Académique Documents
Professionnel Documents
Culture Documents
DECISION ANALYSIS
8-1. The purpose of this question is to make students use a personal experience to distinguish between
good and bad decisions. A good decision is based on logic and all of the available information. A bad
decision is one that is not based on logic and the available information. It is possible for an unfortunate or
undesirable outcome to occur after a good decision has been made. It is also possible to have a favorable
or desirable outcome occur after a bad decision.
8-2. The decision-making process includes the following steps: (1) define the problem, (2) list the
alternatives, (3) identify the possible outcomes, (4) evaluate the consequences, and (5) select an
evaluation criterion and make the appropriate decision. The first four steps or procedures are common for
all decision-making problems. Step 5, however, depends on the decision-making model used.
8-3. An alternative is a course of action over which we have complete control. A state of nature is an
event or occurrence in which we have no control. An example of an alternative is deciding whether or not
to take an umbrella to school or work on a particular day. An example of a state of nature is whether or
not it will rain on a particular day.
8-4. The basic differences between decision-making models under certainty, risk, and uncertainty depend
on the amount of chance or risk that is involved in the decision. A decision-making model under certainty
assumes that we know with complete confidence the future outcomes. Decision-making-under-risk
models assume that we do not know the outcomes for a particular decision but that we do know the
probability of occurrence of those outcomes. With decision making under uncertainty, it is assumed that
we do not know the outcomes that will occur, and furthermore, we do not know the probabilities that
these outcomes will occur.
8-5. EMV is the expected monetary value. This is the expected return that we would realize if the
decision were repeated an infinite number of times. EVwPI is the expected value with perfect
information. This is the return or value of making the same decision an infinite number of times when we
have perfect or complete information. EVPI is the expected value of perfect information. This is simply
the difference between EMV and EVwPI. It is the amount that we would be willing to pay for perfect
information.
8-6. A decision tree is preferred to a decision table when a number of sequential decisions are to be made.
A sequential decision situation is one in which the outcome of one decision becomes an important factor
in making future decisions. For example, if a decision maker is considering the possibility of acquiring
additional information and a decision of whether or not to build a new plant, the decision to acquire the
new information is made first. Then, based on the results of the new information (if it is gathered), the
decision to build the plant is made. Therefore, these decisions are sequential. One is made before the
other.
8-7. A prior probability is one that exists before additional information is gathered. A posterior
probability is one that can be computed based on prior probabilities and additional information.
8-8. The purpose of Bayesian analysis is to determine posterior probabilities based on prior probabilities
and new information. Bayesian analysis can be used in the decision-making process whenever additional
information is gathered. This information can then be combined with prior probabilities in arriving at
posterior probabilities. Once these posterior probabilities are computed, they can be used in the decision-
making process as any other probability value.
8-9. The overall purpose of utility theory is to incorporate a decision maker’s preference for risk in the
decision-making process.
8-10. A utility function can be assessed in a number of different ways. A common way is to use a
standard gamble. With a standard gamble, the best outcome is assigned a utility of 1, and the worst
outcome is assigned a utility of 0. Then, intermediate outcomes are selected and the decision maker is
given a choice between having the intermediate outcome for sure and a gamble involving the best and
worst outcomes. The probability that makes the decision maker indifferent between having the
intermediate outcome for sure and a gamble involving the best and worst outcomes is determined. This
probability then becomes the utility of the intermediate value. This process is continued until utility
values for all economic consequences are determined. These utility values are then placed on a utility
curve.
8-11. When a utility curve is to be used in the decision-making process, utility values from the utility
curve replace all monetary values at the terminal branches in a decision tree or in the body of a decision
table. Then, expected utilities are determined in the same way as expected monetary values. The
alternative with the highest expected utility is selected as the best decision.
8-12. A risk seeker is a decision maker who enjoys and seeks out risk. A risk avoider is a decision maker
who avoids risk even if the economic payoff is higher. The utility curve for a risk seeker increases at an
increasing rate. The utility curve for a risk avoider increases at a decreasing rate.
SOLUTIONS TO PROBLEMS
8-25. See file P8-25.XLS. Note: In this problem, lower times are the preferred outcome. Therefore we
select the alternative with the lowest EMV. Calculate regret as (time taken - best time taken), then select
alternative with lowest EOL. EVPI = Best EMV – EVwPI. Alternatively, we can multiply all payoffs by
-1, and solve this problem like any other problem.
