Decision Making - Certained

MCDM
 Analytic Hierarchy Process

o AHP is a powerful tool that may be used to

make decisions
o when multiple and conflicting
objectives/criteria are present,
o and both qualitative and quantitative aspects
of a decision need to be considered.
• The Analytic Hierarchy Process (AHP) is a multi-criteria
decision making method developed by Thomas Saaty
(1980)*.
• AHP allows decision makers to model a complex problem
in a hierarchical structure, showing the relationships of
 the goal,
 objectives (criteria),
and alternatives.
• AHP is made up of several components such as hierarchical
structuring of complexity, pairwise comparisons, judgments,
an eigenvector method for deriving weights, and
consistency considerations.
*Saaty, Thomas L. [1980] The Analytic Hierarchy Process, McGraw-Hill, New York
AHP is based on three basic principles:
• decomposition,
• comparative judgments, and
• hierarchic composition or synthesis of
priorities.
 Hierarchy
The three major levels
• Goal The goal is a statement of the overall
priority.
• Objectives These are the factors needing
consideration.
• Alternatives We consider the alternatives that
are available to reach the goal.
AHP Process
Define /develop conceptual framework for
decision problem
• Example: Buying a car

Affinity
Diagram

Goal Buy the bes t

Car

General Criteria Handling Economy Power

Secondary Braking Dist Turning Radius Purchase Cos t Maint Cost Gas Mileage Time 0-60
Criteria

Alternatives Ford Taurus Lexus Saab 9000

Saaty’s fundamental scale for pairwise
comparison
Saaty's Fundamental Scale for Pairwise Comparison

Intensity of Definition Explanation

Importance

1 Equal Importance Two criteria contribute equally to the objective

3 Moderate Importance One criteria is moderately favoured over another

5 Strong Importance One criteria is strongly favoured over another

One criteria is strongly favoured over another and its

7 Very strong Importance
dominance is demonstrable

9 Extreme Importance One criteria has the highest possible importance

2,4,6,8 Intermediate values between two adjacent judgments

 Pairwise comparision

For instance I strongly favor banana to apple then I give mark lik
 No of comparisons

Number of things 1 2 3 4 5 6 7
Number of
comparisons 0 1 3 6 10 15 21 ?
Priority vectors-( Eigen value and Eigen
vector)
 Consistency

Since B>A ,A>C and, logically, we hope that B>C or

Banana must be preferable than Cherry. This logic of
preference is called transitive property.
If the answer is Banana more than Cherry, then his
judgment is consistent. On the contrary, if preference
is Cherry over Banana then the answer is inconsistent.
Thus consistency is closely related to the transitive
property.
 Measuring Consistency

• Recall that for consistent 3x3 comparison 3
matrix,
• Compare with from inconsistent matrix
max

 max is calculated by
with the priority weight
multiplying the column total of reciprocal matrix

 max
n
• Use test statistic:
C.I.   Consistency Index
n 1

• From Car Example:

C.I. = (3.039–3)/(3-1) = 0.0195
• Another measure compares C.I. with randomly generated
ones
C.R. = C.I./R.I. where R.I. is the random index
n 1 2 3 4 5 6 7 8
R.I. 0 0 .52 .89 1.11 1.25
Example - Pairwise Comparisons
• Consider following criteria
Purchase Cost Maintenance Cost Gas Mileage

• Want to find weights on these criteria

• AHP compares everything two at a time

(1)
Purchase Cost to Maintenance Cost
Compar
e
– Which is more important?
Say purchase cost
– By how much? Say moderately
3
Example - Pairwise Comparisons
(2) Compare Purchase Cost to Gas Mileage

– Which is more important?

Say purchase cost
– By how much? Say more important
5

(3) Compare Maintenance Cost

to Gas Mileage

– Which is more important?

Say maintenance cost
– By how much? Say more important
3
 Example - Pairwise Comparisons
• This set of comparisons gives the following
matrix: P M G

P 1 3 5

M 1/3 1 3

G 1/5 1/3 1

• Ratings mean that P is 3 times more important than M

and P is 5 times more important than G
• What’s wrong with this matrix?
The ratings are inconsistent!
Techniques that apply to decision
making under uncertainty.
Techniques that apply to decision
making under Risk.
Decision Process-a glance

1
• Identify the problem.

