Questions

question 1 -

a) Acceptance sampling is a common quality control procedure to determine whether to accept

or reject a batch of products. The procedure is to take a random sample of size n from the batch
for a test, and each sampled unit is then classified as good or defective. The product batch will
be rejected if there are r or more defective units in the sample.

Kentwood Electronics just received a large batch of component parts from one of its suppliers
and tested a random sample of 10 items from this batch of component parts. The company
rejects a batch from a supplier if they find one or more defective components in the sample.
Assume that 1% of this batch of component parts are actually defective. What is the probability
that Kentwood Electronics will reject this batch?

b) KJS is an online beauty store. They would like to make personalized product
recommendations on the home page to visitors of different age groups. Management estimates
that 70% of the visitors are aged under 40 and 30% are aged 40 and over. Historical data
suggests that 10% visitors aged under 40 have browsing history in the “Anti-Ageing” section on
KJS’s website while 60% visitors aged 40 and over have browsing history in the “Anti-Ageing”
section. Suppose a customer is visiting KJS’s website and she has browsing history in the “Anti-
Ageing” section, what is the probability that she is aged 40 and over?

Question 2 -

The human resources manager of NexaCore, Inc., wants to predict the annual salaries of given
employees using the potential explanatory variables in the spreadsheet Q2_Salary in the
file Data.xlsx. The data include the following variables:

• Salary: Current annual salary (in dollars)

• Years Previous Experience: Number of years of relevant work experience prior to coming to
NexaCore.
• Years Employed: Number of years employed at NexaCore.
• Years Education: Number years of education beyond high school.
• Gender: 0 = Female, 1 = Male.
• Number Supervised: Number of employees supervised by this employee.
• Sales Department: 0 = Not in the sales department, 1 = In the sales department.

a) Estimate an appropriate multiple regression equation to predict the annual salary of a given
NexaCore employee using all the six explanatory variables: Years Previous Experience, Years
Employed, Years Education, Gender, Number Supervised, and Sales Department. According to
the estimated regression model, is there a difference between the mean salaries earned by male
and female employees at NexaCore? If so, how large is the difference?

b) Given the estimated regression model in part (a), predict the annual salary of a female
employee who served in another company for 10 years prior to coming to work at NexaCore.
This woman, a graduate of a four-year collegiate business program, has been supervising 12
subordinates in the purchasing department since joining the organization five years ago.

c) Add an interaction term between Years Employed and Sales Department, and run the
regression again. Does the inclusion of the interaction term improve the model’s goodness of fit?
Why or why not?
d) Given the estimated regression model in part (c), how do you interpret the coefficient for the
interaction term between Years Employed and Sales Department?

Question 3

BioView, a tech company specializing in MRI brain scan software, has developed advanced 3D
imaging software for brain tumor diagnosis and is planning to develop an operational version of
this software and market it. They were recently offered £250, 000 by another medical company to
purchase the software in its current form, along with the rights to further develop and market it.
BioView’s managers are deciding whether to accept this offer or reject it and continue developing
the software themselves.

If they reject the offer, BioView would need to invest £200, 000 to develop an operational version,
with a 60% chance of success. If development fails, they could create a less advanced software
expected to generate £300, 000 in revenue. If successful, BioView has two options for funding
and marketing the operational version:

1. Applying for a Entrepreneur Innovative Research (EIR) grant: There is a 75% chance of
winning the grant to develop and market the software.

2. Accepting the investment from a venture capital (VC) firm: The VC firm offered a larger
amount of investment than the EIR grant to finance and market the software in exchange for 30%
of future revenue of the operational version of the software.

Due to regulations of the EIR grant sponsor, BioView cannot receive the funding from both the
EIR grant and the VC firm. Moreover, due to some time constraint from the VC firm, if BioView
applied for the EIR grant, they would have to turn down the VC’s offer.

BioView’s managers estimate potential revenues for the operational software under three
scenarios: ”high revenue,” ”medium revenue,” and ”low revenue.” Table 1 provides the
probabilities and revenue estimates for each scenario, showing higher expected revenues with
VC funding (before deducting the VC’s 30% share) compared to the EIR grant.

BioView would like to make proper decisions to maximise its expected payoff, which is its
expected revenue minus all the expected cost.

a) Construct a decision tree to model BioView’s decision problem. Your tree should clearly
display decision nodes, event nodes and terminal nodes as well as any cash flows and
probabilities associated with the nodes and branches. Possible actions and outcomes should be
clearly labelled on the tree. At each terminal node in the tree you should include the total cash
flow for the path leading to that terminal node. You do not need to solve the tree in this part.

b) Should BioView apply for the EIR grant or accept the VC investment, if they continued their
own development of the operational version and succeeded? Why?

c) Should BioView accept the medical company’s offer of buying the software? Why?

Question 4 -

Coke is an essential fuel and reactant in the blast furnace process for steel production and is
produced from metallurgical coal. Steelco, a steel manufacturer in the United States, needs to
decide the procurement plan for metallurgical coal to be used next year. The company has
received five bids from different coal mining companies. The relevant information of each bid is
summarised in Table 2. For example, Ames Mining Co. bids to supply up to 600 ktons (1 kton =
1,000 tons) metallurgical coal at a price of 150 thousand pounds per kton (i.e., £150 per ton).
The coal supplied by Ames has 16% volatile matter. Ames is a union (United Mine Workers)
mine and the delivery mode of its coal is by rail.

Steelco plans to procure at least 1,400 ktons of metallurgical coal for next year. However, its
transportation capacity is limited to 700 ktons by rail and 850 ktons by truck next year. Moreover,
to hedge against adverse labour relations, Steelco also needs to procure at least 50% of its
metallurgical coal from union mines (i.e., coal procured from union mines must exceed that from
non-union mines). Steelco must decide the amount of coal to contract from each supplier next
year in order to minimise its total procurement cost of metallurgical coal.

a) Formulate a linear programming model for Steelco’s problem. Be sure to clearly specify your
decision variables, objective function, and constraints.

b) Solve the problem using Excel Solver. What is the optimal procurement plan of Steelco? What
will be Steelco’s optimal total procurement cost?

c) Suppose Steelco is able to increase its capacity for transporting metallurgical coal by truck
next year by 100 ktons at a cost of 5 thousand pounds per kton. Do you think Steelco should
expand its capacity for transporting metallurgical coal by truck next year? Why?
d) Due to the requirements of steel making process, the average proportion of volatile matter of
all the procured metallurgical coal must be at least 20%. In this case, how would you adjust your
linear programming formulation in part (a)? What is the new optimal procurement plan of
Steelco?