Decision Analysis: Eaching Uggestions
Decision Analysis: Eaching Uggestions
Decision Analysis: Eaching Uggestions
3
C H A P T E R
Decision Analysis
17
18 CHAPTER 3 DECISION ANALYSIS
of Rob1. The robots will be used to perform a variety of repair The Hurwicz approach uses a coefficient of realism value of
opera- tions on large industrial equipment. Of course, George can 0.7, and a weighted average of the best and the worst payoffs for
always do nothing and not buy any robots (alternative 3). The market each alternative is computed. The results are as follows:
for the re- pair could be either favorable (event 1) or unfavorable Weighted average (alternative 1) = ($50,000)(0.7)
(event 2). George has constructed a payoff matrix showing the + (—$40,000)(0.3)
expected returns of each alternative and the probability of a favorable
= $23,000
or unfavorable market. The data are presented below:
Weighted average (alternative 2) = ($30,000)(0.7)
+ (—$20,000)(0.3)
EVENT 1 EVENT 2 = $15,000
Probability 0.6 0.4 Weighted average (alternative 3) = 0
Alternative 1 50,000 —40,000 The decision would be alternative 1.
Alternative 2 30,000 —20,000 The minimax regret decision minimizes the maximum oppor-
Alternative 3 0 0 tunity loss. The opportunity loss table for Goleb is as follows:
First
Decision
Point
Second
Decision
Point
–$5,000
1
Favorable Market (0.158)
$45,000
4 Unfavorable Market (0.842)
–$45,000
Favorable Market (0.158)
$25,000
Rob2
5 Unfavorable Market (0.842)
–$25,000
–$5,000
$–5,000
$0
SOLUTIONS TO DISCUSSION
QUESTIONS AND PROBLEMS 3-10. The purpose of Bayesian analysis is to determine poste-
rior probabilities based on prior probabilities and new
3-1. The purpose of this question is to make students use a per- information. Bayesian analysis can be used in the decision-
sonal experience to distinguish between good and bad decisions. making process whenever additional information is gathered. This
A good decision is based on logic and all of the available informa- information can then be combined with prior probabilities in
tion. A bad decision is one that is not based on logic and the avail- arriving at posterior probabilities. Once these posterior
able information. It is possible for an unfortunate or undesirable probabilities are computed, they can be used in the decision-
outcome to occur after a good decision has been made. It is also making process as any other prob- ability value.
possible to have a favorable or desirable outcome occur after a
bad decision. 3-11. The expected value of sample information (EVSI) is the
increase in expected value that results from having sample infor-
3-2. The decision-making process includes the following steps: mation. It is computed as follows:
(1) define the problem, (2) list the alternatives, (3) identify the
pos- sible outcomes, (4) evaluate the consequences, (5) select an EVSI = (expected value with sample information)
evalua- tion criterion, and (6) make the appropriate decision. The + (cost of information) — (expected value
without sample information)
first four steps or procedures are common for all decision-making
problems. Steps 5 and 6, however, depend on the decision-making 3-12. The overall purpose of utility theory is to incorporate a de-
model. cision maker’s preference for risk in the decision-making process.
3-3. An alternative is a course of action over which we have 3-13. A utility function can be assessed in a number of different
complete control. A state of nature is an event or occurrence in ways. A common way is to use a standard gamble. With a
which we have no control. An example of an alternative is decid- standard gamble, the best outcome is assigned a utility of 1, and
ing whether or not to take an umbrella to school or work on a par- the worst outcome is assigned a utility of 0. Then, intermediate
ticular day. An example of a state of nature is whether or not it outcomes are selected and the decision maker is given a choice
will rain on a particular day. between having the intermediate outcome for sure and a gamble
involving the best and worst outcomes. The probability that makes
3-4. The basic differences between decision-making models
the decision maker indifferent between having the intermediate
under certainty, risk, and uncertainty depend on the amount of
outcome for sure and a gamble involving the best and worst
chance or risk that is involved in the decision. A decision-making
outcomes is determined. This probability then becomes the utility
model under certainty assumes that we know with complete confi-
of the intermediate value. This process is continued until utility
dence the future outcomes. Decision-making-under-risk models
values for all economic conse- quences are determined. These
assume that we do not know the outcomes for a particular
utility values are then placed on a utility curve.
decision but that we do know the probability of occurrence of
those out- comes. With decision making under uncertainty, it is 3-14. When a utility curve is to be used in the decision-making
assumed that we do not know the outcomes that will occur, and process, utility values from the utility curve replace all monetary
furthermore, we do not know the probabilities that these outcomes values at the terminal branches in a decision tree or in the body of
will occur. a decision table. Then, expected utilities are determined in the
same way as expected monetary values. The alternative with the
3-5. The techniques discussed in this chapter used to solve deci-
highest expected utility is selected as the best decision.
