Chapter 9.
DECISION
RECOGNIZING RISK
©3522017Batangas State
University
Introduction
In previous chapters, we stated specific assumptions concerning applicable revenues,
costs, and other quantities important to an engineering economy analysis. It was
assumed that a high degree of confidence could be placed in all estimated values. That
degree of confidence is sometimes called assumed certainty. Decisions made solely on
the basis of this kind of analysis are sometimes called decisions under certainty. The
term is rather misleading in that there rarely is a case in which the best of estimates of
quantities can be assumed as certain.
We now consider the more realistic situation in which estimated future quantities are
uncertain and project outcomes are risky. The motivation for dealing with risk and
uncertainty is to establish the bounds of error such that another alternative being
considered may turn out to be a better choice than the one we recommended under
assumed certainty.
353 ©2017 Batangas State University
Learning Objective
• Discuss and illustrate several
probabilistic methods that are useful
in analyzing risk and uncertainty
associated with engineering
economic studies
354 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Expected Monetary Value (EMV) uses the probabilities to
calculate the average payoff for each alternative, choose the
alternative that has the best expected payoff.
EMV for alternative i = n(probability of outcome)*(payoff of
outcome)
355 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Expected Monetary Value Method
Outcomes (Demand)
Alternatives EMV
High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0
Probability
0.3 0.5 0.2
of outcome
356 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Expected Opportunity Loss (EOL)
How much regret do we expect based on the probabilities?
EOL for alternative i = n(probability of outcome)*(regret of outcome)
Steps:
1. Construct Opportunity Loss Table
• Choose the highest value per outcome and subtract with each alternatives
2. Multiply the opportunity loss by the probability of that loss for each outcome or
each alternative and add these together
3. Minimum EOL will always result in the same decision as maximum EMV.
4. Minimum EOL will always equal EVPI
357 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Outcomes (Demand)
Alternatives EMV
High Moderate Low
Large plant 200,000 100,000 -120,000
Small plant 90,000 50,000 -20,000
No plant 0 0 0
Probability
0.3 0.5 0.2
of outcome
356 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Regret (Opportunity Loss) Values
Outcomes (Demand)
Alternatives EOL
High Moderate Low
Large plant
Small plant
No plant
Probability
0.3 0.5 0.2
of outcome
358 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Perfect Information
It would tell us with certainty which outcome is going to
occur
Having perfect information before making a decision would
allow choosing the best payoff for the outcome
359 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Expected Value with Perfect Information (EVwPI)
The expected payoff of having perfect information before
making a decision
EVwPI = n(probability of outcome)*(best payoff of outcome)
360 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Expected Value of Perfect Information (EVPI)
The amount by which perfect information would increase our
expected payoff.
Provides an upper bound on what to pay for additional
information
EVPI = EVwPI - EMV
EVwPI = Expected Value with Perfect Information
EMV = the best EMV without perfect information
361 ©2017 Batangas State University
Expected Monetary Value of Alternatives
Demand
Alternatives
High Moderate Low EVwPI=
Large plant 200,000 100,000 -120,000 EVPI = EVwPI - EMV
Small plant 90,000 50,000 -20,000 EVPI =
No plant 0 0 0
Probability of
0.3 0.5 0.2
outcome
Best Payoff 200,000 100,000 0
36 ©2017 Batangas State University
2
Expected Monetary Value of Alternatives
The operations manager for a well-drilling company must recommend whether to build
a new facility, expand his existing one or do nothing. He estimates that long-run profits
in($000) will vary with the amount of precipitation (rainfall) as follows:
Alternative Low Normal High
Do Nothing -100 100 300
Expand 350 500 200
Build New 750 300 0
Probability 0.30 0.20 0.50
Solve for:
a. EMV
b. EVPI
c. EOL
361 ©2017 Batangas State University
Decision Tree Analysis
Decision trees, also called decision flow networks and
decision diagrams, are powerful means of depicting and
facilitating the analysis of important problems, especially
those that involve sequential decisions and variable outcomes
over time.
Decision trees are used in practice because they make it
possible to breakdown a large, complicated problem into a
series of smaller simple problems, and they enable objective
analysis and decision making that includes explicit
consideration of the risk and effect of the future.
363 ©2017 Batangas State University
Decision Tree Analysis
Decision trees
It can be used instead of a table to show alternatives,
outcomes and payoffs
Consists of nodes and arcs
Shows the order of decisions and outcomes
364 ©2017 Batangas State University
Decision Tree Analysis
Decision trees techniques
1. Alternatives
2. State of nature with probability values
The objective is to make decision that maximizes the expected
value of the alternatives.
365 ©2017 Batangas State University
Decision Tree Analysis
Given
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
Probability
0.3 0.5 0.2
of outcome
356 ©2017 Batangas State University
Decision Tree Analysis
86,000
High (0.3)
200,000
Moderate(0.5)
100,000
Low (0.2)
48,000 -120,000
High(0.3)
90,000
Small Moderate (0.5)
Plant 50,000
Low (0.2)
0 -20,000
High (0.3)
0
Moderate (0.5)
0
Low (0.2)
0
Sample Tree Diagram Select Large Plant with EMV =
86,000
366 ©2017 Batangas State University
Sample Problems
Southern Hospital Supplies, a company that makes hospital growns, is considering the
capacity expansion. It's alternatives are to do nothing, build a small plant, build a medium
plant or build a large plant. The new facility would produce a new type of grown and
currently the potential or marketability for this product is unknown. If a large plant is built
and favourable market exists, a profit of $100,000 could be realized. An unfavourable market
would yield a $90,000 loss. However, a medium plant would earn $60,000 profit with a
favourable market. A $10,000 loss would result from an unfavourable market. A small plant,
on the otherhand, would return $40,000 with favourable market conditions and lose $5,000 in
an unfavourable market. Of course, there are always the option of doing nothing.
Recent market research indicates that there is 0.4 probability of a favourable market, which
means that there is also 0.6 probability of an unfavourable market. With this information, the
alternatives that will result in the highest expected monetary value (EMV) can be selected.
a. Prepare a decision tree and compute the EMV for each branch.
b. If a new estimate of the loss from a medium plant is an unfavourable market increases to -
$20,000, what is the new EMVfor this branch?
367 ©2017 Batangas State University
Sample Problems
Favourable Market
(0.4)
Unfavourable Market
(0.6)
Favourable Market
(0.4)
Unfavourable Market
(0.6)
Favourable Market
(0.4)
Unfavourable Market
(0.6)
Favourable Market
(0.4)
Unfavourable Market
(0.6)
368 ©2017 Batangas State University
Problems
A firm that plans to expand it's product line must decide whether to build a
small or large plant facility to produce the new products. If it build a small
facility and demand is low, the NPV after deducting for building cost will be
$400,000. If demand is high, the firm can either maintain the small facility
or expand it. Expansion would have a net present value of $450,000 and
maintaining the small facility would have a NPV of $50,000. If a large
facility is build and demand is high, the estimated net present value is
$800,000. If demand returns out to be low, the net present value will be -
$10,000. The probability that demand is high is estimated to be 0.6 and the
probability that the demand is low is estimated to be 0.4. Prepare a decision
tree analysis and what is the best alternative?
370 ©2017 Batangas State University
