Decision Tree Analysis
Maydison Ginting
maydison.ginting@prasetiyamulya.ac.id
Why (business) Decision Analysis
• Help us make a better business decision that
makes companies stay competitive in a
dynamically-changing environment
Innovate or Die (?)
success
Reward = +++
innovate
failure
Reward = - - -
Don’t
Reward = 0
innovate
Uncertainty vs Risk
• Decisions Under Uncertainty (Decision Making
without Probabilities)
– A decision model for which there is more than one state of
nature, but without knowledge of the probabilities for
occurrence of the various state of nature.
• Decisions Under Risk (Decision Making with
Probabilities)
– A decision model for which there is more than one state of
nature, and a probability for occurrence of each state of
nature is known.
Uncertainty vs Risk
Payoff Table
States of Nature
Decision
a (x%)* b ((100-x)%)*
1 Payoff 1a Payoff 1b
2 Payoff 2a Payoff 2b
* if known, decisions under risk
if not known, decisions under uncertainty
States of nature: the events that may occur in the future
- high or low demand
- good or bad economic conditions
Decision Making without Probabilities
Decision situation:
States of Nature
Decision
(Purchase) Good Economic Poor Economic
Conditions Conditions
Apartment building $ 50,000 $ 30,000
Office building 100,000 - 40,000
Warehouse 30,000 10,000
Decision Making without Probabilities
The Maximax Criterion
• In the Maximax criterion the decision maker selects the decision
that will result in the maximum value of the maximum return of
each decision
• An optimistic criterion
States of Nature
Decision
(Purchase) Good Economic Poor Economic Maximum
Conditions Conditions payoff
Apartment building $ 50,000 $ 30,000 50,000
Office building 100,000 - 40,000 100,000
Warehouse 30,000 10,000 30,000
What if the payoff table consists of costs?
The selection would be the minimum of the minimum costs Minimin
Decision Making without Probabilities
The Maximin Criterion
• In the Maximin criterion the decision maker selects the decision
that will result in the maximum value of the minimum return of
each decision
• A pessimistic criterion
States of Nature
Decision
(Purchase) Good Economic Poor Economic Minimum
Conditions Conditions payoff
Apartment building $ 50,000 $ 30,000 30,000
Office building 100,000 - 40,000 - 40,000
Warehouse 30,000 10,000 10,000
What if the payoff table consists of costs?
The selection would be the minimum of the maximum costs Minimax
Decision Making with Probability
GOOP algorithm:
• Tentukan apa yang menjadi tujuan dan/atau value yang
ingin dicapai. (Goals)
• Tentukan sekumpulan alternatif keputusan. (Options)
• Tentukan peristiwa/event/state of the world yang akan
terjadi beserta besar payoff/reward untuk masing-masing
alternatif keputusan di setiap event. (Outcomes)
• Tentukan nilai probabilitas dari masing-masing event.
(Probability)
• Tentukan expected value untuk setiap alternatif keputusan.
• Pilih alternatif yang memiliki expected value terbesar.
Decision Making with Probabilities
The Expected Value
• There is more one state of nature
• The decision maker can arrive at a probability estimate for the occurrence of the
states of nature
• Expected value is computed by multiplying each decision outcome under each
state of nature by the probability of its occurrence
• Management should make the decision that maximizes the Expected Value
States of Nature
Decision
(Purchase) Good Economic Poor Economic
Conditions Conditions
Apartment building $ 50,000 $ 30,000
Office building 100,000 - 40,000
Warehouse 30,000 10,000
Probability 0,60 0,40
Decision Making with Probabilities
States of Nature
Decision
(Purchase) Good Economic Poor Economic
Conditions Conditions
Apartment building $ 50,000 $ 30,000
Office building 100,000 - 40,000
Warehouse 30,000 10,000
Probability 0,60 0,40
The Expected Value
• EV(Apartment) = $50,000(0.6) + 30,000(0.4) = 42,000
• EV(Office) = $100,000(0.6) + (-40,000)(0.4) = 44,000 (Max value)
• EV(Warehouse) = $30,000(0.6) + 10,000(0.4) = 22,000
Decision Trees
A dicision tree is a diagram consisting of
• decision nodes (represented as squares)
• probability nodes (circles)
• decision alternatives (branches)
Chance Branch
Choice
Decision Making with Probabilities
A Decision Tree
Sequential Decision Trees
• A sequential decision tree is used to illustrate a
situation requiring a series of decisions.
• Used where a payoff table, limited to a single
decision, cannot be used.
Real estate investment example modified to encompass a ten-year period in
which several decisions must be made:
Buying a Used Car
• Costs $1000
• Can sell it for $1100, $100 profit
• Every car is a lemon or a peach
• 20% are lemons 80% peach
• Costs $40 to repair a peach, $200 to
repair a lemon
• Should we buy a car?
THANK YOU