sion problems under uncertainty include maximax, maximin, equally
likely, coefficient of realism, and minimax regret. The maximax 3-15. A risk seeker is a decision maker who enjoys and seeks
decision-making criterion is an optimistic decision-making criterion, out risk. A risk avoider is a decision maker who avoids risk even
while the maximin is a pessimistic decision-making criterion. if the potential economic payoff is higher. The utility curve for a
risk seeker increases at an increasing rate. The utility curve for a
3-6. For a given state of nature, opportunity loss is the difference
risk avoider increases at a decreasing rate.
between the payoff for a decision and the best possible payoff for
that state of nature. It indicates how much better the payoff could 3-16. a. Decision making under uncertainty.
have been for that state of nature. The minimax regret and the b. Maximax criterion.
mini- mum expected opportunity loss are the criteria used with c. Sub 100 because the maximum payoff for this is
this. $300,000.
3-7. Alternatives, states of nature, probabilities for all states of Row Row
nature and all monetary outcomes (payoffs) are placed on the Equipmen Favorabl Unfavorabl Maximu Minimum
deci- sion tree. In addition, intermediate results, such as EMVs for t e e m
mid- dle branches, can be placed on the decision tree. Sub 100 300,000 —200,000 300,000 —200,000
Oiler J 250,000 —100,000 250,000 —100,000
3-8. Using the EMV criterion with a decision tree involves
Texan 75,000 —18,000 75,000 —18,000
starting at the terminal branches of the tree and working toward
the origin, computing expected monetary values and selecting the 3-17. Using the maximin criterion, the best alternative is the
best alternatives. The EMVs are found by multiplying the proba- Texan (see table above) because the worst payoff for this ($—
bilities of the states of nature times the economic consequences
18,000) is better than the worst payoffs for the other decisions. 3-
and summing the results for each alternative. At each decision
18. a. Decision making under risk—maximize expected
point, the best alternative is selected.
monetary value.
3-9. A prior probability is one that exists before additional in-
formation is gathered. A posterior probability is one that can be
computed using Bayes Theorem based on prior probabilities and
additional information.
CHAPTER 3 D E C I S I O N A N A LY S I S 21
b. EMV (Sub 100) = 0.7(300,000) + 0.3(–200,000) 3-22. a. Expected value with perfect information is
= 150,000 1,400(0.4) + 900(0.4) + 900(0.2) = 1,100; the
maxi-
EMV (Oiler J) = 0.7(250,000) + 0.3(–100,000) mum EMV without the information is 900. Therefore,
= 145,000 Allen should pay at most EVPI = 1,100 – 900 = $200.
EMV (Texan) = 0.7(75,000) + 0.3(–18,000) b. Yes, Allen should pay [1,100(0.4) + 900(0.4) +
= 47,100 900(0.2)] — 900 = $80.
Optimal decision: Sub 100. 3-23. a. Opportunity loss table
c. Ken would change decision if EMV(Sub 100) is less
Strong Fair Poor Max.
than the next best EMV, which is $145,000. Let X = Market Market Market Regret
payoff for Sub 100 in favorable market.
Large 0 19,000 310,000 310,000
(0.7)(X) + (0.3)(—200,000) < 145,000 Medium 250,000 0 100,000 250,000
0.7X < 145,000 + 60,000 = 205,000 Small 350,000 29,000 32,000 350,000
X < (205,000)/0.7 = 292,857.14 None 550,000 129,000 0 550,000
The decision would change if this payoff were less than
292,857.14, so it would have to decrease by about $7,143. b. Minimax regret decision is to build medium.
EQUALLY CRIT. OF
MARKET MAXIMAX MAXIMIN LIKELY REALISM
Decision Row Row Row Weighte
Alternatives Good Fair Poor Maximu Minimu Average d
m m Averag
e
Small 50,000 20,000 — 50,000 —10,000 20,000 38,000
10,000
Medium 80,000 30,000 — 80,000 —20,000 30,000 60,000
20,000
Large 100,000 30,000 — 100,000 —40,000 30,000 72,000
40,000
Very Large 300,000 25,000 —160,000 300,000 —160,000 55,000 208,000
Payoff
Favorable Market (0.5)
$100,000
1
–$40,000
Unfavorable Market (0.5)
$30,000
$0
EMV for no clinic is $0
CHAPTER 3 D E C I S I O N A N A LY S I S 23
3-29. a.
Payoff
Favorable Market (0.82)
$95,000
CONSTRUCT 2 Unfavorable Market (0.18)
–$45,000
$69,800
1
Favorable Market (0.11)
$36,140 Unfavorable Market (0.89) $95,000
CONSTRUCT
3
–$45,000
$36,140 –$5,000
$30,000
$0
DO NOT CONSTRUCT
3-30.
