Decision Analysis

Comparing Decisions
Planning Horizon = 1 month

Product Table (T) Chair (C) Availability

Resource

Wood (𝑚3 ) 3 2 2000 / 1500 Cost = $150 per 𝑚3

Labor (hours) 3 1 1600 / 1200
Price per unit $1000 $600

Optimal

Scenario Wood Availability Labor Availability Tables Chairs Profit

I 2000 1600 400 400 340000
II 2000 1200 133 800 313333
III 1500 1600 500 0 275000
IV 1500 1200 300 300 255000
 Comparing Decisions
Optimal
Scenario Wood Availability Labor Availability Tables Chairs Profit
I 2000 1600 400 400 340000
II 2000 1200 133 800 313333
III 1500 1600 500 0 275000
IV 1500 1200 300 300 255000

Scenarios
Strategy Prod. Mix I II III IV
Tables first
Or A 400, 400
Chairs first? B 133, 800
C 500, 0
D 300, 300
 Decision Analysis
An analytic and systematic approach to the study of
decision making which is based on logic.

Five Steps
1. Clearly define the problem
2. List all possible alternatives
3. Identify all possible outcomes for each alternative
4. Identify the payoff for each alternative and outcome
combination
5. Use a decision modeling technique to choose an
alternative
 Thompson Lumber
1. Decision: Should he make and sell backyard storage sheds
2. Alternatives:
1. Build a large plant
2. Build a small plant
3. Do nothing
3. Outcomes: Demand for sheds will be
1. High
2. Moderate
3. Low
 Thompson Lumber

4. Payoff table

OUTCOMES
HIGH MODERATE LOW
ALTERNATIVES DEMAND DEMAND DEMAND
Build large plant $200,000 $100,000 –$120,000
Build small plant $ 90,000 $ 50,000 –$ 20,000
No plant $ 0 $ 0 $ 0

5. Select and apply decision analysis model

 Decision-Making Environments
Type 1: Decision making under certainty:
➢ Decision makers know for sure the payoff for every decision alternative
➢ Typically, there is only one outcome for each alternative
Type 2: Decision making under uncertainty:
➢ Decision makers have no information at all about the various outcomes.
➢ They do not know the likelihood (or probability) that a specific outcome will
occur
Type 3: Decision making under risk:
➢ Decision makers have some knowledge regarding the probability of occurrence
of each outcome.
➢ The probability could be a precise measure or an estimate
 Decision making under Uncertainty

• Probabilities of possible outcomes are not known

• Decision making methods:
1. Maximax
2. Maximin
• No universal right decision
3. Criterion of realism • Depends on the outlook of decision maker
4. Equally likely
5. Minimax regret
 Thompson Lumber
• Maximax Criterion
– Maximizes the maximum payoff

OUTCOMES
HIGH MODERATE LOW
ALTERNATIVES DEMAND DEMAND DEMAND
Build large plant $200,000 $100,000 –$120,000
Build small plant $ 90,000 $ 50,000 –$ 20,000
No plant $ 0 $ 0 $ 0

Maximax

Maximax is an optimistic approach

 Thompson Lumber

• Maximin Criterion
– Maximizes the minimum payoff

OUTCOMES
HIGH MODERATE LOW
ALTERNATIVES DEMAND DEMAND DEMAND
Build large plant $200,000 $100,000 –$120,000
Build small plant $ 90,000 $ 50,000 –$ 20,000
No plant $ 0 $ 0 $ 0

Maximin
Maximin is a pessimistic approach
 Thompson Lumber
• Criterion of Realism (Hurwicz)
Thompson’s coefficient of realism a = 0.45
Realism payoff for = a x (Maximum payoff for alternative)
alternative + (1 – a) x (Minimum payoff for alternative)

OUTCOMES
HIGH MODERATE LOW WT. AVG. FOR
ALTERNATIVES DEMAND DEMAND DEMAND ALTERNATIVE
Build large plant $200,000 $100,000 –$120,000 $24,000
Build small plant $ 90,000 $ 50,000 –$ 20,000 $29,500
No plant $ 0 $ 0 $ 0 $ 0

Realism
The criterion of realism uses the weighted average approach
 Thompson Lumber

• Equally Likely (Laplace) Criterion

– Highest average payoff

OUTCOMES
HIGH MODERATE LOW AVERAGE FOR
ALTERNATIVES DEMAND DEMAND DEMAND ALTERNATIVE
Build large plant $200,000 $100,000 –$120,000 $60,000
Build small plant $ 90,000 $ 50,000 –$ 20,000 $40,000
No plant $ 0 $ 0 $ 0 $ 0

Equally
likely
The equally likely criterion selects the highest average alternative
 Thompson Lumber
• Minimax Regret Criterion
Outcomes (Demand)
Alternatives High Moderate Low
Minimax regret is based on
Large plant 200,000 100,000 -120,000 opportunity loss.
Small plant 90,000 50,000 -20,000
No plant 0 0 0

Regret (Opportunity Loss) Table Max

Alternatives High Moderate Low Regret
Large plant 0 0 120,000 120,000
Small plant 110,000 50,000 20,000 110,000
No plant 200,000 100,000 0 200,000 Minimax
 Waldo Books (Q22, P357)
Waldo Books needs to decide how many copies of a new hardcover release to purchase for its
shelves. The store has assumed that demand will be 50, 100, 150, or 200 copies next month, and it
needs to decide whether to order 50, 100, 150, or 200 books for this period. Each book costs
Waldo $20 and can be sold for $30. Waldo can sell any unsold books back to the supplier for $4.

