Decision Making
Megha Sharma
Indian Institute of Management Calcutta
Decision Trees
Any problem that can be presented in a decision table can also
be graphically represented in a decision tree
Decision trees are most beneficial when a sequence of
decisions must be made
All decision trees contain decision points or nodes and state-of-
nature points or nodes
A decision node from which one of several alternatives may be chosen
A state-of-nature node out of which one state of nature will occur
Five Steps to Decision Tree Analysis
1. Define the problem
2. Structure or draw the decision tree
3. Assign probabilities to the states of nature
4. Estimate payoffs for each possible combination of
alternatives and states of nature
5. Solve the problem by computing expected monetary values
(EMVs) for each state of nature node
Structure of Decision Trees
Trees start from left to right
Represent decisions and outcomes in sequential order
Squares represent decision nodes
Circles represent states of nature nodes
Lines or branches connect the decisions nodes and the
states of nature nodes
TLC’s Decision Tree
A State-of-Nature Node
Favorable Market
A Decision Node
1
Unfavorable Market
Favorable Market
Construct
2
Small Plant Unfavorable Market
TLC’s Decision Tree
EMV for Node 1 = = (0.5)(Rs. 200,000) + (0.5)(–Rs.180,000)
Rs. 10,000
Payoffs
Favorable Market (0.5)
Rs. 200,000
Alternative with best EMV is
selected 1
Unfavorable Market (0.5)
–Rs. 180,000
Favorable Market (0.5)
Rs. 100,000
Construct
2
Small Plant Unfavorable Market (0.5)
–Rs.20,000
EMV for Node 2 = = (0.5)(Rs. 100,000)
Rs. 40,000 + (0.5)(–Rs.20,000)
Rs. 0
Should TLC conduct market research?
TLC can conduct market survey on its own at a cost of $10000.
The survey of course will not provide perfect information but it
may help quite a bit nevertheless. There is a 45% chance that
survey will indicate a favorable market for storage sheds. TLC
knows that there is a 78% chance that the market will actually
be favorable if the survey indicates so. While there is a 27%
chance of a favorable market when survey indicates an
unfavorable market.
ShouldTLC conduct this market research?
Should TLC conduct market research?
First Decision Second Decision Point Payoffs
Point
Favorable Market (0.78)
Rs. 190,000
2 Unfavorable Market (0.22)
–Rs. 190,000
Favorable Market (0.78)
Small Rs. 90,000
Plant
3 Unfavorable Market (0.22)
–Rs. 30,000
No Plant
–Rs. 10,000
1 Favorable Market (0.27)
Rs. 190,000
4 Unfavorable Market (0.73)
–Rs. 190,000
Favorable Market (0.27)
Small Rs. 90,000
Plant
5 Unfavorable Market (0.73)
–Rs. 30,000
No Plant
–Rs. 10,000
Favorable Market (0.50)
Rs. 200,000
6 Unfavorable Market (0.50)
–Rs. 180,000
Favorable Market (0.50)
Small Rs. 100,000
Plant
7 Unfavorable Market (0.50)
–Rs. 20,000
No Plant
Rs. 0
Should TLC conduct market research?
1. Given favorable survey results,
EMV(node 2) = EMV(large plant | positive survey)
= (0.78)($190,000) + (0.22)(–$190,000) = $106,400
EMV(node 3) = EMV(small plant | positive survey)
= (0.78)($90,000) + (0.22)(–$30,000) = $63,600
EMV for no plant = –$10,000
2. Given negative survey results,
EMV(node 4) = EMV(large plant | negative survey)
= (0.27)($190,000) + (0.73)(–$190,000) = –$87,400
EMV(node 5) = EMV(small plant | negative survey)
= (0.27)($90,000) + (0.73)(–$30,000) = $2,400
EMV for no plant = –$10,000
Should TLC conduct market research?
3. Compute the expected value of the market survey,
EMV(node 1) = EMV(conduct survey)
= (0.45)($106,400) + (0.55)($2,400)
= $47,880 + $1,320 = $49,200
4. If the market survey is not conducted,
EMV(node 6) = EMV(large plant)
= (0.50)($200,000) + (0.50)(–$180,000) = $10,000
EMV(node 7) = EMV(small plant)
= (0.50)($100,000) + (0.50)(–$20,000) = $40,000
EMV for no plant = $0
5. Best choice is to seek marketing information
Should TLC conduct market research?