Favorable Market
Large Shop
2 Unfavorable Market
No Shop
Favorable Market
Small Shop
3 Unfavorable Market
Favorable Market
Large Shop
4 Unfavorable Market
No Shop
Favorable Market
Small Shop
5 Unfavorable Market
Favorable Market
Large Shop
6 Unfavorable Market
No Shop
Favorable Market
Small Shop
7 Unfavorable Market
3-31.
a. EMV(node 2) = (0.9)(55,000) + (0.1)(–$45,000)
= 49,500 — 4,500 = $45,000
EMV(node 3) = (0.9)(25,000) + (0.1)(–15,000)
= 22,500 — 1,500 = $21,000
EMV(node 4) = (0.12)(55,000) + (0.88)(–45,000)
= 6,600 — 39,600 = –$33,000
EMV(node 5) = (0.12)(25,000) + (0.88)(–15,000)
= 3,000 — 13,200 = –$10,200
EMV(node 6) = (0.5)(60,000) + (0.5)(–40,000)
= 30,000 — 20,000 = $10,000
EMV(node 7) = (0.5)(30,000) + (0.5)(–10,000)
= 15,000 — 5,000 = $10,000
EMV(node 1) = (0.6)(45,000) + (0.4)(–5,000)
= 27,000 — 2,000 = $25,000
Since EMV(market survey) > EMV(no survey), Jerry should con-
duct the survey. Since EMV(large shop | favorable survey) is
larger than both EMV(small shop | favorable survey) and EMV(no
shop | favorable survey), Jerry should build a large shop if the sur-
vey is favorable. If the survey is unfavorable, Jerry should build
nothing since EMV(no shop | unfavorable survey) is larger than
both EMV(large shop | unfavorable survey) and EMV(small shop
| unfavorable survey).
CHAPTER 3 D E C I S I O N A N A LY S I S 25
Payoff
$45,000 Favorable Market (0.9)
Large Shop $55,000
2 Unfavorable Market (0.1)
$45,000 –$45,000
No Shop
–$5,000
$21,000 Favorable Market (0.9)
Small Shop $25,000
3 Unfavorable Market (0.1)
$25,000 –$15,000
1
–$33,000 Favorable Market (0.12)
Large Shop $55,000
4 Unfavorable Market (0.88)
–$5,000 –$45,000
No Shop
–$5,000
–$10,200 Favorable Market (0.12)
Small Shop $25,000
5 Unfavorable Market (0.88)
–$15,000
3-32.
Payoff
$8,500 2 (0.9)
A3 $500 $12,000
(0.1)
3 (0.9) –$23,000
$8,500 (0.1)
A4 $2,000
–$13,000
A5
–$3,000
$2,750
1
–$9,000 4 (0.4)
A3 –$7,000 $12,000
(0.6)
4
5 (0.4) –$23,000
–$3,000 (0.6)
A $2,000
–$13,000
A5
–$3,000
$4,500 6 (0.7)
A3 $500 $15,000
(0.3)
7 (0.7) –$20,000
$4,500 (0.3)
A4 $5,000
–$10,000
A5
$0
3-35. a.
3-36. a.
Payoff
Favorable Market
$95,000
Produce
Razor 3 Unfavorable Market
Survey –$65,000
Favorable
Do Not Produce Razor
–$5,000
1 Favorable Market $95,000
Produce
Razor 4 Unfavorable Market
Survey –$65,000
Unfavorable
Do Not Produce Razor
–$5,000
Favorable Market $80,000
Study Produce
Razor 5 Unfavorable Market –$80,000
Favorable Do Not Produce Razor –$20,000
Conduct
Pilot Favorable Market
Study 2 $80,000
Produce
6
Razor Unfavorable Market
Study –$80,000
Unfavorable
Do Not Produce Razor
–$20,000
Utility
0.85 Market Favorable (0.78)
0.95
Produce
3
Survey Razor Market Unfavorable (0.22)
0.5
Produce
0.78 Market Favorable (0.5)
Razor 1
7
Market Unfavorable (0.5)
0.55
Do Not Produce Razor
0.81
P(poor economy | prediction of 3-39. The expected value of the payout by the insurance com-
0.1(0.4) pany is
good economy) = 0.077
0.8(0.6) 0.1(0.4) EV = 0(0.999) + 100,000(0.001) = 100
The expected payout by the insurance company is $100, but the
P(good economy | prediction of policy costs $200, so the net gain for the individual buying this
poor economy) = 0.2(0.6) 0.25 policy is negative (–$100). Thus, buying the policy does not maxi-
0.2(0.6) 0.9(0.4) mize EMV since not buying this policy would have an EMV of 0,
which is better than –$100. However, a person who buys this pol-
P(poor economy | prediction of
icy would be maximizing the expected utility. The peace of mind
poor economy) = 0.9(0.6) 0.75 that goes along with the insurance policy has a relatively high util-
0.2(0.6) 0.9(0.4) ity. A person who buys insurance would be a risk avoider.