2
• Specify the objectives and criteria for solution.

3
• Develop suitable alternatives.

4
• Analyze and compare alternatives.

5
• Select the best alternative.

6
• Implement the solution.

7
• Monitor to see that desired result is achieved.
 DECESION MAKING ENVIRONMENT
Decisions are made under three types of
Environment.

Certainty Uncertainty Risk

More than one State of More than one state of

Only one state of nature Nature exists but decision
exists i.e. the complete maker lacks sufficient
certainty about the knowledge to allow him
future. assign prob to various
State of Nature
nature exists but the
decision maker has
sufficient information to
allow him assign problem
to each of States.
Decision Making characterized as follows:

Set of
future
conditions

Decision
Making

Known
payoff List of
alternatives alternatives
The information for a decision is often
summarized in a payoff table.

Payoff Table
Table showing the expected payoffs for each
alternative in every possible state of nature.
Payoff Table:

POSSIBLE FUTURE
DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.
Causes for Poor Decisions

Mistakes in decision process

Bounded Rationality

Suboptimization
Bounded Rationality
Limitations on decision making caused by costs,
human abilities, time, technology, and availability of
information.

Because of these limitations, managers can’t always

expect to reach decisions that are optimal in the sense
of providing the best possible outcome. They might
instead, resort to a satisfactory solution.
Decision Environments
Environment in
Environment which it is
in which impossible to
relevant asses the
parameters likelihood of
have known various future
values. events.

Environment at
which certain
future events have
probable
outcomes. Risk
Decision Making Under Certainty

When it is known for

certain which is of the
possible future conditions
will happen, just choose
the alternative that has
the best payoff under the
state of nature.
Decision Making Under Certainty
POSSIBLE FUTURE
DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.

What will you choose to build if the demand will be low,

moderate and high? Or Decision with the perfect Information
 If the demand will be low, just choose the small facility
with a payoff of $10 Million.
 If the demand is moderate choose to build a medium
facility with a payoff $12 Million.
 If the demand is high just build large facility with a $16
Million.
Decision Making Under Uncertainty
Decisions are sometimes made under complete
uncertainty. No information is available on how likely
the various states of nature are:

Laplace
Choose the alternative with the best average period
of any of the alternatives.

Maximin
Choose the alternative with the best of the worst
possible payoff.
Minmax regret
Choose the alternative that has the least of worst
regrets.
Hurwicz
Choose the alternative that has in between most
optimistic and most pessimistic
Decision Making Under Uncertainty
POSSIBLE FUTURE
DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.
Using the laplace approach what will we choose?

Row in Total Row Average

(in $ Million ) (in $ Million)
Because the medium facility
$30 $10.00 has the highest average, it would
be chosen under the Laplace
31 10.33
criterion.
14 4.67
Decision Making Under Uncertainty
POSSIBLE FUTURE
DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.

Using the maximin approach what will we choose?

The worst payoffs for the alternatives are:
Small Facility : $10 million
Medium Facility : 7 million
Large Facility : (4) million

Hence, since $10 million is the best we choose to build

a small facility.
Decision Making Under Uncertainty
POSSIBLE FUTURE
DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.
Using the minimax regret approach what will we choose?
Regrets (in $ Millions)
Alternatives Low Moderate High Worst
Small Facility $0 $2 $6 $6
Medium Facility 3 0 4 4
Large Facility 14 10 0 14

The best of these worst regrets would be chosen using a

minimax regret. The lowest regret is 4, which is for medium
facility, Hence, it would be chosen.
 Decision Making Under Uncertainty
POSSIBLE FUTURE DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.
Using the Hurwicz approach what will we choose?
α the index of optimism(0<=α<=1),The best payoffs for the alternatives:
Max{ α. Maximum + (1-α). Minimum}/ Min{ α. Minimum+ (1-α). Maximum}

