Decision Theory

• Decision making deals with methods for determining the optimal course of
action when a number of alternatives are available
• Decision making is an integral part of management planning, organizing,
controlling and motivation processes
• The decision maker selects one strategy (course of action) over others
depending on some criteria, like utility, sales, cost or rate of return, etc.
Decision
• Significance of decision
• Time availability
• Cost of analysis
• Degree of complexity
Types of environments
✓ Decision making under certainty
✓ Decision making under uncertainty
✓ Decision making under risk
✓ Decision making under conflict
Decision making under certainty
• A state of certainty exists when a decision maker knows, with reasonable
certainty,
✓ what the alternatives are and
✓ what conditions are associated with each alternative.
Decision making under Uncertainty
• A state of uncertainty exists when a decision maker
• does not know all of the alternatives,
• the risks associated with each,
• the consequences each alternative is likely to have
• Relevant information
• Approach the situation from a logical and rational view
• Intuition, judgment and experience
Decision making under conflict
• Situations exists in which two or more opponents with conflicting objectives try
to make decisions with each trying to gain at the cost of others.
Decision Making Framework
• Defining the problem.
• Establish the decision criteria
• Formulation of a model
• Generating alternatives
• Evaluation of the alternatives
• Implementation and monitoring
Decision Methodology
• Complete Certainty Methods:
• Algebra, Calculus Mathematical Programming
• Risk and Uncertainty Methods:
• Statistical Analysis
• Extreme Uncertainty Methods:
• Game theory
Decision making under Uncertainty
The five approaches/criterion for Decision making under Uncertainty are as follows:
✓ Criterion of Optimism (Maximax)
✓ Criterion of Pessimism (Maximin) (Conservative Approach)
✓ Minimax Regret Criterion (Regret payoff)
✓ Hurwicz Criterion (Criterion of Realism )
✓ Laplace Criterion (Equally Likelihood) (Criterion of Rationality )

Your newly launched product is likely to have different states of demand such as high,
moderate, low or poor demand.
The likely payoff by using three different strategies are forecasted as follows:
How will you make decision under uncertainty?
Alternatives States of Nature (Product Demand)
High Moderate Low Poor
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000
Contract 70000 30000 -40000 -80000
Subcontract 30000 15000 -1000 -10000

Criterion of Optimism (Maximax)

• Determine maximum payoff for every strategy
• Select the strategy with best out of all maximum possible payoff
Alternatives States of Nature (Product Demand) Maximum
High Moderate Low Poor of the
Row
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000 50000
Contract 70000 30000 -40000 -80000 70000
Subcontract 30000 15000 -1000 -10000 30000

Maximax= 70000
Therefore , Decision with Criterion of Optimism should be to go with CONTRACT

Criterion of Pessimism (Maximin) (Conservative Approach)

This criterion identifies the worst outcome of each alternative and then select the best
of those worst outcomes.
It states that the decision maker maximises his minimum possible payoffs.
• Determine the minimum possible payoff for each alternative
• Then chooses the alternative with maximum payoff within this group.
Alternatives States of Nature (Product Demand) Minimum
High Moderate Low Poor of the
Row
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000 -45000
Contract 70000 30000 -40000 -80000 -80000
Subcontract 30000 15000 -1000 -10000 -10000

Maximax= -10000
Therefore , Decision with Criterion of Pessimism should be to go with
SUBCONTRACT

Minimax Regret Criterion (Regret payoff)

Step 1: Determine the best payoff for each state of nature and create a regret table.
Alternatives States of Nature (Product Demand)
High Moderate Low Poor
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000
Contract 70000 30000 -40000 -80000
Subcontract 30000 15000 -1000 -10000
Best Payoff 70000 30000 -1000 -10000

Step 2: Determine the amount of regret for each event.

(Best Payoff- Event Payoff)
Alternatives States of Nature (Product Demand)
High Moderate Low Poor
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 70000- 5000 24000 35000
50000
=20000
Contract 0 0 39000 70000
Subcontract 40000 15000 0 0
Best Payoff 70000 30000 -1000 -10000

Step 3: Determine the maximum regret for each decision.

Alternatives States of Nature (Product Demand) Maximum
High Moderate Low Poor Regret
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 20000 5000 24000 35000 35000
Contract 0 0 39000 70000 70000
Subcontract 40000 15000 0 0 40000

Step 4: Select the decision with the minimum value from the column of maximum
regrets.
Alternatives States of Nature (Product Demand) Maximum
High Moderate Low Poor Regret
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 20000 5000 24000 35000 35000
Contract 0 0 39000 70000 70000
Subcontract 40000 15000 0 0 40000

Maximum Regret= 35000

Therefore , Decision with Criterion of Regret should be to go with EXPAND
 Hurwicz Criterion (Criterion of Realism )
The Hurwicz criterion attempts to find a compromise between the extremes posed by
the optimist and pessimist criteria. Instead of assuming total optimism or pessimism,
Hurwicz incorporates a measure of both by assigning a certain percentage weight to
optimism and the balance to pessimism.
The criterion of realism consists of the following states:
1. Choose appropriate degree of optimism (α), so that (1- α) represents the degree
of pessimism
2. Determine the maximum as well as minimum of each alternative and obtain
P= α * Maximum + (1- α) * Minimum
3. Choose the alternative that yields the maximum value of P
Determine Maximum & Minimum payoff for each row
Alternatives States of Nature (Product Demand) Maximum Minimum
High Moderate Low Poor of the of the
Row Row
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000 50000 -45000
Contract 70000 30000 -40000 -80000 70000 -80000
Subcontract 30000 15000 -1000 -10000 30000 -10000

