You are on page 1of 55

EM6113-Engineering Management

Techniques

Week 8 Decision Analysis

1
Learning Outcomes and Reading

 The learning outcomes of this lecture are


 Define and determine the types of decisions

 The formulation process in decision analysis.

 Present decision analysis without and with


probabilities

• Reading
Anderson, Sweeney, Williams “An Introduction to
Management Science: Quantitative Approaches to
Decision Making”.Chapter “Decision Analysis”

2
Importance of Decisions
 Decisions shape the life and destiny of
people and organisations.
 People and organisations make decisions
continuously over their life period.
 Examples of decisions include the university
and course to choose, the company where to
work, the country where to invest, the level of
spending in research and development, and
so on.
3
Type of Decisions
 Simple decisions which do not require too
much analysis to make a decision.
Eg: Do I have to accept the offer of free
accommodation during my holidays.
 Complex decisions which involve difficult
choices and which require complicated
analysis before choosing an alternative.
Eg: Should my company acquire a new
supply chain management software.
4
Decision Analysis
 For evaluating and choosing among
alternatives

 Considers all the possible alternatives and


possible outcomes
Five Steps in Decision Making

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 &
outcome combination
5. Use a decision modeling technique to
choose an alternative
Amazing Co. Example
1. Decision: Whether or not to make and sell
storage sheds
2. Alternatives:
• Build a large plant
• Build a small plant
• Do nothing
3. Outcomes: Demand for sheds will be high,
moderate, or low
4. Payoffs

Outcomes (Demand)
Alternatives High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0

5. Apply a decision modeling method


Types of Decision
Modeling Environments

Type 1: Decision making under certainty

Type 2: Decision making under uncertainty

Type 3: Decision making under risk


Decision Making Under Certainty

 The consequence of every alternative is


known
 Usually there is only one outcome for each
alternative
 This seldom occurs in reality
Decision Making Under Uncertainty
 Probabilities of the possible outcomes are
not known
 Decision making methods:
1. Maximax
2. Maximin
3. Criterion of realism
4. Equally likely
5. Minimax regret
Maximax Criterion
 The optimistic approach
 Assume the best payoff will occur for each
alternative
Outcomes (Demand)
Alternatives High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0

Choose the large plant (best payoff)


Maximin Criterion
• The pessimistic approach
• Assume the worst payoff will occur for each
alternative
Outcomes (Demand)
Alternatives High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0

Choose no plant (best payoff)


Criterion of Realism
 Uses the coefficient of realism (α) to
estimate the decision maker’s optimism
 0<α<1

α x (max payoff for alternative)


+ (1- α) x (min payoff for alternative)
= Realism payoff for alternative
Suppose α = 0.45
Realism
Alternatives Payoff
Large plant 24,000
Small plant 29,500
No plant 0

Choose small plant


Equally Likely Criterion
Assumes all outcomes equally likely and uses
the average payoff

Average
Alternatives Payoff
Large plant 60,000
Small plant 40,000
No plant 0

Chose the large plant


Minimax Regret Criterion
 Regret or opportunity loss measures much
better we could have done
Regret = (best payoff) – (actual payoff)
Outcomes (Demand)
Alternatives High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0
The best payoff for each outcome is highlighted
Regret Values
Outcomes (Demand) 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

We want to minimize the amount of regret


we might experience, so chose small plant
Decision Making 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)
Expected Monetary Value (EMV) Method
Outcomes (Demand)
Alternatives High Moderate Low EMV
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
0.3 0.5 0.2
of outcome

Chose the large plant


Expected Opportunity Loss
(EOL)
 How much regret do we expect based on the
probabilities?

EOL (for alternative i) =


∑(probability of outcome) x (regret of outcome)
Regret (Opportunity Loss) Values
Outcomes (Demand)
Alternatives High Moderate Low EOL
Large plant 0 0 120,000 24,000
Small plant 110,000 50,000 20,000 62,000
No plant 200,000 100,000 0 110,000

Probability
0.3 0.5 0.2
of outcome

Chose the large plant


Decision Trees
 Can be used instead of a table to show
alternatives, outcomes, and payofffs
 Consists of nodes and arcs
 Shows the order of decisions and outcomes
Decision Tree for Amazing Co.
Folding Back a Decision Tree
 For identifying the best decision in the tree
 Work from right to left
 Calculate the expected payoff at each
outcome node
 Choose the best alternative at each decision
node (based on expected payoff)
Amazing Co. Tree with EMV’s
Decision Trees for Multistage
Decision-Making Problems