PAYOFFS Outcomes
Alternatives None Mild Severe EMV Choice
Tennessee 15 30 45 25.00
Back roads 20 25 35 24.17 Best
Expressway 30 30 30 30.00
REGRET Outcomes
Alternatives None Mild Severe EOL Choice
Tennessee 0 5 15 4.17
Back roads 5 0 5 3.33 Best
Expressway 15 5 0 9.17
8-26.
(a) EMV if we construct the clinic = 0.5 * $100,000 + 0.5 * (-$40,000) = $30,000. EMV if we do nothing
= $0. Therefore, construct clinic.
(b) See file P8-26.XLS for the TreePlan solution. Conduct study. Construct clinic if result is positive. Do
not construct clinic if result is negative. EMV = $36,140
(c) EVSI = $11,140. Thus, the physicians would pay up to $11,140 more for the survey. Note: Since
EVSI should be calculated assuming no cost to gather the sample information, $5,000 had to be added
back to $36,140.
(d) EVwPI = $50,000. Best EMV = $30,000. EVPI = $20,000. Efficiency = 55.70%.
8-27.
(a) See file P8-27.XLS, sheet (a) for the utility curve. They are risk avoiders.
(b) See file P8-27.XLS, sheet (a) for the Tree Plan solution. Expected utility if we conduct the survey =
0.761. Expected utility if we do not conduct the survey = 0.90. Therefore, the medical professionals
should not conduct the survey, and should not construct the clinic.
(b) EVSI = $11,000. EVwPI = $35,400. Best EMV = $19,000. EVPI = $16,400. Efficiency = 67.07%.
8-29. See file P8-29.XLS for the TreePlan solution. Note: All costs have been expressed as negative
amounts in the decision tree. Hence, we select highest expected values at all nodes.
(a) Use supplier A. Expected cost is $90 with supplier A versus $113 with supplier B. Note: In the tree,
we pick the EMV of -$90 over the EMV of -$113.
8-34.
(a)
Prior Probabilities
P(Facility works) = 0.60
P(Facility fails) = 0.40
Conditional probabilities
P(Pilot works | Facility works) = 0.80
P(Pilot fails | Facility works) = 0.20
P(Pilot works | Facility fails) = 0.15
P(Pilot fails | Facility fails) = 0.85
(b) See file P8-34.XLS for the revised TreePlan solution. Build the pilot plant. If the pilot plant succeeds,
build facility. If the pilot plant fails, don't build facility. Expected profit = $244,300.
8-35. See file P8-35.XLS for the TreePlan solution. Accept the wager. Use a hard first serve. If the first
serve is out of play, use a hard second serve. Expected payoff = $17.60.
8-36.
(a) See file P8-36.XLS, sheet (a) for the TreePlan solution. Accept the wager. Use a hard first serve. If the
first serve is out of play, use a hard second serve. Expected utility = 0.588.
(b) See file P8-36.XLS, sheet (b) for the utility curve. You are a risk avoider.
8-38.
(a)
Prior Probabilities
P(Oil well) = 0.20
P(Dry well) = 0.80
Conditional probabilities
P(Positive test | Oil well) = 0.85
P(Negative test | Oil well) = 0.15
P(Positive test | Dry well) = 0.25
P(Negative test | Dry well) = 0.75
8-39.
(a) See file P8-39.XLS, sheet (a) for the TreePlan solution. Test the land. If the test result is positive, drill
for oil. If the test result is negative, sell the land. Expected utility = 0.246.
(b) See file P8-39.XLS, sheet (b) for the utility curve. Shamrock Oil is a risk seeker.
8-40.
(a) See file P8-40.XLS for the TreePlan solution. Conduct the survey questionnaire. If the response is
positive, produce razor. If the response is negative, do not produce razor. Expected return = $24,160.
8-41.
(a) See file P8-41.XLS, sheet (a) for the TreePlan solution. Conduct the survey questionnaire. If the
survey response is positive, produce razor. If the survey response is negative, do not produce razor.
Expected utility = 0.823.
(b) See file P8-41.XLS, sheet (b) for the utility curve. Jim is a risk avoider.
