Blake Electronics Case Study
Group 2 Module 2
Blake Electronics Case Study
(1) Does Steve need additional information from Iverstine and Walker
Our group consensus is that additional information is not needed. We believe that all quantitative information needed to form a decision tree is available in the problem
(2) What would you recommend? In progress… Subject to change based on final tree.
Our group recommends that Steve moves forward with the production of the Master Control Center residential product. We recommend for him to utilize the research services of Marketing Associates, Inc. (MAI). The revised Excel document(attached) shows that the EMV for MAI and Iverstine & Walker are identical ($1.2M). However, when you take into account the cost of each firm the best return on investment is to go with MAI. Of course, Iverstine & Walker have the ability to predict the success or failure of the master control boxes with a greater amount of certainty compared to MAI. Since it was stated in the case that Steve Blake is relatively conservative, perhaps additional analysis is needed to place a price tag on their more-certain information through an Expected Value of Perfect Information (EVPI) analysis. The EVPI may be an unnecessary step since it may not affect our decision
Work
Probability from Blake Internal
P(Successful venture) =
0.6
P(Unsuccessful venture) =
0.4
MAI Probabilities
P(Successful venture | Fav survey) =
0.7
P(Unsuccessful venture | Fav survey) =
0.3
P(Successful venture | Unfav survey) =
0.4
P(Unsuccessful venture | Unfav survey) =
0.6
Iverstine & Walker Probabilities
P(Fav survey | Succes) =
0.9
P(Unfav survey | Success) =
0.1
P(Fav survey | Failure) =
0.2
P(Unfav survey | Failure) =
0.8
Calculations for Posterior Probabilities (Bayes Theorem) included in decision tree
Posterior Probabilities for Iverstine & Walker if Survey is Favorable
Outcome
P(Fav survey | Outcome)
Prior Probability
Join Prob
Posterior
Joint Prob Equation
Posterior Prob Equation
Blake Electronics Case Study
(1) Does Steve need additional information from Iverstine and Walker
Our group consensus is that additional information is not needed. We believe that all quantitative information needed to form a decision tree is available in the problem
(2) What would you recommend? In progress… Subject to change based on final tree.
Our group recommends that Steve moves forward with the production of the Master Control Center residential product. We recommend for him to utilize the research services of Marketing Associates, Inc. (MAI). The revised Excel document(attached) shows that the EMV for MAI and Iverstine & Walker are identical ($1.2M). However, when you take into account the cost of each firm the best return on investment is to go with MAI. Of course, Iverstine & Walker have the ability to predict the success or failure of the master control boxes with a greater amount of certainty compared to MAI. Since it was stated in the case that Steve Blake is relatively conservative, perhaps additional analysis is needed to place a price tag on their more-certain information through an Expected Value of Perfect Information (EVPI) analysis. The EVPI may be an unnecessary step since it may not affect our decision
Work
Probability from Blake Internal
P(Successful venture) =
0.6
P(Unsuccessful venture) =
0.4
MAI Probabilities
P(Successful venture | Fav survey) =
0.7
P(Unsuccessful venture | Fav survey) =
0.3
P(Successful venture | Unfav survey) =
0.4
P(Unsuccessful venture | Unfav survey) =
0.6
Iverstine & Walker Probabilities
P(Fav survey | Succes) =
0.9
P(Unfav survey | Success) =
0.1
P(Fav survey | Failure) =
0.2
P(Unfav survey | Failure) =
0.8
Calculations for Posterior Probabilities (Bayes Theorem) included in decision tree
Posterior Probabilities for Iverstine & Walker if Survey is Favorable
Outcome
P(Fav survey | Outcome)
Prior Probability
Join Prob
Posterior
Joint Prob Equation
Posterior Prob Equation