Teratex, a textile company that has a productive experience in the foreign market of 25 years, mu
the contrary the purchase from an external supplier. The profits depend on the demand of the pro
Decision alternative
Demand low-utility
Manufacture 221
Subcontract 210
Buy 195
Probabilities Ʃ = 1 0.35
PART 1. DECISION TREES, EVPI and EVMI
According to the corresponding information in Table 1 and the Predicted Value of Perfect Informat
and Decision Trees, respond:
a. Use EVPI to determine if the company should try to get a better estimate of the demand.
b. A test market study of potential product demand is expected to report a favorable (F) or unfavor
LOW-UTILITY (0,35)
NODE 2
OW AVERAGE -UTILITY
(0,42)
LOW-UTILITY (0,35)
HIGH-UTILITY (0,23)
BUY
LOW-UTILITY (0,35)
�(𝑺𝒋/𝐄)=(�(𝑬/𝑺𝒋)�(𝑺𝒋))/(�(𝑬))
B. BAYES' THEOREM
P(F/low) = 0,3
P(F/low average) = 0,38
P(F/high medium) = 0,4
P(F/high) = 0,55
P (s1) = (0,35) P (s2) = (0,42) P (s3) = (0,23)
P(F/low) = 0,3
P(F/low average) = 0,38
P(F/high medium) = 0,4
P(F/high) = 0,55
FAVORABLE
Estate of nature Previous probabilities P(sj) Conditional probabilities P(F/sj)
LOW 0.35 0.3
LOW AVERAGE 0.42 0.38
HIGH 0.23 0.55
P (F)
P (s1) = (0,35) P (s2) = (0,42) P (s3) = (0,23)
P(D/low) = 0,7
P(D/ low average) = 0,62
P(D/ high medium) = 0,6
P(D/high) = 0,45
UNFAVORABLE
Estate of nature Previous probabilities P(sj) Conditional probabilities P(D/sj)
LOW 0.35 0.7
LOW AVERAGE 0.42 0.62
HIGH 0.23 0.45
P (D)
NOD
E4
MANUFACTURE
0,39
NODE 2
FAVORABLE
NODE 2
FAVORABLE
SUBCONTRAC NOD
E5
BUY
NOD
E6
NODE 1
MANUFACTURE NOD
E7
0,61
NODE 3
UNFAVORABLE
SUBCONTRAC NOD
E8
NOD
BUY E9
E=𝐄𝐕𝐌𝐈/𝐄𝐕𝐏𝐈
E=𝟎/𝟎 𝐗 𝟏𝟎𝟎 = 0% OF
𝐗 𝟏𝟎𝟎
EVMI = │EVcMI - EVsMI│
EVcMI= 254.00
EVsMI= 254.07
EVMI 0
CISION TREES, EVPI and EVMI
market of 25 years, must decide if it manufactures a new product in its main plant, or if on
the demand of the product. The table shows projected profits, in millions of dollars.
251 310
225 278
236 289
0.42 0.23
ue of Perfect Information (EVPI) theory, the Expected Value of Sample Information (EVMI)
of the demand.
avorable (F) or unfavorable (U) condition. The relevant conditional probabilities are:
ITY (0,35)
221
GE -UTILITY
42)
251
ITY (0,23)
310
LITY (0,35)
GE -UTILITY 210
42)
225
TY (0,23)
278
TY (0,35)
E -UTILITY
2) 195
TY (0,23)
236
289
S $ 254.07
P(D/low) = 0,7
P(D/ low average) = 0,62
P(D/ high medium) = 0,6
P(D/high) = 0,45
Joint probabilities P(F n sj) Later probabilities P(sj/F)
0.105 0.27
0.1596 0.41
0.1265 0.32
0.39
LE
Later probabilities
Joint probabilities P(D n sj) P(sj/D)
0.245
LOW-UTILITY (0,27) 0.40
0.2604 0.43
0.1035 0.17
OW AVERAGE -UTILITY
0.61(0,41)
NOD
E4
HIGH-UTILITY (0,32)
TURE
LOW-UTILITY (0,27)
LOW-UTILITY (0,27)
OW AVERAGE -UTILITY
NOD (0,41)
E5
HIGH-UTILITY (0,32)
LOW-UTILITY (0,27)
Y OW AVERAGE -UTILITY
NOD (0,41)
E6
HIGH-UTILITY (0,32)
LOW-UTILITY (0,40)
OW AVERAGE -UTILITY
(0,43)
CTURE NOD
E7
HIGH-UTILITY (0,17)
LOW-UTILITY (0,40)
OW AVERAGE -UTILITY
(0,43)
NTRAC NOD
E8
HIGH-UTILITY (0,17)
LOW-UTILITY (0,40)
LOW-UTILITY (0,40)
OW AVERAGE -UTILITY
(0,43)
NOD
Y E9
HIGH-UTILITY (0,17)
𝐗 𝟏𝟎𝟎 = 0% OF EFFECTIVENESS
77.35
105.42 254.07
71.3
73.5
94.5 231.94
63.94
68.25
99.12 233.84
66.47
221
251 261.78
310
210
237.91
225
278
195
241.89
236
289
221
249.03
251
310
210
228.01
225
225
278
195
236 228.61
289
261.78
254.00
249.03
249.03
Problem 2. DEC
ElectroCom, a company that manufactures electronic components for the introduction in its produ
supervision or if it buys it from an external supplier. The profits depend on the demand of the prod
Table 2. Decision pro
Decision alternative
Manufacture
Subcontract
Buy
Lease
Outsource
Probabilities Ʃ = 1
According to the corresponding information in Table 2 and the Predicted Value of Perfect Informat
e. Use EVPI to determine if the company should try to get a better estimate of the demand.