 Multistage problems involve a sequence of


several decisions and outcomes
 It is possible for a decision to be immediately
followed by another decision
 Decision trees are best for showing the
sequential arrangement
Expanded Amazing Co
Problem
 Suppose they will first decide whether to pay
$4000 to conduct a market survey
 Survey results will be imperfect
 Then they will decide whether to build a large
plant, small plant, or no plant
 Then they will find out what the outcome and
payoff are
Amazing Co.
Optimal Strategy
1. Conduct the survey

2. If the survey results are positive, then build


the large plant (EMV = $141,840)

If the survey results are negative, then build


the small plant (EMV = $16,540)
Utility as a
Decision Making Criterion

 Construct the decision tree as usual with the


same alternative, outcomes, and probabilities
 Utility values replace monetary values
 Fold back as usual calculating expected utility
values
Decision Tree Example for Omar
Omar’s Decision Tree With Utility
Values
Example Time
Lets Practice ……

35
Formulation of Decision Analysis
Problems

 The formulation of Decision Analysis problems


involve the determination of the following:
 The decision alternative di: These indicate the
possible set of actions which can be taken by the
decision maker. These actions are under the control
of the decision maker.
 The states of nature Sj: These indicate the possible
external events which may occur in the world and
impact the outcomes from the decisions. These
states are beyond the control of the decision maker.
 The pay off Vij: These indicates the outcomes for
decision di and state of nature Sj
36
Example of a Decision Analysis
Problem
 An oil company is considering investing in a new field
after receiving encouraging report from its sales
department which forecast high oil prices in the future. If
the sales department’s forecast is correct, the oil
company can expect high profits of about 1 Million
Pounds a day. If the forecast turned out to be incorrect,
the revenues will offset the initial investment and the
company will just break even (that is the profit will be nil).
The company can also invest in an old less productive
field, which involve less investment. If the sales
department’s forecast is correct, the profit will be around
600,000 Pounds a day. However, if the forecast is
incorrect, the profit will be around 200,000 Pounds a
day. Formulate the decision problem for this company.
37
Formulation of the Oil Company
Decision Problem
 Decisions: The company has two possible decisions
d1: Invest in the new field
d2: Invest in the old field
 States of nature: There are two states of nature
S1: High oil price
S2: Low oil price
 Pay off Matrix:

S1 S2
d1 1,000,000 0
d2 600,000 200,000
38
Decision Trees
 Decision trees are graphical representation of te
evolution of decision problems over time.
 They include two type of nodes
 Decision nodes which represent the possible
decisions which can be taken at the corresponding
periods in the decision process.
 Chance nodes which represent the occurrence of the
states of nature at the corresponding periods in the
decision process.
 The payoff for each decision di and state of nature Sj.
 Decisions nodes and chance nodes alternate
on the decision tree. 39
The Oil Company Decision
Tree
S1
1000000
d1

S2
0

S1
600000

d2
200000
S2

40
Decision Analysis Without Probabilities (1)
 Optimistic Approach: The best payoff for each
decision is determined. The optimal decision d* is the
one which generates the maximum payoff among the
best pay offs determined for each individual decision.
V *  Max{Max Vij}
di Sj

 Conservative Approach: The worst payoff for each


decision is determined. The optimal decision d* is the
one which generates the maximum payoff among the
best pay offs determined for each individual decision.

V *  Max{Min Vij}
di Sj

41
Decision Analysis Without Probabilities (2)
 The Minimax Regret Approach: This approach includes
two steps:
 Determination of the regret matrix: The regret Rij associated with
decision di and state of nature Sj is determined as follows

Where Rij  Vj * Vij


Rij: Regret associated with decision di and state of nature Sj
Vj*: The maximum payoff under state of nature Sj
Vij: Payoff associated with decision di and state of nature Sj.
 Determination of the minimax regret: The optimal decision d* is
the one which generates the minimum regret among the
maximum regrets evaluated for each decision.

R*  Min{Max Rij}
di Sj
42
Solution of the Oil Company Decision
Problem: The Optimistic Approach
Step 1: Calculate the maximum payoff associated with
every decision

S1 S2 Max
d1 1,000,000 0 1,000,000
d2 600,000 200,000 600,000

Step 2: Select the decision which maximises the


maximum payoff for every decision

V* =Max{1000000,600000}=1000000 and d*=d1


43
Solution of the Oil Company Decision
Problem: The Conservative Approach
Step 1: Calculate the minimum payoff associated with
every decision

S1 S2 Min
d1 1,000,000 0 0
d2 600,000 200,000 200,000

Step 2: Select the decision which maximises the


minimum payoff for every decision

V* =Max{0,200000}=200000 and d*=d2

44
Solution of the Oil Company Decision Problem:
The Minimax Regret Approach (1)
Step 1: Determine the maximum payoff associated with every state of
nature
S1 S2
d1 1,000,000 0
d2 600,000 200,000
Max 1,000,000 200,000