First Decision Second Decision Point Payoffs
Point
Rs. 106,400 Favorable Market (0.78)
Rs. 190,000
Unfavorable Market (0.22)
Rs. 106,400
–Rs.190,000
Rs. 63,600 Favorable Market (0.78)
Small Rs. 90,000
Unfavorable Market (0.22)
Plant –Rs.30,000
No Plant
–Rs.10,000
–Rs. 87,400 Favorable Market (0.27)
Rs. 190,000
Unfavorable Market (0.73)
Rs. 2,400 –Rs.190,000
Rs. 2,400 Favorable Market (0.27)
Small Rs. 90,000
Unfavorable Market (0.73)
Plant –Rs.30,000
No Plant
–Rs.10,000
Rs. 49,200
Rs. 10,000 Favorable Market (0.50)
Rs. 200,000
Unfavorable Market (0.50)
–Rs.180,000
Rs. 40,000
Rs. 40,000 Favorable Market (0.50)
Small Rs. 100,000
Unfavorable Market (0.50)
Plant –Rs.20,000
No Plant
Rs. 0
How do we get these probability values?
Many ways of getting probability data
It can be based on
Management’s experience and intuition
Historical data
Computed from other data using Bayes’ theorem
Bayes’ theorem incorporates initial estimates and
information about the accuracy of the sources
Allows the revision of initial estimates based on new
information
Calculating Revised Probabilities
In the Thompson Lumber case we used these four conditional
probabilities
P (favorable market(FM) | survey results positive) = 0.78
P (unfavorable market(UM) | survey results positive) = 0.22
P (favorable market(FM) | survey results negative) = 0.27
P (unfavorable market(UM) | survey results negative) = 0.73
The prior probabilities of these markets are
P (FM) = 0.50
P (UM) = 0.50
Calculating Revised Probabilities
Through discussions with experts TLC has learnt that
it can use this information and Bayes’ theorem to calculate posterior probabilities
STATE OF NATURE
RESULT OF SURVEY FAVORABLE MARKET UNFAVORABLE MARKET
(FM) (UM)
Positive (predicts P (survey positive | FM) P (survey positive | UM)
favorable market for
product) = 0.70 = 0.20
Negative (predicts P (survey negative | FM) P (survey negative | UM)
unfavorable market for
product) = 0.30 = 0.80
Calculating Revised Probabilities
Recall Bayes’ theorem is
P ( B | A) P ( A)
P( A | B)
P ( B | A) P ( A) P ( B | A ) P ( A )
where
A, B any two events
A complement of A
For this example, A will represent a favorable market and B will
represent a positive survey
Calculating Revised Probabilities
P (FM | survey positive)
P ( survey positive | FM ) P ( FM )
P(survey positive |FM) P(FM) P(survey positive |UM) P(UM)
(0.70 )(0.50 ) 0.35
0.78
(0.70 )(0.50 ) (0.20 )(0.50 ) 0.45
P (UM | survey positive)
P ( survey positive | UM ) P (UM )
P(survey positive |UM) P(UM) P(survey positive |FM) P(FM)
(0.20)(0.50 ) 0.10
0.22
(0.20 )(0.50 ) (0.70 )(0.50 ) 0.45
Calculating Revised Probabilities
POSTERIOR PROBABILITY
CONDITIONAL
PROBABILITY P(STATE OF
P(SURVEY NATURE |
STATE OF POSITIVE | STATE PRIOR JOINT SURVEY
NATURE OF NATURE) PROBABILITY PROBABILITY POSITIVE)
0.35/0.45
FM 0.70 X 0.50 = 0.35 0.78
=
0.10/0.45
UM 0.20 X 0.50 = 0.10 0.22
=
P(survey results positive) = 0.45 1.00
Calculating Revised Probabilities
P (FM | survey negative)
P ( survey negative | FM ) P ( FM )
P(survey negative |FM) P(FM) P(survey negative |UM) P(UM)
(0.30 )(0.50 ) 0.15
0.27
(0.30 )(0.50 ) (0.80 )(0.50 ) 0.55
P (UM | survey negative)
P ( survey negative | UM ) P (UM )
P(survey negative |UM) P(UM) P(survey negative |FM) P(FM)
(0.80)(0.50 ) 0.40
0.73
(0.80 )(0.50 ) (0.30 )(0.50 ) 0.55
Calculating Revised Probabilities
POSTERIOR PROBABILITY
CONDITIONAL
PROBABILITY
P(SURVEY P(STATE OF
NEGATIVE | NATURE |
STATE OF STATE OF PRIOR JOINT SURVEY
NATURE NATURE) PROBABILITY PROBABILITY NEGATIVE)
0.15/0.55
FM 0.30 X 0.50 = 0.15 0.27
=
0.40/0.55
UM 0.80 X 0.50 = 0.40 0.73
=
P(survey results positive) = 0.55 1.00