f. A test market study of potential product demand is expected to report a favorable (F) or unfavora
DECISION TREE
MANUFACTURE
SUBCONTRACT
NODE 1 BUY
LEASE
OUTSOURCE
NODE 4 = 202.9
EVPI= │EVcPI-EVsPI│
EVcPI = 202.9
EVPI= │202,9-202,9│ 0
BAYES' THEOREM
�(𝑺𝒋/𝐄)=(�(𝑬/𝑺𝒋)�(𝑺𝒋))/(�(𝑬))
P (s1) = (0,19) P (s2) = (0,21) P (s3) = (0,28) P (s4) = (0,32)
P(F/low) = 0,2
P(F/low average) = 0,2
P(F/high medium) = 0,35
P(F/high) = 0,5
FAVORABLE
Estate of nature Previous probabilities P(sj)
LOW 0.19
LOW AVERAGE 0.21
HIGH MEDIUM 0.28
HIGH 0.32
P (s1) = (0,19) P (s2) = (0,21) P (s3) = (0,28) P (s4) = (0,32)
P(D/low) = 0,8
P(D/ low average) = 0,8
P(D/ high medium) = 0,65
P(D/high) = 0,5
UNFAVORABLE
Estate of nature Previous probabilities P(sj)
LOW 0.35
LOW AVERAGE 0.42
HIGH MEDIUM 0.28
HIGH 0.23
FAVORABLE
NODE 1
UNFAVORABLE
NODE 4 200
NODE 5 203.09
NODE 6 204.85
NODE 7 131.49
NODE 8 199.37
NODE 9 188.68
NODE 10 196.24
NODE 11 198.97
NODE 12 130.12
NODE 13 196.09
EVcMI= 268.619
EVsMI= 202.9
EVMI = 65.719
Problem 2. DECISION TREES, EVPI and EVMI
ents for the introduction in its product catalog, must decide whether to manufacture a new product in its main plant, s
depend on the demand of the product. The table shows projected profits, in millions of dollars .
Table 2. Decision process for the commercialization of the product
States of nature
Demand High Medium -
Demand low-utility Demand low average - utility utility
173 183 195
181 192 207
183 197 207
125 128 131
188 192 198
0.19 0.21 0.28
Predicted Value of Perfect Information (EVPI) theory, the Expected Value of Sample Information (EVMI) and Decision
to report a favorable (F) or unfavorable (U) condition. The relevant conditional probabilities are:
PROBABILITY DEMAND
PROBABILITY DEMAND
Millions
Millions
Millions
Millions
Millions
s Node 6)
Millions
FAVORABLE
Conditional probabilities P(F/sj) Joint probabilities P(F n sj) Later probabilities P(sj/F)
0.2 0.038 0.11
0.2 0.042 0.12
0.35 0.098 0.29
0.5 0.16 0.47
P (F) 0.34
UNFAVORABLE
Later probabilities
Conditional probabilities P(D/sj) Joint probabilities P(D n sj)
P(sj/D)
0.8 0.28 0.31
0.8 0.336 0.37
0.65 0.182 0.20
0.5 0.115 0.13
P (D) 0.91
MANUFACTURE
SUBCONTRACT
0.34 NODE 2 BUY
LEASE
OUTSOURCE
MANUFACTURE
SUBCONTRACT
LEASE
OUTSOURCE
Millions
Millions
HIGHER NODE 4, NODE 5
Millions NODE 6, NODE 7 AND 204,85 NODE 6
NODE 8 =
Millions
Millions
Millions
Millions
HIGHER NODE 9, NODE
Millions 10 NODE 11, NODE 12 198,97 NODE 11
AND NODE 13 =
Millions
Millions
Millions
E =
Millions
𝐸𝑉𝑀𝐼/𝐸𝑉𝑃𝐼 X
Millions 100
E = 65,719/0 X 0
100 =
a new product in its main plant, subcontract it with company
s of dollars .
Demand High - utility
218
213
215
137
209
0.32
abilities are:
32.87
38.43
195.66
54.6
69.76
34.39
40.32
200.83
57.96
68.16
34.77
41.37
202.9
57.96
68.8
23.75
26.88
131.15
36.68
43.84
35.72
40.32
198.36
55.44
66.88
LOW (0,11) 173
250.7117
198.97
Problem 3. DEC
Teratextyl, a textile company that has a productive experience in the foreign market of 30 years, m
supplier. The profits depend on the demand of the product. The table shows projected profits, in m
According to the corresponding information in Table 3 and the Predicted Value of Perfect Information (EVPI) theory,
i. Use EVPI to determine if the company should try to get a better estimate of the demand.
j. A test market study of potential product demand is expected to report a favorable (F) or unfavorable (U) condition
k. What is the expected value of market research information?
l. What is the efficiency of the information?