Step 2: Calculate the regret matrix

S1 S2
d1 0 200,000
d2 400,000 0

45
Solution of the Oil Company Decision Problem:
The Minimax Regret Approach (2)

Step 3: Calculate the maximum regret for every decision

Step 4: Select the decision which minimises the maximum


regret for every decision

R* =Min{200000,400000}=200000 and d*=d1

46
Decision Analysis With Probabilities
 In this case, there are probabilities associated with the occurrence of
the states of nature.
 The Expected Value Approach:
n
EVi  Vij  P( Sj ) for i  1, 2,....n
j 1
n

 P( S )  1
j 1
j

Where
EVi: Expected value of decision di
Vij: Payoff associated with decision di and state of nature Sj.
P(Sj): The probability that the state of nature Sj occurs
The optimal expected value EV* is determined as follows

EV *  Max EVi
i 47
Solution of the Oil Company Decision Problem:
The Expected Value Method (1)

 In the oil company example described before, the sales


department believe that there is a 40% chance that the
oil price will be high in the future and 60% chance that it
will go down. Given this new information, what is the
best decision for this company?

 The new information means that


P(S1)=0.4 and P(S2)=0.6

Step 1: Calculate the expected value associated with


every decision
-EV1= (1000000 x 0.4) + (0 x 0.6) =400000
-EV2= (600000 x 0.4) + (200000 x 0.6) =360000
48
Solution of the Oil Company Decision
Problem: The Expected Value Method (2)

The expected value matrix is as follows

Step 2: Select the decision which maximises the expected


value for every decision

EV*=Max(400000,360000)=400000 and d*=d1

49
Decision Tree for the Oil Company Problem
(Decision Analysis With Probabilities)

S1 (0.4)
1000000
D1 400000

S2 (0.6)
0
400000
S1 (0.4)
600000
360000

d2
200000
S2 (0.6)

50
Sensitivity Analysis
 The role of sensitivity analysis is to determine how the
decision is sensitive to changes in the probabilities
associated with the states of nature.
 Example: How will changes in the probabilities associated with
future oil prices affect the investment decision of the oil company?
Let P1=P(S1), therefore P2=P(S2)=1-P1, therefore
EV1=1000000P1+0P2=1000000P1
EV2=600000P1+200000P2=600000P1+200000(1-P1)
=200000+400000P1
To determine the indifference point we must have EV1=EV2
EV1=EV2 → 1000000P1=200000+400000P1
600000P1=200000 and P1=200000/600000=1/3
If P1>1/3 , EV1>EV2 and the best decision is d*=d1
If P1<1/3, EV1<EV2 and the best decision is d*=d2
51
Graphical Representation of the
Sensitivity Analysis

Sensitivity Analysis for the Oil Company Decision Problem


1000000

800000

600000
EV

400000

200000

0
0 1
EV1 EV2
P1
52
Expected Value of Perfect Information
 If the oil company had a perfect information about the oil prices in
the future, the decision will be

-d=d1 if S=S1 and the payoff would be V=1000000


-d=d2 if S=S2 and the payoff would be V=200000

Therefore the expected value with perfect information is


EVwPI=1000000 x 0.4 + 200000 x 0.6 = 520000

The expected value without perfect information EVwoPI is 400000

Therefore the expected value of perfect information EVPI is

EVPI  EVwPI  EVwoPI =120000


53
References
 Association for Project Management (2006) APM Body of Knowledge 5th
edn. High Wycombe: Association for Project Management
 Chapman, C and Ward, S. (2011) How to Manage Project Opportunity and
Risk. Chichester: John Wiley & Sons, Ltd.
 Degraeve, Z. & Nicholson, N. (2004) Risk: how to make decisions in an
uncertain world. Norwich:Format
 Gardiner, P.D. (2005) Project Management: A Strategic Planning Approach
Basingstoke: Palgrave McMillan
 Gray, C.F. and Larson, E.W. (2008) Project management: the managerial
process 4th edn. McGraw-Hill Irwin
 Hillson,D. (2009) Managing Risk in Projects. Farnham:Gower
 Kerzner, H.(2009) Project Management: a systems approach to planning,
scheduling and controlling. 10th edn. Hoboken, New Jersey: John Wiley &
Sons, Inc
 Marchewka, J.T. (2006) Information Technology Project Management:
Providing Measurable Organisational Value. 2nd edn. J. Wiley & Sons, Inc.
 Project Management Institute (2004) A Guide to the Project Management
Body of Knowledge (PMBOK © Guide) 3rd edn. Project Management
Institute 54
Questions?

55

You might also like