Alternative Row min Row max α. Maximum + (1-α). Minimum

Small Facility 10 10 10
Medium Facility 7 12 5α+7
Large Facility -4 16 20α-4
if α=.5,Small facility will be the best alternative
If α=.75,Large facility will be the best alternative
 Activity
ALTERNATIVES STATE OF NATURE

HIGH MOD LOW NIL

Small Facility 50,000 25,000 - 25,000 - 45,000

Medium Facility 70,000 30,000 - 40,000 - 80,000
Large Facility 30,000 15,000 - 1,000 - 10,000
Workout 1
Decision Making Under Risk
Decisions made under the
condition that the probability of
occurrence for each state of nature can
be estimated
A widely applied criterion is
expected monetary value (EMV).
 Decision Making under risk

Expected Value Expected Opportunity

Criterion Loss Criterion Expected Value for
or or Perfect Information
Expected Monetary Expected Value of
Value Criterion Regret
Decision Making Under Risk

EMV
Determine the expected payoff
of each alternative, and choose
the alternative that has the
best expected payoff
This approach is most
appropriate when the decision
maker is neither risk averse nor
risk seeking
Decision Making Under Risk
POSSIBLE FUTURE
DEMAND
Alternatives Low Moderate High
Small Facility $10* $10 $10
Medium Facility 7 12 12
Large Facility (4) 2 16
*Present value in $ millions.

Using the EMV criterion, identify the best alternative for these
probabilities: low=.30,moderate=.50 and high=.20.
EVSmall = .30($10)+.50($10)+.20($10) = $10
EVMedium = .30($7) +.50($12)+.20($12) = $10.5
EVLarge = .30($-4) +.50($2) +.20($16) = $3

Hence, choose the medium facility because it has the

highest expected value.
Expected Value of Perfect Information (EVPI)

The difference between the

expected payoff with perfect
information and the expected payoff
under risk.
Expected Value of Perfect Information (EVPI)
There are two ways to determine EVPI:

Expected Expected
Payoff Payoff
EVPI
Under Under
Certainty Risk

or
Expected Value of Perfect Information (EVPI)
(Using the first method)

.30($10) + .50($12) + .20($16) = $12.2

The expected payoff risk based on Example

4.0 is $10.5.
EVPI = $12.2 - $10.5 = $1.7
Expected Value of Perfect Information (EVPI)
(Using the second method)

Using the table of regrets we can compute the

expected regret for each alternative. Thus:
Small Facility .30(0) + .50(2) + .20(6) = 2.2
Medium Facility .30(3) + .50(0) + .20(4) = 1.7
Large Facility .30(14)+.50(10)+.20(0) = 9.2

The lowest expected regret is 1.7.

Therefore, EVPI = 1.7.
Problem
A newspaper boy has the following probabilities of
selling a magazine:
No. of copies sold Probability
10 0.10
11 0.15
12 0.20
13 0.25
14 0.30
Cost of the copy is 30 paisa and sale price is 50 paisa.
He cannot return the unsold copies. How many
should he order?
 SOLUTION
Cost Price = 30 paisa.
Selling Price = 50 paisa.
Profit = Selling price – Cost price = 20 paisa.
STEP I : CONSTRUCT CONDITIONAL PROFIT TABLE

Profit * S.P. = 20 S.P. ;When D ≥ S

Utility = S.P. * D – C.P. * S = 50D – 30S; When D < S
Possible Possible Stock Action
Demand Probability 10 11 12 13 14
(No. of
Copies Copies Copies Copies Copies
Copies)
10 0.10 200 170 140 110 80
11 0.15 200 220 190 160 130
12 0.20 200 220 240 210 180
13 0.25 200 220 240 260 230
14 0.30 200 220 240 260 280
 STEP II: CONSTRUCT EXPECTED PROFIT TABLE:
Possible Possible Stock Action
Demand Probability 10 11 12 13 14
(No. of
Copies Copies Copies Copies Copies
Copies)
10 0.10 20 17 14 11 8
11 0.15 30 33 28.5 24 19.5
12 0.20 40 44 48 42 36
13 0.25 50 55 60 65 57.5
14 0.30 60 66 72 78 84
TOTAL EXPECTED
PROFIT
200 215 222.5 220 205

