CHAPTER FIVE

DECISION THEORY/ANALYSIS
 INTRODUCTION

In dealing with LP, models were formulated and

solved in order to aid the manager in making
decision.
The solutions to the models were represented
by values for the decision variables. However,
these LP models are formulated under the
assumption that certainty existed.
In actual practice, however, many decision
making situations occur under conditions of
uncertainty.
 Decision Theory/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.
• The decision model includes a list of courses of action
available and the possible consequences of each course of
action.
• An important factor in making a decision is the degree of
certainty associated with consequences.
 Characteristics of Decision Theory Problems

• Decision theory problems are characterized by the

following:
– A list of alternatives: a set of mutually exclusive and
collectively exhaustive options available to the decision
maker.
– A list of possible future states of nature: a set of possible
future conditions or events, beyond the control of the
decision maker that determine the decision’s consequence.
 Cont…

- Payoffs: the payoffs might be profits, revenues, costs, or

other measures of value. Usually the measures are financial.
Usually payoffs are estimated values.
for the various alternative/state of nature combination.
combinations.
– Degreeof certainty: These include complete
certainty, risk and complete uncertainty.
– Decision criteria: the decision maker’s attitudes toward
the decision as well as the degree of certainty that
surrounds a decision. Example; maximize the expected
payoffs.
 Decision Making Environments

• Decisions are made under three types of environments:

1. Under conditions of certainty
– The decision maker has perfect knowledge about the
future outcomes, and selects the one with best payoff.
2. Under conditions of uncertainty/With out probabilities/
– The decision maker lacks the knowledge about the
probabilities of their occurrence.
– the decision maker is uncertain which state of nature will occur
in the future, and has no control over them.
 Cont…

• Examples:
– Manufacturer introducing a new product in the
marketplace
.
3. Decision making under conditions of risk/With
Probabilities/

– but the decision maker has sufficient knowledge to

allow him/her to assign probabilities to the various
states of nature.
 The Payoff Table

• A payoff table is a device used to summarize and

organize information relevant to a particular decision.
• It includes:
– a list of alternatives,
– possible future states of nature, and
– the payoffs associated with each of the alternative/state
of nature combinations, and
– probabilities [if available].
 Cont…

• 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.
• General Format:
 Example

• The following payoff table provides data about profits of the various
states of nature/alternative combination.
Alternatives State of Nature(Level of Demand)
High Moderate Low
A1 4 16 12
A2 5 6 10
A3 -1 4 15

Under conditions of complete certainty, the decision maker simply picks up the best
payoff in that state of nature and chooses the associated alternative.

If we know that S2 will occur, the decision maker then can focus on the first raw of the payoff
table. Because alternative A1 has the largest profit (16), it would be selected.
 Decision Making Under Complete Uncertainty

• There are approaches (criteria) to decision making under

complete uncertainty.
• These are:
A. Maximin (Pessimism)
B. Maximax (Optimism)
C. Minimax Regret, and
D. Principle of Insufficient Reason (Laplace)
 A. Maximin Criterion

• It is the pessimistic or conservative approach.

• It consists of identifying the worst payoff for each alternative,
and, then selecting the alternative that has the best of the
worst payoffs.
Example
• Indicate the optimal decision for the problem using Maximin

Decision: A2 will be chosen.

 B. Maximax Criterion

• It is an optimistic approach, and associated with

choosing the alternative that could result in the best payoff.
• Consists of identifying the best payoff for each alternative and
selecting the alternative that has the best of the best payoffs.
Example
• Indicate the optimal decision for the problem using Maximax

Decision: A1 will be chosen.

 C. Minimax regret Criterion

• It is based on a regret the decision maker may experience after the

decision has been made.
• The objective is to minimize regret before making decision.
• Then, identify the maximum regret for each alternative and selecting the
one with the smallest of the max. regrets.
 In order to use this approach, it is necessary to develop an
opportunity loss table. The opportunity loss reflects the difference
between each payoff and the best possible payoff in a column (i.e.,
given a state of nature). Hence, opportunity loss amounts are found
by identifying the best payoff in a column and, then, subtracting
each of the other values in the column from that payoff.
- Therefore, this decision avoids the greatest regret by selecting the
decision alternative that minimizes the maximum regret.
 Cont’d

Example
• Indicate the optimal decision for the problem using
Minmax regret
Opportunity Loss table:

Decision: A1 will be chosen.

 D. Principle of Insufficient Reason (Laplace)

• It is based on the point that there is not enough information

to conclude that occurrences of each state of nature will be
different.
• Therefore, equal weights should be assigned to all events of each
alternative, and the one with the best average payoff will be selected.
Example
• Indicate the optimal decision for the problem using Laplace

Decision: A1 will be chosen.