3-40.
Payoff Utility
U = 0.8118 Favorable Market (0.82)
$95,000 0.99
Construct
Survey Clinic 2 Unfavorable Market (0.18)
–$45,000 0
Favorable Do Not Construct Clinic
U = 0.76 (0.55) –$5,000 0.7
1 U = 0.1089 Favorable Market (0.11)
$95,000 0.99
Construct
Clinic 3 Unfavorable Market (0.89)
Survey –$45,000 0
Unfavorable
(0.45) Do Not Construct Clinic
– 0.7
$5,000
U = 0.55 Favorable Market (0.5)
1.0
Construct $100,000
Clinic 4 Unfavorable Market (0.5)
0.1
–$40,000
Do Not Construct Clinic
$0 0.9
0.6
Utility
0.4
0.2
0
EU(small plant | survey negative) = 0.27(0.5) + 0.73(0.10)
= 0.208 3-43. Selling price = $20 per gallon; manufacturing cost =
$12 per gallon; salvage value = $13; handling costs = $1 per
EU(no plant | survey negative) = 0.2 gallon; and advertising costs = $3 per gallon. From this informa-
EU(large plant | no survey) = 0.5(1) + 0.5(0.05) = 0.525 tion, we get:
EU(small plant | no survey) = 0.5(0.6) + 0.5(0.15) = marginal profit = selling price minus the manufacturing,
0.375 EU(no plant | no survey) = 0.3 handling, and advertising costs
EU(conduct survey) = 0.45(0.741) + 0.55(0.2565) = marginal profit = $20 — $12 — $1 — $3 = $4 per gallon
0.4745 If more is produced than is needed, a marginal loss is incurred.
EU(no survey) = 0.525
marginal loss = $13 — $12 — $1 — $3 = $3 per gallon
John’s decision would change. He would not conduct the survey
In addition, there is also a shortage cost. Coren has agreed to fulfill
and build the large plant.
any demand that cannot be met internally. This requires that
3-42. a. Expected travel time on Broad Street = 40(0.5) + Coren purchase chemicals from an outside company. Because the
15(0.5) = 27.5 minutes. Broad Street has a lower ex- cost of obtaining the chemical from the outside company is $25
pected travel time. and the price charged by Coren is $20, this results in
Expressway Broad 1
Street
30 Minutes, that has to be purchased from an outside company due to a
shor
U = 0.7 tage shortage.
Congestion (0.5) No
Congestion (0.5) 40 Minutes, cost
U = 0.2 =
$5
15 Minutes,
per
U = 0.9 gall
on
I
n
o
t
h
e
r
w
o
r
d
s
,
C
o
r
e
n
w
i
l
l
l
o
s
e
$
5
f
o
r
e
v
e
r
y
g
a
l
l
o
n
t
h
a
t
i
s
s
o
l
d
CHAPTER 3 D E C I S I O N A N A LY S I S 31
Decision Tree
(0.2) 500
–$1,000 = (500)(4) – (3)(1,000)
(0.3) 1,000
$2,500 = (1,000)(4) – (3)(500)
(0.4) 1,500 $6,000 = (1,500)(4)
Stock 2,000 (0.1) 2,000 $3,500 = (1,500)(4) – (5)(500)
(0.2) 500
$2,400 –$2,500 = (500)(4) – (3)(1,500)
(0.3) 1,000
$1,000 = (1,000)(4) – (3)(1,000)
(0.4) 1,500
$4,500 = (1,500)(4) – (3)(500)
(0.1) 2,000
$8,000 = (2,000)(4)
medium survey results, and a final table is used for high survey re- Decision Tree–Survey
sults. These tables are shown below. These probabilities will be used
in the decision tree that follows. L 450,000
For low survey results—A1: Small M
450,000
State of P(Bi) P(Ai | Bj) P(Bj and Ai) P(Bj | HL
Nature Ai) 450,000
B1 0.150 0.700 0.105 0.339
B2 0.400 0.400 0.160 0.516 150,000
B3 0.450 0.100 0.045 0.145 Medium M
650,000
P(A1) = 0.310
For medium survey results—A2: HL 750,000
M
HL –250,000
State of P(Bi) P(Ai | Bj) P(Bj and Ai) P(Bj |
Nature Ai) 350,000
B1 0.150 0.200 0.030 0.082 Large
B2 0.400 0.500 0.200 0.548 950,000
B3 0.450 0.300 0.135 0.370
P(A2) = 0.365
450,000
For high survey results—A3: Small M
HL 450,000
State of P(Bi) P(Ai | Bj) P(Bj and Ai) P(Bj |
Nature Ai) 450,000
B1 0.150 0.100 0.015 0.046 150,000
B2 0.400 0.100 0.040 0.123
$646,000 Medium M
B3 0.450 0.600 0.270 0.831 650,000
Medium (0.365)
P(A3) = 0.325
HL
Large 750,000
When survey results are low, the probabilities are P(L) =
0.339; P(M) = 0.516; and P(H) = 0.145. This results in –250,000
EMV(Small) = 450,000; EMV(Medium) = 495,000; and M
EMV(Large) = 233,600. 350,000
When survey results are medium, the probabilities are P(L) H
= 0.082; P(M) = 0.548; and P(H) = 0.378. This results in EMV 950,000
(Small) = 450,000; EMV(Medium) = 646,000; and EMV(Large)
L 450,000
= 522,800.