THE NEWS BOY MUST, THEREFORE, ORDER 12 COPIES TO EARN THE

HIGHEST POSSIBLE AVERAGE DAILY PROFIT OF 222.5 PAISE
 EXPECTED OPPORTUNITY LOSS CRITERION:
Cost Price = 30 paisa.
Selling Price = 50 paisa.
Profit = Selling price – Cost price = 20 paisa.
STEP I : CONSTRUCT CONDITIONAL PROFIT TABLE

Profit * S.P. = 20 S.P. ;When D ≥ S

Utility = S.P. * D – C.P. * S = 50D – 30S; When D < S
Possible Possible Stock Action
Demand Probability 10 11 12 13 14
(No. of
Copies Copies Copies Copies Copies
Copies)
10 0.10 200 170 140 110 80
11 0.15 200 220 190 160 130
12 0.20 200 220 240 210 180
13 0.25 200 220 240 260 230
14 0.30 200 220 240 260 280
 STEP II: CONSTRUCT CONDITIONAL LOSS
TABLE
ROW-WISE SUBSTRACTION

ROW MAX – OTHER ELEMENTS OF ROW

Possible Possible Stock Action
Demand Probability 10 11 12 13 14
(No. of
Copies Copies Copies Copies Copies
Copies)
10 0.10 0 30 60 90 120
11 0.15 20 0 30 60 90
12 0.20 40 20 0 30 60
13 0.25 60 40 20 0 30
14 0.30 80 60 40 20 0
 STEP III: CONSTRUCT EXPECTED LOSS TABLE:
Possible Possible Stock Action
Demand
Probability 10 Copies 11 Copies 12 Copies 13 Copies 14 Copies
(No. of
Copies)
10 0.10 0 3 6 9 12
11 0.15 3 0 4.5 9 13.5
12 0.20 8 4 0 6 12
13 0.25 15 10 5 0 7.5
14 0.30 24 18 12 6 0

E.O.L. 50 35 27.5 30 45

THE OPTIMUM STOCK ACTION IS THE ONE WHICH WILL

MINIMIZE EXPECTED OPPORTUNITY LOSS; THIS
ACTION CALLS FOR THE STOCKING OF
12 COPIES EACH DAY AT WHICH POINT THERE IS
MINIMUM EXPECTED LOSS OF 27.5 PAISE.
 Expected Value for Perfect Information:
Cost Price = 30 paisa.
Selling Price = 50 paisa.
Profit = Selling price – Cost price = 20 paisa.
STEP I : CONSTRUCT CONDITIONAL PROFIT TABLE

Possible Possible Stock Action

Demand Probability 10 11 12 13 14
(No. of
Copies Copies Copies Copies Copies
Copies)
10 0.10 200
11 0.15 220
12 0.20 240
13 0.25 260
14 0.30 280
STEP II: CONSTRUCT EXPECTED PROFIT TABLE
WITH PERFECT INFORMATION:
DEMAND Probability Conditional Profit Expected Profit With
(No. Of Copies) Under Certainty Perfect Information

10 0.10 200 20
11 0.15 220 33
12 0.20 240 48
13 0.25 260 65
14 0.30 280 84
EPPI = 250

= min E.O.L.
Workout 2
 Workout 3
Under an employment promotion program, it is proposed to
allow sale of newspapers on the buses during off peak hours.
The vendor can purchase the newspaper at a special
concessional rate of 25 paise per copy against the selling price
of 40 paise. Any unsold copies are, however a dead loss. A
vendor has estimated the following probability distribution for
theofno.
No. of copies
copies 15 demanded:
16 17 18 19 20
Probability 0.04 0.19 0.33 0.26 0.11 0.07

a) How many copies should he ordered so that his expected

profit will be maximum?
b) Compute EPPI
c) The vendor is thinking of spending on a small market survey
to obtain additional information regarding the demand levels.
How much should he be willing to spend on such a survey?
Decision
Trees
Decision Trees
 What is a Decision Tree?

• A Visual Representation of Choices,

Consequences, Probabilities, and
Opportunities.
• A Way of Breaking Down
Complicated Situations Down to
Easier-to-Understand Scenarios.
 Easy Example

• A Decision Tree with two choices.