8-43.
(a) See file P8-43.XLS, sheet (a) for the utility curve. Jason is a risk seeker.
(b) See file P8-43.XLS, sheet (b) for the TreePlan solution. The use of utilities does not affect Jason’s
decision. The recommended course is still to not lock-in now. If the rate increases after one month, lock-
in then. If the rate is unchanged or decreases after one month, don't lock-in and accept current rate at
settlement. Expected utility = 0.261.
8-45.
(a) See file P8-45.XLS, sheet (a) for the TreePlan solution. Get the information. If it is favorable, build
the store. If it is unfavorable, don't build the store. Expected utility = 0.62.
(b) See file P8-45.XLS, sheet (b) for the utility curve. Sue is a risk seeker.
Prior Probabilities
P(Low demand) = 0.15
P(Fair demand) = 0.40
P(High demand) = 0.45
Conditional probabilities
P(Low survey | Low demand) = 0.70
P(Fair survey | Low demand) = 0.20
P(High survey | Low demand) = 0.10
P(Low survey | Fair demand) = 0.40
P(Fair survey | Fair demand) = 0.50
P(High survey | Fair demand) = 0.10
P(Low survey | High demand) = 0.10
P(Fair survey | High demand) = 0.30
P(High survey | High demand) = 0.60
(1) See file P8-Ski.XLS. The best decision is to have Leadville Barts make the helmets and have
Progressive Products make the rest. Expected value = $2,600. The option of not using Progressive was,
however, very close with an expected value of $2,500.
PAYOFFS Outcomes
Alternatives Poor Average Good Excellent EMV Choice
PP -$5,000 -$2,000 $2,000 $5,000 $700
LB and PP -$10,000 -$4,000 $6,000 $12,000 $2,600 Best
TR and PP -$15,000 -$10,000 $7,000 $13,000 $900
CC and PP -$30,000 -$20,000 $10,000 $30,000 $1,000
LB, CC, and TR -$60,000 -$35,000 $20,000 $55,000 $2,500
Probability 0.1 0.3 0.4 0.2
REGRET Outcomes
Alternatives Poor Average Good Excellent EOL Choice
PP $0 $0 $18,000 $50,000 $17,200
LB and PP $5,000 $2,000 $14,000 $43,000 $15,300 Best
TR and PP $10,000 $8,000 $13,000 $42,000 $17,000
CC and PP $25,000 $18,000 $10,000 $25,000 $16,900
LB, CC, and TR $55,000 $33,000 $0 $0 $15,400
Probability 0.1 0.3 0.4 0.2
(4) Clearly, there are a number of options that Bob did not consider.
(1) MAI’s proposal directly gives Steve the conditional probabilities he needs (e.g., probability of a
successful venture given a favorable survey). Although the information from Iverstine and Kinard (I&K)
is different, we can easily use Bayes’ theorem to on I&K information to compute the revised probabilities
(see file P8-Blake.XLS, sheet Posterior). As such, does not need any additional information from I&K.
(2) Steve’s problem involves three decisions. First, should he contract the services of an outside research
agency? Second, if a survey is warranted, should he employ MAI or I&K? Third, in any case, should the
new product line be introduced?
If Steve decides not to conduct a survey, the decision is to introduce the product with an EMV of
$700,000 [= (0.6)($1,500,000) + (0.4)(-$500,000)].
If Steve decides to conduct the survey, he has to choose between MAI and I&K. If he chooses MAI for
the survey, the best choice is to introduce the product irrespective of whether the survey results are
favorable or unfavorable. The EMV is $800,000 if the survey results are favorable, while the EMV is
only $200,000 if the survey results are unfavorable. The overall EMV of hiring MAI is $500,000 [= (0.5)
($800,000) + (0.5)($200,000)].
If Steve chooses I&K for the survey, the best choice is to introduce the product if survey results are
favorable, for an EMV of $940,000. On the other hand, if the survey results are unfavorable, the best
decision is to not introduce the product for an EMV of -$300,000 (the cost of the survey). The overall
EMV of hiring MAI is $468,800 [= (0.62)($940,000) + (0.38)(-$300,000)].
Comparing these alternatives, Steve should not hire either firm to do the survey. He should simply choose
to introduce the produce right away for an EMV of $700,000.