When survey results are high, the probabilities are P(L) = Small M
0.046; P(M) = 0.123; and P(H) = 0.831. This results in HL 450,000
EMV(Small) = 450,000; EMV(Medium) = 710,100; and
EMV(Large) = 821,000. 450,000
If the survey results are low, the best decision is to build the 150,000
medium facility with an expected return of $495,000. If the survey Medium M
results are medium, the best decision is also to build the medium HL 650,000
plant with an expected return of $646,000. On the other hand, if
the survey results are high, the best decision is to build the large 750,000
facility with an expected monetary value of $821,000. The ex- –250,000
pected value of using the survey is computed as follows: Large M
350,000
EMV(with Survey) = 0.310(495,000) + 0.365(646,000)
+ 0.325(821,000) = 656,065 H
950,000
Because the expected monetary value for not conducting the sur-
vey is greater (670,000), the decision is not to conduct the survey
and to build the medium-sized facility.
CHAPTER 3 D E C I S I O N A N A LY S I S 33
3-45. a.
Payoff
Succeed (0.5)
$75,000 $250,000
1
Succeed (0.6)
$140,000 $300,000
Mall 2
Succeed (0.75)
$250,000 $400,000
3
$0
First
Second
Decision
Decision
Point
Point
Payoff
SD (0.78)
Downtown SD (0.22) $220,000
2 SM (0.84)
–$130,000
Mall $270,000
A 3 SM (0.16)
SC (0.91) –$130,000
Circle $370,000
4 SC (0.09)
–$230,000
1 No Grocery Store
–$30,000
SD (0.27)
Downtown $220,000
SD (0.73)
5
SM (0.36) –$130,000
Mall $270,000
D B 6 SM (0.64)
–$130,000
SC (0.53)
Circle $370,000
7 SC (0.47)
–$230,000
No Grocery Store
–$30,000
SD (0.5)
Downtown SD (0.5) $250,000
8 SM (0.6)
–$100,000
Mall SM (0.4) $300,000
C 9 SC (0.75)
–$100,000
SC (0.25) $400,000
Circle
10
–$200,000
No Grocery Store
$0
3-46. a. Sue can use decision tree analysis to find the best solu-
information, Branch 10 (6–10) is a bad market given favorable in-
tion. The results are presented below. In this case, the best
formation, Branch 11 (5–7) is the decision to build the retail store
decision is to get information. If the information is favorable, she
given unfavorable information, Branch 12 (5–14) is the decision
should build the retail store. If the information is not favorable,
not to build the retail store given unfavorable information, Branch
she should not build the retail store. The EMV for this decision is
13 (7–12) is a successful retail store given unfavorable
$29,200.
information, Branch 14 (7–13) is an unsuccessful retail store given
In the following results (using QM for Windows), Branch 1
unfavorable information, Branch 15 (8–15) is a successful retail
(1–2) is to get information, Branch 2 (1–3) is the decision to not
store given that no information is obtained, and Branch 16 (8–16)
get information, Branch 3 (2–4) is favorable information, Branch
is an unsuccess- ful retail store given no information is obtained.
4 (2–5) is unfavorable information, Branch 5 (3–8) is the decision
to build the retail store and get no information, Branch 6 (3–17) is
the decision to not build the retail store and to get no information,
Branch 7 (4–6) is the decision to build the retail store given
favorable information, Branch 8 (4–11) is the decision to not build
given favor- able information, Branch 9 (6–9) is a good market
given favorable
CHAPTER 3 D E C I S I O N A N A LY S I S 35
b. The suggested changes would be reflected in Branches 4 and 5. The decision stays the same, but the EMV
increases to $46,000. The results are provided in the tables that follow:
Results for 3-46. a.