Go to Graduate School to get my

MBA.

Go to Work “in the Real World”

Notation Used in Decision Trees

• A box is used to show a choice that the

manager has to make.

• A circle is used to show that a probability

outcome will occur.

• Lines connect outcomes to their choice

or probability outcome.
 Easy Example - Revisited
What are some of the costs we should take into
account when deciding whether or not to go to
business school?

• Tuition and Fees

• Rent / Food / etc.
• Opportunity cost of salary
• Anticipated future earnings
Simple Decision Tree Model
2 Years of tuition: $55,000, 2 years of
Room/Board: $20,000; 2 years of Opportunity
Cost of Salary = $100,000
Total = $175,000.
PLUS  Anticipated 5 year salary after
Business School = $600,000.
Go to Graduate NPV (business school) = $600,000 - $175,000 =
School to get my $425,000
MBA.

Go to Work “in the

Real World”
First two year salary = $100,000 (from
above), minus expenses of $20,000.
Final five year salary = $330,000
NPV (no b-school) = $410,000

Go to Business School
Is this a realistic model?
Utility Function

• Future uncertainty (interest rates, future salary, etc)

• Cost of Living differences
• Type of Job [utility function = f($, enjoyment)]
• Girlfriend / Boyfriend / Family concerns
• Others?
Utility Function = f ($, enjoyment, family, location, type of job
/ prestige, gender, age, race) Human Factors Considerations
 Mary’s Factory
Mary is a manager of a gadget factory. Her factory has been quite
successful the past three years. She is wondering whether or not it
is a good idea to expand her factory this year. The cost to expand
her factory is $1.5M. If she does nothing and the economy stays
good and people continue to buy lots of gadgets she expects $3M
in revenue; while only $1M if the economy is bad.
If she expands the factory, she expects to receive $6M if economy
is good and $2M if economy is bad.
She also assumes that there is a 40% chance of a good economy
and a 60% chance of a bad economy.
(a) Draw a Decision Tree showing these choices.
 Decision Tree Example
40 % Chance of a Good Economy
Profit = $6M

Expand Factory
Cost = $1.5 M 60% Chance Bad Economy
Profit = $2M
Good Economy (40%)
Profit = $3M
Don’t Expand Factory
Cost = $0 Bad Economy (60%)
Profit = $1M

NPVExpand = (.4(6) + .6(2)) – 1.5 = $2.1M

NPVNo Expand = .4(3) + .6(1) = $1.8M

$2.1 > 1.8, therefore you should expand the factory

 Joe’s Garage

Joe’s garage is considering hiring another mechanic. The

mechanic would cost them an additional $50,000 / year in salary
and benefits. If there are a lot of accidents in Providence this
year, they anticipate making an additional $75,000 in net
revenue. If there are not a lot of accidents, they could lose
$20,000 off of last year’s total net revenues. Because of all the
ice on the roads, Joe thinks that there will be a 70% chance of “a
lot of accidents” and a 30% chance of “fewer accidents”.
Assume if he doesn’t expand he will have the same revenue as
last year.
Draw a decision tree for Joe and tell him what he should do.
 Answer
70% chance of an increase
in accidents
Profit = $70,000
Hire new
mechanic
Cost = $50,000 30% chance of a
decrease in accidents
Profit = - $20,000
Don’t hire new
mechanic
Cost = $0

• Estimated value of “Hire Mechanic” =

NPV =.7(70,000) + .3(- $20,000) - $50,000 = - $7,000
• Therefore you should not hire the mechanic
Decision Trees
Determine the product of the chance probabilities and their respective
payoffs of the branches and the expected value of each initiative:
Decision Trees
Build Small
Low Demand .4($40) = $16
High Demand .6($55) = $33

Build Large
Low Demand .4($50) = $20
High Demand .6($70) = $42

______________________________________________________________________

Build Small
$16 + $33 = $49

Build Large
$20 + $42 = $62

Hence, the choice should be to build the large facility

because it has a larger expected value than the small facility.
Workout 4