 Decision Making Under Risk

• Under the conditions of risk, probabilities will be assigned for

the occurrence of the various states of nature.
There are three decision criteria for decisions under risk
1. Expected Monetary Value(EMV)
2. Expected Opportunity Loss (EOL)
3. Expected Value of Perfect Information (EVPI)
• For example:
For the given example above assume the estimated probability of
high demand at 0.2, the probability of moderate demand at 0.5, and
the probability of low demand at 0.30.
 1. Expected Monetary Value(EMV)

• This approach provides the decision maker with a value which

represents an average payoff for each alternative.
• The best alternative is, then, the one that has the highest
expected monetary value.
– The average or expected payoff of each alternative is a weighted
average using probabilities of events.
• Thus, EMV is:
 Cont…

Example:
• Find the optimal decision for the problem using the EMV approach.
Solution:

Decision: A1 will be chosen.

 2. Expected Opportunity Loss (EOL)

• In this case, a table of opportunity losses will be used rather

than a table of payoffs.
• Hence, the opportunity losses for each alternative are
weighted by the probabilities of their respective event to
compute a long run average opportunity loss.
• The alternative with the smallest expected loss will be
selected.
• The EOL will always result in the same choice as the EMV.
• It is because they are equivalent ways of combining the values;
– maximizing payoffs is equivalent to minimizing the
opportunity losses.
 Cont
…
Example:
• Find the optimal decision for the problem using the EOL approach.
Solution:
Expected Opportunity Loss table

Decision: A1 will be chosen.

 3. Expected Value of Perfect Information

• It is a measure of the difference between payoff that could be

realized under condition of certainty (EPC) and the
expected payoff under a condition involving risk (EMV).
 Example:
• The EVPI for the problem solved above can be computed as:
EPC = 0.2 (5) + 0.5(16)+0.3(15) = 13.5
EMV = 12.4 (computed earlier)
• EVPI = EPC – EMV
EVPI = 13.5 – 12.4 = 1.10
 Cont…

• Note that the EVPI is exactly equal to EOL (In fact, these two
quantities will always be equal).
• The EOL indicates the expected opportunity losses due to
imperfect information, OR
– the expected payoff that could be achieved by having perfect
information.
• The EVPI represents an upper bounded on the amount of money
the company would spend to obtain perfect information.
 II. Decision Tree

• The Payoff Table approach is useful for a non-sequential or

single stage decision situations.
• Many real-world decision problems consists of a sequence of
dependent decisions.
• Decision Trees are useful in analyzing multi-stage
decision processes.
• A Decision Tree is a chronological
representation of the decision process.
• The decision tree is composed of nodes (squares and
circles) and branches (lines).
 Cont…

• The squares indicate decision Typical Decision tree format:

points
• Circles represent chance events
(circles and squares are called
nodes)
• The lines emanating from
squares represent alternatives.
• The lines from circles represent
states of nature
• The tree is read from right to
left.
 Example

• Given the following Pay off table for Real Estate investment:
a. Draw up a decision tree.
b. Determine which decision maximizes the Real Estate
investor’s expected payoff?
c. How much will be the optimal expected payoff for the
investment
State of Nature (Economic Condition)
Decision Good (0.6) Poor (0.4)
(Purchase)
Apartment 50,000 30,000
building
Office building 100,000 -40,000
Warehouse 30,000 10,000
 Solution

a) The decision tree is developed as follows.

Purchase $50,000

2
$30,000
Decision $100,000
Office
1 3
building $-40,000

$30,000

4
$10,000
 Cont…

b) Determining the optimum strategy

– The expected value of the payoffs is computed at each
probability node.
– Start with the final outcomes (payoffs) and work
backward through the decision tree toward node 1.
EV (node 2) = 0.60($ 50,000) + 0.40($ 30,000) = $42,000
EV (node 3) = 0.60($100,000) + 0.40($-40,000) = $44,000
EV (node 4) = 0.60($ 30,000) + 0.40($ 10,000) = $22,000

• Each of these three expected values at nodes 2, 3, and 4 are

the outcomes of a possible decision that can occur at node 1.
 Cont…

• Now we will be following the folding-back procedure.

 Folding-Back Procedure
• Starting from the right of the decision tree and working
back to the left:
1. At each probability node, calculate an EMV—a sum of
products of monetary values and probabilities.
2. At each decision node, take a maximumof EMVs to
identify the optimal decision.
 Cont…

• Moving toward node 1, we select the branch that comes from

the probability node with the highest expected payoff.
• The branch corresponding to the highest payoff ($44,000)
will be considered. This branch represents the decision to
purchase the office building.
Decision: The optimal strategy is to purchase office
building
C. The optimal expected payoff for the investment is $44, 000.
 Cont…

• Decisiontree showing the optimal strategyand the associated

expected payoff value.
Decision $42,000 $50,000

2
$30,000
Office $44,000 $100,000
Building
Office
1 3
building $-40,000

$22,000 $30,000

4
$10,000