Calculate weighted average payoff using suitable degree of optimism (α) and degree
of pessimism (1-α)
Assuming, Degree of Optimism = α= 0.8
Alternatives States of Nature (Product Demand) Maximum Minimum P= α *
of the of the Maximum
High Moderate Low Poor
Row Row + (1- α) *
(Rs.) (Rs.) (Rs.) (Rs.)
Minimum

Expand 50000 25000 -25000 -45000 50000 -45000 31000

Contract 70000 30000 -40000 -80000 70000 -80000 40000
Subcontract 30000 15000 -1000 -10000 30000 -10000 22000

Maximum yield=Pmax= 40000

Therefore , Decision with Criterion of Realism should be to go with CONTRACT

Laplace Criterion (Equally Likelihood) (Criterion of Rationality) (Bayes Criterion)

This criterion is based upon what is known as the principle of insufficient reason.
This criterion assigns equal probabilities to all the events of each alternative decision
and selects the alternative associated with the maximum expected pay off.
Alternatives States of Nature (Product Demand) Expected Payoff
High Moderate Low Poor
= ∑ Event payoff/n
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000 50000 + 25000 − 25000 − 45000
= 1250
4
Contract 70000 30000 -40000 -80000 -5000
Subcontract 30000 15000 -1000 -10000 8500
Maximum of expected payoff= 8500
Therefore , Decision with Criterion of rationality should be to go with
SUBCONTRACT

Decision making under risk

•The decision maker does not know which state
of nature will occur but can estimate the probability of occurrence for each state
•These probabilities may be subjective (they usually represent estimates from experts
in a particular field), or they may reflect historical frequencies.
•Future probabilities estimated or known:
• Expected Value (EV) Approach
• Expected Opportunity Loss (EOL) Approach
• Expected Value of Perfect Information (EVPL) approach
Determine the expected pay-off of each alternative, and choose the alternative that
has the best-expected pay-off using the expected monetary value criterion
Expected Monetary value for Expand Strategy=
=Σ Expected value of each * Probability of Occurrence
=50000*0.20+ 25000*0.40+ (-25000)* 0.30+ (-45000)*0.10
=8000
Alternatives States of Nature (Product Demand) Expected Monetary value
High Moderate Low Poor
(Rs.) (Rs.) (Rs.) (Rs.)
Expand 50000 25000 -25000 -45000 8000
Contract 70000 30000 -40000 -80000 6000
Subcontract 30000 15000 -1000 -10000 10700
Probability of 0.20 0.40 0.30 0.10
Occurrence

Maximum of expected payoff= 10700

Therefore , Decision with Expected monetary value criteria should be to go with
SUBCONTRACT Strategy

Case:
A newspaper boy has the following probabilities of selling a magazine:

No of Copies sold Probability

10 0.10
11 0.15
12 0.20
13 0.25
14 0.30

Cost of a copy is 30 paise and the sale price is 50 paise. He cannot return unsold
copies.
How many copies should be ordered?
Conditional Profit table:
Payoff= 20*Copies Sold – 30* Copies Unsold
= 20*10 - 30*0= 200

Possible Probability Possible Stock Option

Demand 10 11 12 13 14
Copies Copies Copies Copies 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 Profit Table:

Expected Profit= Payoff* Associated Probability
= 200*0.10= 20

Possible Probability Possible Stock Option

Demand 10 11 12 13 14
Copies Copies Copies Copies Copies
10 0.10 20 17 14 11 8
11 0.15 30 33 28.5 24 19.5
12 0.20 40 44 48 42 36
13 0.25 50 55 60 65 57.5
14 0.30 60 66 72 78 84
Total Expected Profit 200 215 222.5 220 205

As maximum total expected profit is 222.5

Therefore, newsboy should order 12 copies as a best possible stock option.

Expected Opportunity Loss (EOL) approach:

• EOL represents the amount by which maximum possible profit will be reduced
under various possible stock options.
• The course of actions that minimises these losses is the optimal decision
alternative as per following:
• Prepare the conditional profit table for each decision event combination and
write the associated probabilities.
• For each event determine the conditional opportunity loss (COL) by subtracting
the payoff from the maximum payoff for that event.
• Calculate the EOL for each decision alternative by multiplying the COL by the
associated probabilities and then adding the values.
• Select the alternative that yields the lowest EOL.

Conditional Loss table:

Having unsold copy will have a loss of 30 paise
Not having copy will have opportunity loss of 20paise

Possible Probability Possible Stock Option

Demand 10 11 12 13 14
Copies Copies Copies Copies 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

Expected Loss Table:

Possible Probability Possible Stock Option

Demand 10 11 12 13 14
Copies Copies Copies Copies Copies
10 0.10 0 3 6 9 12
11 0.15 3 0 4.5 9 13.5
12 0.20 8 4 0 6 12
13 0.25 15 10 5 0 7.5
14 0.30 24 18 12 30 0
Total Expected Loss 50 35 27.5 30 45

As minimum total expected loss is 27.5

Therefore, newsboy should order 12 copies as a best possible stock option.

Expected Value of Perfect Information (EVPI):

If decision-makers can get perfect (complete and accurate) information about the
occurrence of various states of nature, then choosing a course of action that yields the
desired payoff in the presence of any state of nature is easy.

Conditional Profit table:

Possible Probability Possible Stock Option
Demand 10 11 12 13 14
Copies Copies Copies Copies Copies
10 0.10 200 - - - -
11 0.15 - 220 - - -
12 0.20 - - 240 - -
13 0.25 - - - 260 -
14 0.30 - - - - 280

Expected profit with perfect information

=200*0.10 + 220*0.15+ 240*0.20+ 260*0.25+ 280*0.30 = 250