In d u s t rial En gi n eeri n g| MARMARA UNIVERSIT Y

IE4031

DECISION ANALYSIS
Evaluating Alternatives and Making Choices-I

Dr. Zeynep Tuğçe Kalender 04.11.2021

Making Choices

• Clearly define the problem

• List all possible alternatives
• Identify all possible outcomes for each alternative
• Identify the payoff for each alternative & outcome combination
• Use a decision modeling technique to choose an alternative
Evaluating Alternatives

Type 1 Type 2 Type 3

Decision Decision Decision

making under making under making under
certainty uncertainty risk
 Evaluating Alternatives

Type 1 Type 2 Type 3

Decision making Decision making Decision making

under certainty under uncertainty under risk

• The consequence of every alternative is known

• Usually there is only one outcome for each alternative
• This seldom occurs in reality
 Evaluating Alternatives

Type 1 Type 2 Type 3

Decision making Decision making Decision making

under certainty under uncertainty under risk

• Maximax
• Probabilities of the possible • Maximin
outcomes are not known • Criterion of realism
• Equally likely
• Decision making methods: • Minimax regret
 In d u s t rial En gi n eeri n g| MARMARA UNIVERSIT Y

CASE STUDY
Thompson Lumber Co. Example
Thompson Lumber Co. Example
• Decision: Whether or not to make and sell Outcomes (Demand)

storage sheds Alternatives High Moderate Low

Large plant 200,000 100,000 -120,000
• Alternatives:
Small plant 90,000 50,000 -20,000
• Build a large plant No plant 0 0 0

• Build a small plant

• Do nothing
Apply a decision
• Outcomes: Demand for sheds will be high,
modeling method
moderate, or low
Thompson Lumber Co. Example
Thompson Lumber Co. -Maximax Criterion
• The optimistic approach Outcomes (Demand)
• Best of Bests Alternatives High Moderate Low
Large plant 200,000 100,000 -120,000
• Assume the best payoff will occur for each
Small plant 90,000 50,000 -20,000
alternative
No plant 0 0 0

Outcomes (Demand)
High Moderate Low
Alternatives Choose the large
Large plant 200,000 100,000 -120,000
plant (best payoff)
Small plant 90,000 50,000 -20,000
=200,000
No plant 0 0 0
Thompson Lumber Co. -Maximin Criterion
• The pesimistic approach Outcomes (Demand)
• Best of Worsts Alternatives High Moderate Low
Large plant 200,000 100,000 -120,000
• Assume the worst payoff will occur for each
Small plant 90,000 50,000 -20,000
alternative
No plant 0 0 0

Outcomes (Demand)
High Moderate Low
Alternatives Choose no plant
Large plant 200,000 100,000 -120,000
(best payoff)
Small plant 90,000 50,000 -20,000
=0
No plant 0 0 0
Thompson Lumber Co. – Criterion of Realism
• Uses the coefficient of realism (α) to estimate
Outcomes (Demand)
the decision maker’s optimism (0 < α < 1 ) High Moderate Low
Alternatives
Large plant 200,000 100,000 -120,000
α x (max payoff for alternative)
Small plant 90,000 50,000 -20,000
+ (1- α) x (min payoff for alternative)
No plant 0 0 0
= Realism payoff for alternative

Suppose α = 0.45
Choose small plant

0.45 x (200,000) + (1-0.45) x (-120,000) = 29,500

0.45 x (90,000) + (1-0.45) x (-20,000)
Thompson Lumber Co. – Equally likely
• Assumes all outcomes equally likely and uses
Outcomes (Demand)
the average payoff High Moderate Low
Alternatives
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0

Average
Alternatives
Payoff

Large plant 60,000 200,000 + 100,000 +(−120,000) Choose large plant

𝟑
90,000 + 50,000 +(−20,000) = 60,000
Small plant 40,000
𝟑

No plant 0
Thompson Lumber Co. – Minimax regret
• Regret or opportunity loss measures much Outcomes (Demand)
better Alternatives High Moderate Low

• We want to minimize the amount of regret Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
Regret = (best payoff) – (actual payoff)
No plant 0 0 0

1. The best payoff for each outcome is determined

2. Regret values are calculated

Outcomes (Demand) Max Choose small plant

Alternatives High Moderate Low Regret
Large plant 0 0 120,000 120,000 = 110,000
Small plant 110,000 50,000 20,000 110,000
No plant 200,000 100,000 0 200,000
 Evaluating Alternatives

Type 1 Type 2 Type 3

Decision making Decision making Decision making

under certainty under uncertainty under risk

• Maximax
• Probabilities of the possible • Maximin
outcomes are not known • Criterion of realism
• Equally likely
• Decision making methods: • Minimax regret
Let’s assume that we have an
example in which payoffs are
given in terms of costs
 Be careful about the values in the payoff table!
Maximax Minimax Regret

Minimax
Let’s assume that we have an
example in which payoffs are
given in terms of profits
Be careful about the values in the payoff table!

Equally Likely
(Laplace Approach)

Criterian of Realism
(Hurwicz Approach)
 Evaluating Alternatives

Type 1 Type 2 Type 3

Decision making Decision making Decision making

under certainty under uncertainty under risk

• Where probabilities of outcomes are available

• Expected Monetary Value (EMV) uses the probabilities
to calculate the average payoff for each alternative
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)

Outcomes (Demand)
Alternatives EMV
High Moderate Low

Large plant 200,000 100,000 -120,000 86,000

Small plant 90,000 50,000 -20,000 48,000

No plant 0 0 0 0

Probability of
0.3 0.5 0.2
outcome
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)
EMV (for alternative i) =
∑(probability of outcome) x (payoff of outcome)

Optimal decision:
Counteroffer $5 Billion

Optimal decision strategy:

Counteroffer $5 Billion and
if Texaco counteroffers $3 Billion,
then refuse this counteroffer.
Analyze the decision tree below. What is the
expected payoff for the best alternative?

HINT
First
calculate the
missing
probabilities.
NET SAVINGS

$1,200,000

- $575,000

$2,000,000

- $300,000

$250,000

$1,500,000

$200,000

$10,000
 ($1,200,000 x 0,6 ) + (- $575,000 X 0.40) NET SAVINGS
=$490,000

$1,200,000

- $575,000
$1,540,000
$1,160,000
$2,000,000
$1,540,000

- $300,000

$1,160,000
$250,000

$590,000
$1,500,000
$590,000
$200,000

$10,000
Any
Questions?
 In d u s t rial En gi n eeri n g| MARMARA UNIVERSIT Y

See you next lecture…