Business Decision Analytics under Uncertainty, Fall 2021 Midterm Exam 1
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Business Decision Analytics under Uncertainty, Fall 2021
Midterm Exam 1 (October 20, 2021)
Discussion version, posted for all sections
Question 1 (40 points)
The manager of the fresh fruit section at Jersey Foods wants to order raspberries today for tomorrow’s needs. Since these raspberries are ripe, Jersey Foods will need to sell them tomorrow and then discard the items that remain unsold. (Note that this is a simplification of reality.) The manager estimates that Jersey Foods will sell 12, 13, 14, or 15 cases tomorrow. Cases consist of 6 smaller packages: for simplicity, we consider only entire cases as the unit to consider. Jersey Foods purchases the raspberries for $7 per case and sells them for $18 per case. The manager checked their recent records on daily raspberry sales: based on this information, she estimates the probabilities for selling 12, 13, 14, and 15 cases of raspberries tomorrow are 0.1, 0.3, 0.4, and 0.2, respectively.
1.1 Develop a decision analysis formulation of this problem, by identifying the decision alternatives, the states of nature, and display the payoff table.
1.2 How many cases of raspberries should be purchased, according to the maximin payoff criterion?
1.3 How many cases should be purchased according to the maximum likelihood criterion?
1.4 How many cases should be purchased according to Bayes’ decision rule?
1.5 The manager is a bit uncertain about the probability estimates for selling 13 cases and 14 cases. Reapply Bayes’ decision rule when the prior probabilities of selling 13 and 14 cases are (i) 0.2 and 0.5, (ii) 0.4 and 0.3, and (iii) 0.5 and 0.2.
1.6 What seems to be the most reasonable final decision, regarding the order volume?
Question 2 (20 points)
You developed a new software product to play high-quality video games on a computer. The software has unique features that you patented. However, it appears that some of these features were copied and used in a competitive software product recently released by MegaVideo Corporation. Therefore, you consider suing MegaVideo for patent infringement. With legal fees and other expenses, your cost of going to trial is expected to be $1 million. You think that you have a 60% chance of winning the case, in which case you would receive $5 million in punitive damages. If you lose the case, then you will receive nothing. Moreover, in the latter case, there is a 50% chance that the judge would also order you to pay for court expenses and legal fees incurred by MegaVideo implying an additional cost of $1 million. As an alternative to consider, MegaVideo has offered you $1.5 million to settle this case out of court.
2.1 Without drawing a decision tree, determine your Bayesian decision, by maximizing the expected payoff.
2.2 and 2.3 How would you decide if you were to use the two other decision paradigms discussed in class and used also in Question 1 above.
2.4 Comment on the outcome and state your own (personally preferred) decision.
Question 3 (40 points)
Disclaimer: this is a strongly simplified model with fictitious data, used for illustration only.
You decided to plan a healthy diet for yourself at minimum cost. The food types considered are shown below (with symbolic names), together with their unit amount cost and unit amount contribution to your dietary requirements. For simplicity, we assume that all foods can be portioned and consumed in arbitrary continuous amounts.
3.1 Develop a corresponding Excel model and solve it.
Food1 Food2 Food3 Food4 Food5 Food6 Nutrient requirement /week
Unit cost [$] 1.39 3.19 0.99 2.59 2.19 1.49
Nutrient1[mg] 10 60 3 40 45 20 min.: 2200 max.: 3000
Nutrient2[mg] 12 30 20 28 35 10 min.: 2800 max.: 3500
Nutrient3[mg] 50 40 7 30 20 18 min.: 1600 max.: 2000
Nutrient4[mg] 24 10 40 13 20 16 min.: 3000 max.: 4000
3.2 Is it possible (in principle) that your model has no solution? If yes, then how would you modify it to make it feasible?
2023-10-28