Start Endin Branch Profit Use Node Nod
Node g Probabilit (End Node) Branch Type e
Node y ? Valu
e
Start 0 1 0 0 Decision 29,200
Branch 1 1 2 0 0 Yes Chance 29,200
Branch 2 1 3 0 0 Decision 28,000
Branch 3 2 4 0.6 0 Decision 62,000
Branch 4 2 5 0.4 0 Decision —20,000
Branch 5 3 8 0 0 Yes Chance 28,000
Branch 6 3 17 0 0 Final 0
Branch 7 4 6 0 0 Yes Chance 62,000
Branch 8 4 11 0 —20,000 Final —20,000
Branch 9 6 9 0.9 80,000 Final 80,000
Branch 10 6 10 0.1 —100,000 Final —100,000
Branch 11 5 7 0 0 Chance —64,000
Branch 12 5 14 0 —20,000 Yes Final —20,000
Branch 13 7 12 0.2 80,000 Final 80,000
Branch 14 7 13 0.8 —100,000 Final —100,000
Branch 15 8 15 0.6 100,000 Final 100,000
Branch 16 8 16 0.4 —80,000 Final —80,000
c. Sue can determine the impact of the change by changing the probabilities and recomputing EMVs. This analysis
shows the decision changes. Given the new probability values, Sue’s best decision is build the retail store without
getting additional information. The EMV for this decision is $28,000. The results are presented below:
Results for 3-46. c.
Start Endin Branch Profit Use Node Nod
Node g Probabilit (End Node) Branch Type e
Node y ? Valu
e
Start 0 1 0 0 Decision 28,000
Branch 1 1 2 0 0 Chance 18,400
Branch 2 1 3 0 0 Yes Decision 28,000
Branch 3 2 4 0.6 0 Decision 44,000
Branch 4 2 5 0.4 0 Decision —20,000
Branch 5 3 8 0 0 Yes Chance 28,000
Branch 6 3 17 0 0 Final 0
Branch 7 4 6 0 0 Yes Chance 44,000
Branch 8 4 11 0 —20,000 Final —20,000
Branch 9 6 9 0.8 80,000 Final 80,000
Branch 10 6 10 0.2 —100,000 Final —100,000
Branch 11 5 7 0 0 Chance —64,000
Branch 12 5 14 0 —20,000 Yes Final —20,000
Branch 13 7 12 0.2 80,000 Final 80,000
Branch 14 7 13 0.8 —100,000 Final —100,000
Branch 15 8 15 0.6 100,000 Final 100,000
Branch 16 8 16 0.4 —80,000 Final —80,000
d. Yes, Sue’s decision would change from her original decision. With the higher cost of information, Sue’s decision
is to not get the information and build the retail store. The EMV of this decision is $28,000. The results are given
below:
Results for 3-46. d.
e. The expected utility can be computed by replacing the monetary values with utility values. Given the utility values in the prob-
lem, the expected utility is 0.62. The utility table represents a risk seeker. The results are given below:
Results for 3-46. e.
Start Endin Branch Profit Use Endin Node Nod
Node g Probabilit (End Node) Branch g Type e
Node y ? Node Valu
e
Start 0 1 0 0 1 Decision 0.62
Branch 1 1 2 0 0 2 Chance 0.256
Branch 2 1 3 0 0 Yes 3 Decision 0.62
Branch 3 2 4 0.6 0 4 Decision 0.36
Branch 4 2 5 0.4 0 5 Decision 0.1
Branch 5 3 8 0 0 Yes 8 Chance 0.62
Branch 6 3 17 0 0.2 17 Final 0.20
Branch 7 4 6 0 0 Yes 6 Chance 0.36
Branch 8 4 11 0 0.1 11 Final 0.1
Branch 9 6 9 0.9 0.4 9 Final 0.4
Branch 10 6 10 0.1 0 10 Final 0
Branch 11 5 7 0 0 7 Chance 0.08
Branch 12 5 14 0 0.1 Yes 14 Final 0.1
Branch 13 7 12 0.2 0.4 12 Final 0.4
Branch 14 7 13 0.8 0 13 Final 0
Branch 15 8 15 0.6 1 15 Final 1
Branch 16 8 16 0.4 0.05 16 Final 0.05
f. This problem can be solved by replacing monetary values with utility values. The expected utility is 0.80. The utility table
given in the problem is representative of a risk avoider. The results are presented below:
Results for 3-46. f.
