Decision Analysis: Management Science Decision Trees

• A decision tree is a chronological representation of

the decision problem.
Decision Analysis
• Each decision tree has two types of nodes; round
• Problem Formulation
nodes correspond to the states of nature while
• Decision Making without Probabilities square nodes correspond to the decision
• Decision Making with Probabilities alternatives.

• Risk Analysis and Sensitivity Analysis • The branches leaving each round node represent
the different states of nature while the branches
• Decision Analysis with Sample Information leaving each square node represent the different
• Computing Branch Probabilities with Baye’s decision alternatives.
Theorem • At the end of each limb of a tree are the payoffs
• Utility Theory attained from the series of branches making up that
limb.

Problem Formulation
Decision Making without Probabilities
• A decision problem is characterized by decision
alternatives, states of nature, and resulting payoffs. • Three commonly used criteria for decision making
when probability information regarding the
• The decision alternatives are the different possible likelihood of the states of nature is unavailable are:
strategies the decision maker can employ.
– The optimistic approach
• The states of nature refer to future events, not
under the control of the decision maker, which may – The conservative approach
occur. States of nature should be defined so that – The minimax regret approach.
they are mutually exclusive and collectively
exhaustive. Optimistic Approach

Influence Diagrams • The optimistic approach would be used by an

optimistic decision maker.
• An influence diagram is a graphical device
showing the relationships among the decisions, the • The decision with the largest possible payoff is
chance events, and the consequences. chosen.

• Squares or rectangles depict decision nodes. • If the payoff table was in terms of costs, the
decision with the lowest cost would be chosen.
• Circles or ovals depict chance nodes.
• Diamonds depict consequence nodes.
Conservative Approach
• Lines or arcs connecting the nodes show the
direction of influence. • The conservative approach would be used by a
conservative decision maker.
• For each decision the minimum payoff is listed
Payoff Tables and then the decision corresponding to the
• The consequence resulting from a specific maximum of these minimum payoffs is selected.
combination of a decision alternative and a state of (Hence, the minimum possible payoff is
nature is a payoff. maximized.)
• A table showing payoffs for all combinations of • If the payoff was in terms of costs, the maximum
decision alternatives and states of nature is a payoff costs would be determined for each decision and
table. then the decision corresponding to the minimum of
these maximum costs is selected. (Hence, the
• Payoffs can be expressed in terms of profit, cost, maximum possible cost is minimized.)
time, distance or any other appropriate measure.
Minimax Regret Approach
• The minimax regret approach requires the
construction of a regret table or an opportunity loss
table.
• This is done by calculating for each state of nature
the difference between each payoff and the largest
payoff for that state of nature.
• Then, using this regret table, the maximum regret
for each possible decision is listed. Minimax Regret Approach
• The decision chosen is the one corresponding to  For the minimax regret approach, first compute a
the minimum of the maximum regrets. regret table by subtracting each payoff in a column
from the largest payoff in that column. In this
example, in the first column subtract 4, 0, and 1
Example from 4; etc. The resulting regret table is:
Consider the following problem with three decision
alternatives and three states of nature with the
following payoff table representing profits:

Optimistic Approach
 An optimistic decision maker would use the
optimistic (maximax) approach. We choose the Practice Exercise
decision that has the largest single value in the Consider the following problem with three decision
payoff table. alternatives and four states of nature with the
following payoff table representing profits:

Conservative Approach
 A conservative decision maker would use the
conservative (maximin) approach. List the
minimum payoff for each decision. Choose the
decision with the maximum of these minimum
payoffs.
Decision Making with Probabilities
• Expected Value Approach
 If probabilistic information regarding the states of
nature is available, one may use the expected value
(EV) approach.
 Here the expected return for each decision is
calculated by summing the products of the payoff
under each state of nature and the probability of the
respective state of nature occurring.
The decision yielding the best expected return is
chosen.

Expected Value of a Decision Alternative

Expected Value of Perfect Information

• Frequently information is available which can
improve the probability estimates for the states of
nature.
• The expected value of perfect information (EVPI)
Example: is the increase in the expected profit that would
Burger Prince result if one knew with certainty which state of
nature would occur.
•Burger Prince Restaurant is contemplating opening
a new restaurant on Main Street. It has three • The EVPI provides an upper bound on the
different models, each with a different seating expected value of any sample or survey
capacity. Burger Prince estimates that the average information.
number of customers per hour will be 80, 100, or • EVPI Calculation
120.
– Step 1: Determine the optimal return
corresponding to each state of nature.
– Step 2: Compute the expected value of these
optimal returns.
– Step 3: Subtract the EV of the optimal decision
from the amount determined in step (2).
 • The outcomes of this analysis are called posterior
probabilities or branch probabilities for decision
Risk Analysis
trees.
• Risk analysis helps the decision maker recognize
Computing Branch Probabilities
the difference between:
• Branch (Posterior) Probabilities Calculation
– The expected value of a decision alternative, and
– Step 1: For each state of nature, multiply the prior
– The payoff that might actually occur
probability by its conditional probability for the
• The risk profile for a decision alternative shows indicator -- this gives the joint probabilities for the
the possible payoffs for the decision alternative states and indicator.
along with their associated probabilities.
– Step 2: Sum these joint probabilities over all states
-- this gives the marginal probability for the
indicator.
– Step 3: For each state, divide its joint probability
by the marginal probability for the indicator - this
gives the posterior probability distribution.

Expected Value of Sample Information

• The expected value of sample information (EVSI)
is the additional expected profit possible through
knowledge of the sample or survey information.
• EVSI Calculation
Sensitivity Analysis
– Step 1: Determine the optimal decision and its
• Sensitivity analysis can be used to determine how
expected return for the possible outcomes of the
changes to the following inputs affect the
sample using the posterior probabilities for the
recommended decision alternative:
states of nature.
– Probabilities for the states of nature
– Step 2: Compute the expected value of these
– Values of the payoffs optimal returns.
• If a small change in the value of one of the inputs – Step 3: Subtract the EV of the optimal decision
causes a change in the recommended decision obtained without using the sample information from
alternative, extra effort and care should be taken in the amount determined in step (2).
estimating the input value.

Efficiency of Sample Information

Bayes’ Theorem and Posterior Probabilities
• Efficiency of sample information is the ratio of
• Knowledge of sample or survey information can EVSI to EVPI.
be used to revise the probability estimates for the
• As the EVPI provides an upper bound for the
states of nature.
EVSI, efficiency is always a number between 0 and
• Prior to obtaining this information, the probability 1.
estimates for the states of nature are called prior
probabilities.
• With knowledge of conditional probabilities for
the outcomes or indicators of the sample or survey
information, these prior probabilities can be revised
by employing Bayes' Theorem.
Example: Burger Prince
• Sample Information
Burger Prince must decide whether or not to
purchase a marketing survey from Stanton
Marketing for $1,000. The results of the survey are
"favorable" or "unfavorable". The conditional
probabilities are:

P(favorable | 80 customers per hour) = .2

P(favorable | 100 customers per hour) = .5
P(favorable | 120 customers per hour) = .9
Should Burger Prince have the survey performed by
Stanton Marketing?
Meaning of Utility
• Utilities are used when the decision criteria must
be based on more than just expected monetary
values.
• Utility is a measure of the total worth of a
particular outcome, reflecting the decision maker’s
attitude towards a collection of factors.
• Some of these factors may be profit, loss, and risk.
• This analysis is particularly appropriate in cases
where payoffs can assume extremely high or
extremely low values.

Example: Risk Avoider