(a) Which option should Waldo choose if it uses the maximax criterion?
(b) Which option should Waldo choose if it uses the maximin criterion?
(c) Which option should Waldo choose if it uses the equally likely criterion?
(d) Which option should Waldo choose if it uses the criterion of realism with 𝛼 = 0.7?
(e) Which option should Waldo choose if it uses the minimax regret criterion?
 Waldo Books (Q22, P357)
Waldo Books needs to decide how many copies of a new hardcover release to purchase for its
shelves. The store has assumed that demand will be 50, 100, 150, or 200 copies next month, and it
needs to decide whether to order 50, 100, 150, or 200 books for this period. Each book costs
Waldo $20 and can be sold for $30. Waldo can sell any unsold books back to the supplier for $4.

Outcomes

Alternatives 50 100 150 200

50 $500.00 $500.00 $500.00 $500.00

100 -$300.00 $1,000.00 $1,000.00 $1,000.00

150 -$1,100.00 $200.00 $1,500.00 $1,500.00

200 -$1,900.00 -$600.00 $700.00 $2,000.00

 Waldo Books (Q22, P357)
Waldo Books needs to decide how many copies of a new hardcover release to purchase for its
shelves. The store has assumed that demand will be 50, 100, 150, or 200 copies next month, and it
needs to decide whether to order 50, 100, 150, or 200 books for this period. Each book costs
Waldo $20 and can be sold for $30. Waldo can sell any unsold books back to the supplier for $4.

𝛼 = 0.7
Outcomes (a) Maximax (b) Maximin (c) Equally Likely (d) Hurwicz

Alternatives 50 100 150 200 Maximum Choice Minimum Choice Average Choice Realism Choice

50 $500.00 $500.00 $500.00 $500.00 $500 $500 Best $500 $500

100 -$300.00 $1,000.00 $1,000.00 $1,000.00 $1,000 -$300 $675 Best $610

150 -$1,100.00 $200.00 $1,500.00 $1,500.00 $1,500 -$1,100 $525 $720

200 -$1,900.00 -$600.00 $700.00 $2,000.00 $2,000 Best -$1,900 $50 $830 Best
 Waldo Books (Q22, P357)
Waldo Books needs to decide how many copies of a new hardcover release to purchase for its
shelves. The store has assumed that demand will be 50, 100, 150, or 200 copies next month, and it
needs to decide whether to order 50, 100, 150, or 200 books for this period. Each book costs
Waldo $20 and can be sold for $30. Waldo can sell any unsold books back to the supplier for $4.

Regret Table
Outcomes Outcomes (e) Minimax

Alternatives 50 100 150 200 Alternatives 50 100 150 200 Maximum Choice
50 $500.00 $500.00 $500.00 $500.00 50 $0 $500 $1,000 $1,500 $1,500
100 -$300.00 $1,000.00 $1,000.00 $1,000.00 100 $800 $0 $500 $1,000 $1,000 Best

150 -$1,100.00 $200.00 $1,500.00 $1,500.00 150 $1,600 $800 $0 $500 $1,600

200 -$1,900.00 -$600.00 $700.00 $2,000.00 200 $2,400 $1,600 $800 $0 $2,400
 Decision making under Risk
Expected Monetary Value (EMV) criterion:
Selects a decision that maximizes the EMV
EMV (Alternative i) = (Payoff of first outcome) x (Probability of first outcome)
+ (Payoff of second outcome) x (Probability of second outcome)
+ … + (Payoff of last outcome) x (Probability of last outcome)

Interpretation?

EMV represents the

long-run average payoff
 Decision making under Risk
Expected Opportunity Loss (EOL) criterion:
Selects a decision that minimizes the EOL
EOL (Alternative i) = (Regret of first outcome) x (Probability of first outcome)
+ (Regret of second outcome) x (Probability of second outcome)
+ … + (Regret of last outcome) x (Probability of last outcome)

Regret (Opportunity Loss) Table

The minimum EOL will always

result in the same decision
alternative as the maximum EMV
 Perfect Information
➢ John Thompson has been approached by Scientific Marketing, Inc., a market research firm
➢ Scientific claims that its analysis will tell John with certainty whether the demand for storage
sheds will be high, moderate, or low and charges $30000 for the information.
➢ It will change John’s problem environment from one of decision making under risk to one of
decision making under certainty
➢ What should John do? Should he hire Scientific to do the marketing study? Is the information
worth $30,000? If not, what is it worth?

Expected Value with Perfect Information (EVwPI)

Expected Value of Perfect Information (EVPI)

 Perfect Information

0.3 0.5 0.2

Expected Value with Perfect Information

EVwPI = $200,000 x 0.3 + $100,000 x 0.5 + $0 x 0.20
= $110,000

Expected Value of Perfect Information

EVPI = EVwPI – Maximum EMV
= $110,000 – $86,000 = $24,000
= Minimum EOL
Because Scientific Marketing wants $30,000 for its analysis, John should reject the offer!
 Perfect Information

Expected Value with Perfect Information (EVwPI)

EVwPI = (Best payoff of first outcome) x (Probability of first outcome)
+ (Best payoff of second outcome) x (Probability of second outcome)
+ … + (Best payoff of last outcome) x (Probability of last outcome)

Expected Value of Perfect Information (EVPI)

EVPI = EVwPI – Maximum EMV EVPI places an upper bound on
what to pay for any information
EVPI = Minimum EOL
 Decision Trees
Presents decision alternatives and outcomes in a sequential manner
Decision node: Arcs (lines) originating from a decision node denote all decision alternatives available to
the decision maker at that node
Outcome node: Arcs (lines) originating from an outcome node denote all outcomes that could occur at
that node

Thompson Lumber

A decision tree usually

begins with a decision node