Start Endin Branch Profit Use Node Nod
Node g Probabilit (End Node) Branch Type e
Node y ? Valu
e
Start 0 1 0 0 Decision 0.80
Branch 1 1 2 0 0 Chance 0.726
Branch 2 1 3 0 0 Yes Decision 0.80
Branch 3 2 4 0.6 0 Decision 0.81
Branch 4 2 5 0.4 0 Decision 0.60
Branch 5 3 8 0 0 Yes Chance 0.76
Branch 6 3 17 0 0.8 Final 0.80
Branch 7 4 6 0 0 Yes Chance 0.81
Branch 8 4 11 0 0.6 Final 0.60
Branch 9 6 9 0.9 0.9 Final 0.90
Branch 10 6 10 0.1 0 Final 0.00
Branch 11 5 7 0 0 Chance 0.18
Branch 12 5 14 0 0.6 Yes Final 0.60
Branch 13 7 12 0.2 0.9 Final 0.90
Branch 14 7 13 0.8 0 Final 0.00
Branch 15 8 15 0.6 1 Final 1.00
3-47. a. The decision table for Chris Dunphy along with the ex-
pected profits or expected monetary values (EMVs) for each alter-
native are shown on the next page.
38 CHAPTER 3 DECISION ANALYSIS
Expected profit:
Alternative Expected
Profit
1 119,500
2 135,500
3 131,500
4 144,500
5 141,500
6 145,000
7 151,500
8 151,000
9 155,500 best alternative
Expected
profit:
Alternative Expected Profit
1 117.100
2 131,500
3 126,300
4 139,700
5 133,900
6 136,200
7 140,700 best alternative:
8 138,600 stock 400,000 watches
9 138,700
CHAPTER 3 D E C I S I O N A N A LY S I S 39
Return in $1,000:
Event 1 Event Event Event Event 5
2 3 4
Probability 0.100 0.280 0.500 0.100 0.020
Alternative 1 100,000 110,000 120,000 135,000 140,000
Alternative 2 90,000 120,000 140,000 155,000 170,000
Alternative 3 85,000 110,000 135,000 160,000 175,000
Alternative 4 80,000 120,000 155,000 170,000 180,000
Alternative 5 65,000 100,000 155,000 180,000 195,000
Alternative 6 50,000 100,000 160,000 190,000 210,000
Alternative 7 45,000 95,000 170,000 200,000 230,000
Alternative 8 30,000 90,000 165,000 230,000 245,000
Alternative 9 20,000 85,000 160,000 270,000 340,000
Solution to 3-51a
Expecte Row Row
d Value Minimu Maximu
($) m ($) m ($)
Probabilities 0.2 0.8
Maint. No Maint.
Cost ($) Cost ($)
No Service Agreement 3,000 0 600 0 3,000
Partial Service Agreement 1,500 300 540 0 1,500
Complete Service Agreement 500 500 500 500 500
Column
3-52. Webest
can use QM for Windows to solve this decision 500
mak-
ing under uncertainty problem. We have made up probability val-
ues, which will be ignored in the analysis. As you can see, the
maximax decision is Option 4, and the maximum decision is Op-
tion 1. To compute the equality likely decision, we used equal
probability values of 0.25 for each of the four scenarios. As seen
below, the equally likely decision, which is the same as the EMV
decision in this case, is Option 3.
Solution to 3-52
Expected Row Row
Value ($) Minimum ($) Miximum
($)
Probabilities 0.25 0.25 0.25 0.25
Judge Trial ($) Court ($) Arbitration
($) ($)
Option 1 5,000 5,000 5,000 5,000 5,000 5,000 5,000
Option 2 10,000 5,000 2,000 0 4,250 0 10,000
Option 3 20,000 7,000 1,000 —5,000 5,750 —5,000 20,000
Option 4 30,000 15,000 —10,000 —20,000 3,750 —20,000 30,000
Payoff table
Laplace Hurwicz
Event 1 Event 2 Average Value Minimum Maximum Value
Alternative 1 0 0 0.0 0 0 0.00
Alternative 2 55,273 —10,000 22,636.5 —10,000 55,273 —2,819.97
Alternative 3 120,000 —15,000 152,500.0 —15,000 120,000 —150.00
Alternative 4 240,000 —30,000 105,000.0 —30,000 240,000 —300.00
Regret table
6 Month—Adopt the 6-month program: if a competitor’s product
Maximum is available at the end of 6 months, then copy; otherwise proceed
Alternative Event 1 Event 2 Regret with research and development.
Alternative 1 240,000 0 240,000 8 Month—Adopt the 6-month program: proceed for 8 months; if
Alternative 2 184,727 10,000 184,727 no competition at 8 months, proceed; otherwise stop development.
Alternative 3 120,000 15,000 120,000
Success or failure of development effort: Ok
Alternative 4 0 30,000 30,000
—Development effort ultimately a success
a. Sue Pansky is a risk avoider and should use the maximin No—Development effort ultimately a failure
decision approach. She should do nothing and not make an Column:
investment in Starting Right. S— Sales revenue
b. Ray Cahn should use a coefficient of realism of 0.11. R—Research and development expenditures
The best decision is to do nothing. E—Equipment costs
c. Lila Battle should eliminate alternative 1 of doing noth- I—Introduction to market costs
ing and apply the maximin criterion. The result is to invest in Market size and Revenues:
the corporate bonds.