DECISION TREE
MANUFACTURE
SUBCONTRACT
NODE 1 BUY
LEASE
OUTSOURCE
NODE 2 = 89.24
EVPI= │EVcPI-EVsPI│
EVcPI = 89.24
EVPI= │89,24-89,24│ 0
BAYES' THEOREM
�(𝑺𝒋/𝐄)=(�(𝑬/𝑺𝒋)�(𝑺𝒋))/(�(𝑬))
P (s1) = (0,30) P (s2) = (0,22) P (s3) = (0,25) P (s4) = (0,23)
P(F/low) = 0,22
P(F/low average) = 0,35
P(F/high medium) = 0,33
P(F/high) = 0,42
FAVORABLE
Estate of nature Previous probabilities P(sj)
LOW 0.3
LOW AVERAGE 0.22
HIGH MEDIUM 0.25
HIGH 0.23
P (s1) = (0,30) P (s2) = (0,22) P (s3) = (0,25) P (s4) = (0,23)
P(D/low) = 0,78
P(D/ low average) = 0,65
P(D/ high medium) = 0,67
P(D/high) = 0,58
UNFAVORABLE
Estate of nature Previous probabilities P(sj)
LOW 0.3
LOW AVERAGE 0.22
HIGH MEDIUM 0.25
HIGH 0.23
FAVORABLE
NODE 1
UNFAVORABLE
NODE 4 90.83
NODE 5 84.78
NODE 6 86.75
NODE 7 87.33
NODE 8 89.31
NODE 9 89.09
NODE 10 82.72
NODE 11 85.97
NODE 12 86.41
NODE 13 88.43
EVcMI= 118.1556
EVsMI= 89.24
EVMI = 28.9156
Problem 3. DECISION TREES, EVPI and EVMI
in the foreign market of 30 years, must decide if it manufactures a new product in its main plant, or if on the contrary
e table shows projected profits, in millions of dollars.
alue of Perfect Information (EVPI) theory, the Expected Value of Sample Information (EVMI) and Decision Trees, respond:
of the demand.
favorable (F) or unfavorable (U) condition. The relevant conditional probabilities are:
PROBABILITY DEMAND
LOW (0,30) 85
HIGH (0,23) 95
LOW (0,30) 78
HIGH (0,23) 89
LOW (0,30) 82
HIGH (0,23) 90
LOW (0,30) 83
HIGH (0,23) 91
LOW (0,30) 85
HIGH (0,23) 93
Millions
Millions
Millions
Millions
Millions
s Node 6)
Millions
nt of: 89.240.000
FAVORABLE
Conditional probabilities P(F/sj) Joint probabilities P(F n sj) Later probabilities P(sj/F)
0.22 0.066 0.20
0.35 0.077 0.24
0.33 0.0825 0.26
0.42 0.0966 0.30
P (F) 0.32
UNFAVORABLE
Later probabilities
Conditional probabilities P(D/sj) Joint probabilities P(D n sj) P(sj/D)
0.78 0.234 0.35
0.65 0.143 0.21
0.67 0.1675 0.25
0.58 0.1334 0.20
P (D) 0.68
MANUFACTURE
SUBCONTRACT
0.32 NODE 2 BUY
LEASE
OUTSOURCE
MANUFACTURE
SUBCONTRACT
LEASE
OUTSOURCE
Millions
Millions
HIGHER NODE 4, NODE 5
Millions NODE 6, NODE 7 AND 90,83 NODE 4
NODE 8 =
Millions
Millions
Millions
Millions
HIGHER NODE 9, NODE
Millions 10 NODE 11, NODE 12 89,09 NODE 9
AND NODE 13 =
Millions
Millions
Millions
E =
Millions
𝐸𝑉𝑀𝐼/𝐸𝑉𝑃𝐼 X
Millions 100
E = 28,9156/0 X 0
100 =
ts main plant, or if on the contrary the purchase from an external
cision Trees, respond:
25.5
19.14
89.24
22.75
21.85
23.4
17.82
82.94
21.25
20.47
24.6
18.7
85.75
21.75
20.7
24.9
18.7
86.28
21.75
20.93
25.5
19.14
88.28
22.25
21.39
LOW (0,20) 85
HIGH (0,30) 95
LOW (0,20) 78
HIGH (0,30) 89
LOW (0,20) 82
HIGH (0,30) 90
LOW (0,20) 83
HIGH (0,30) 91
LOW (0,20) 85
HIGH (0,30) 93
LOW (0,35) 85
HIGH (0,20) 89
LOW (0,35) 82
HIGH (0,20) 90
LOW (0,35) 83
HIGH (0,20) 91
LOW (0,35) 85
HIGH (0,20) 93
90.83
89.6468
89.09