Decision

Theory
CHAPTER 3
Decision Theory

∙ Represents a general
approach to decision
making that is suitable for
a wide range of operations
management decision.
What is involved in making a good decision?

∙ Decision theory is an analytic and systematic approach to the

study of decision making.
∙ A good decision that is based on logic, considers available data
and possible alternatives and applies a quantitative approach.
∙ A bad decision is one that is not based on logic, does not use all
available data does not consider alternatives and does not employ
appropriate quantitative techniques.
∙ It is not base on favorable outcomes.
 1. Identify the possible
2. Develop a list of possible
future conditions (chance 1
alternatives.
event or state of nature)

THE DECISION 3. Determine or estimate

the payoff associated with 3
4. If possible, estimate the
likelihood of each possible
each alternative for every
PROCESS possible future conditions.
future condition.

5. Evaluate alternatives
according to some decision
criterion (maximize
expected profit) and select
the best alternative.
PAYOFF TABLE
∙ Table showing the expected payoff for each alternative in every possible state of nature.

Alternatives .30 .50 .20

Low Moderate High
Small $ 10 $ 10 $ 10
Facility
Medium 7 12 12
Facility
Large -4 2 16
Facility
DECISION ENVIRONMENT

1. Certainty – means that relevant parameters such as

cost, capacity and demand have known value.
2. Risk – means that certain parameters have
probabilistic outcomes.
3. Uncertainty – means that it is impossible to assess
the likelihood of various possible future events.
Decision Making Under CERTAINTY

When it is known for certain which of the possible

future conditions will actually happen, the decision is
relatively straightforward. Alternatives .30 .50 .20
Low Moderate High
Small $ 10 $ 10 $ 10
Facility
Medium 7 12 12
Facility
Large -4 2 16
Facility
Decision Making Under UNCERTAINTY
Four possible decision criteria:
1. Maximin – choose the alternative with theAlt .30 .50 .20 Maxi Maxi Lapla
best of the worst possible payoffs. min max ce
SF = $ 10; MF = $7; LF = $ -4 LD MD HD
Best worst = $ 10 to build SF
2. Maximax – choose the alternative with SF $ 10 $ 10 $ 10 $ 10 $ 10 $ 10
the best possible payoff.
SF = $ 10; MF = $12; LF = $ 16
Best overall payoff = $16, to build LF. MF 7 12 12 7 12 10.33
3. Laplace – choose the alternative with the
best average payoff of any of the alternatives.
SF= (10+10+10)/3= $ 10 LF -4 2 16 -4 16 4.67
MF= (7+12+12)/3= $ 10.33
LF= (-4+2+16)/3= $ 4.67
Best average payoff is to build a MF.
Decision Making Under UNCERTAINTY
(cont.)
4. Minimax Regret – choose the alternative that has the least of the worst regrets.
a. prepare a table of opportunity losses or regret. (subtract every payoff in each column
from the best payoff in that column)
Alt .30 .50 .20 Worst
b. identify the worst regret for each Regret
alternative. Low Moderat High
e
SF= 6; MF= 4; LF= 14 Small $ 0 $ 2 $ 6 6
Lowest regret is 4 which is for MF. Facility
Medium 3 0 4 4
Facility
Large 14 10 0 14
Facility
Decision Making Under RISK
The probability of occurrence of each state of nature is known, these probabilities
must add to 1.00
Expected Monetary Value (EMV) criterion - the best expected value among the
alternatives.
Probabilities: low=.30; medium=.50; high=.20
EMV(SF)= $10(.30) + $10(.50) + $10(.20) = $10
EMV(MF)= $7(.30) + $12(.50) + $12(.20) = $10.5
EMV(LF)= $-4(.30) + 2(.50) + $16(.20) = $ 3
The highest expected value is $10.5, to build a medium facility.
EXPECTED VALUE OF PERFECT INFORMATION
(EVPI solution 1)
The difference between the expected payoff with perfect information and the
expected payoff under risk.
EVPI = Expected payoff under certainty – Expected payoff under risk
Expected payoff under certainty:
∙ $10(.30) + $12(.50) + $16(.20) = $ 12.2
Expected payoff under risk is computed in EMV as $ 10.5.
EVPI (sol.1) = $12.2 - $10.5 = $ 1.7
EXPECTED VALUE OF PERFECT
INFORMATION (EVPI solution 2)

Use the table of regret, compute for the expected regret for each
alternative using the given probabilities. Alternati .30 .50 .20
SF= 0(.30) + 2(.50) + 6(.20) = 2.2 ves Low Moderate High
MF=3(.30) + 0(.50) + 4(.20) = 1.7 Small $ 0 $ 2 $ 6
Facility
LF=14(.30) + 10(.50) + 0(.20) = 9.2
Medium 3 0 4
The lowest expected regret is 1.7
Facility
Large 14 10 0
Facility
DECISION TREE
∙ A schematic representation of the available alternatives and
their possible consequences. The term gets its name from the
treelike appearance. Although tree diagrams can be used in
place of a payoff table, they are particularly useful for analyzing
situations that involve sequential decisions.

∙ A decision tree is composed of a number of nodes that have

branches emanating from them. Square nodes denote decision
points, and circular nodes denote chance events.
Formative assessment using the previous decision
tree:

Alternatives .40 .60

Low High Maximin Maximax Laplace
Demand Demand
Build $ 40 $ 55 40 55 47.5
Small
Build 10 70 10 70 40
Large
Regret Table
Alternatives .4 .60
Low High Worst
Demand Demand Regret
Build $ 0 $ 15 9
Small
Build 30 0 12
Medium
EMV
Builds Small =(0.40)40 + (0.60)55 = 49

Build Large =(0.40)10 + (0.60)70 = 46

Certainty
EMV = (0.40)40 + (0.60)70 = 58
EVPI (sol. 1)
= certainty – risk
= 58 – 49 = 9
EVPI (sol.2)
BS = (0.40) 0 + (0.60)15 = 9

BL = (0.40)30 +(0.60)0 = 12