Without With
d. George Yates should use the equally likely decision cri- Competitio Competitio
terion. The best decision for George is to invest in common n n
stock. S—Substantial (P = 0.1) $800,000 $400,000
e. Pete Metarko is a risk seeker. He should invest in com- M—Moderate (P = 0.6) $600,000 $300,000
mon stock. L—Low (P = 0.3) $500,000 $250,000
f. Julia Day can eliminate the preferred stock alternative
and still offer alternatives to risk seekers (common stock) Competition:
and risk avoiders (doing nothing or investing in corporate C6—Competition at end of 6 months (P = .5)
bonds). No C6—No competition at end of 6 months (P = .
5)
SOLUTIONS TO INTERNET CASES C8—Competition at end of 8 months (P = .6)
No C8—No competition at end of 8 months (P = .
Drink-at-Home, Inc. Case 4)
Abbreviations and values used in the following decision trees: C12—Competition at end of 12 months (P = .8)
No C12—No competition at end of 12 months (P = .
Normal—proceed with research and development at a normal 2)
pace.
42 CHAPTER 3 DECISION ANALYSIS
S (.1) – 80 = – 80
Mkt
S (.1)
50
C12 (.8) M (.6)
–50
L (.3)
Ok (.9) –100
S (.1)
450
Normal No C12 (.2) M (.6)
250
L (.3)
150
No (.1) (Stop)
–100
C8 (.6)
–80
S (.1)
10
8 Month Ok (.9) M (.6)
–90
No C8 (.4) L (.3)
–140
S (.1)
410
No (.1) M (.6)
210
L (.3)
110
S (.1)
C6 (.5) 60
M (.6)
–40
L (.3)
–90
S (.1)
50
C12 (.8) M (.6)
–50
L (.3)
Ok (.9) –100
S (.1) 450
No C6 (.5) No C12 (.2) M (.6)
250
L (.3)
150
No (.1)
–100
CHAPTER 3 D E C I S I O N A N A LY S I S 43
Mkt
S (.1) 50
C (.8) M (.6)
12 –50
–55 L (.3)
Ok (.9) –100
(4) S (.1)
450
Normal –6.4 240 M (.6)
250
No C12 (.2) L (.3)
150
No (.1) (Stop)
–100
C8 (.6)
–80
S (.1)
10
8 Month –74.2 Ok (.9) M (.6)
(–74.2) –90
–95 L (.3)
No C8 (.4) –140
S (.1)
410
No (.1) M (.6)
210
200 L (.3)
110
S (.1)
C6 (.5) 60
M (.6)
–40
–45 L (.3)
–90
S (.1)
50
C12 (.8) M (.6)
–50
–55 L (.3)
Ok (.9) –100
S (.1)
450
No C6 (.5) 240 M (.6)
(19.3) 250
No C12 (.2) L (.3)
150
No (.1)
–100
EXPECTED
POOR AVERAGE GOOD EXCELLENT VALUE
Probabilities 0.1 0.3 0.4 0.2
Option 1—PP —5,000 —2,000 2,000 5,000 700
Option 2—LB and PP —10,000 —4,000 6,000 12,000 2,600
Option 3—TR and PP —15,000 —10,000 7,000 13,000 900
Option 4—CC and PP —30,000 —20,000 10,000 30,000 1,000
Option 5—LB, CC, and TR —60,000 —35,000 20,000 55,000 2,500
result in a grade of B for the course. The table below gives the differ-
ent possibilities – points and grade in the course.
Case 1 Case 2 Case 3
on Exam on Exam on Exam EV Grade in Course
Study 1, 2, 3 12 B 12 B 12 B 12 B
Study 1,2 20 A 20 A 0B 40/3 A 2/3 chance or B 1/3 chance
Study 1,3 20 A 0B 20 A 40/3 A 2/3 chance or B 1/3 chance
Study 2,3 0B 20 A 20 A 40/3 A 2/3 chance or B 1/3 chance
Study 1 25 A 0B 0B 25/3 A 1/3 chance or B 2/3 chance
Study 2 0B 25 A 0B 25/3 A 1/3 chance or B 2/3 chance
Study 3 0B 0B 25 A 25/3 A 1/3 chance or B 2/3 chance
Thus, Raquel should study 2 cases since this will give her a
2/3 chance of an A in the course. Notice that this also has the
high- est expected value. This is a situation in which the values
(points) are not always indicative of the importance of the result
since 0 or 12 results in a B for the course, and 20 or 25 results in
an A for the course.