret
S1 16000 15100 0 0 16000
S2 0 10000 5600 18000 18000
S3 0 0 8000 17000 17000

The decision alternative S1 would be chosen since it corresponds to the minimal of the maximum possible
regrets.

v. For the given payoff matrix the minimum and the maximum payoff for each alternative are
given below.

Minimum Maximum Payoff = α. Maximum + (1- α) minimum

payoff payoff Where α = 0.7
S1 -100 18000 .7 x 18000 + .3 x (-100) = 12570
S2 0 20000 .7 x 20000 + .3 (0) = 14000
S3 -2000 20000 .7 x 20000 + .3 (-2000) = 13400

Thus under Hurwicz rule, alternative S 2 should be chosen as it is associated with the
highest payoff of Rs. 14000.

Exercise: Taylor, p-68. No. S2-1, S2-2, S2-3,S2-5,S2-6, S2-7

Page8 of 8