Case 1– All or some?

(Sampl)
There had been a growing uneasiness, lack of dedication and lethargic attitude - resulting in a escalating attrition rate
among the employees in the data-processing unit of ‘Finex’, the young and reputed consulting company, operating in
Bangalore. Probing into the matter, the HR manager, Ms. Lobo found that the chord of discontent was rising from the
disparity among the salaries of the employees in this department. The discrepancy between the highest and the lowest
monthly income was rather high - approximately Rs. 15,000. Ms. Lobo wanted to solve the matter from the roots,
which obviously called for understanding and evaluating the psychology of these employees, their attitude towards
Finex and other related issues. She was of the opinion that to make a thorough and intricate study, every employee in
this unit should be included – which she was convinced would be a mammoth task.

Being a large company spread all over Bangalore, Finex was proud of its large work-force. After a lot of convincing by
the head of the analysis unit that, Ms. Lobo decided that instead of including every employee in the data-processing
unit, a sample of optimum size should be chosen. It was now required to estimate the size of sample that should be
selected in order to yield an average monthly income, which shall be within Rs.500 around the actual average monthly
income 95% of the time.

Case 2 – How many?

The operations and maintenance manager, Mr. Olann, had just come out of his boss’ office after an arduous ordeal on
this busy Saturday morning at ‘Digitron’, the multinational computer manufacturing company. Corporate giants as
Digitron could very well boast of the ‘zero-error’ computers it manufactures. So large was the scale of operations that,
at any given time there would be 10,000 PCs ready for shipment.

It all started with the Customer Care Executive, who submitted the last month’s report of customer feedback to his
supervisor, as was his job – a large number of PCs were reported to be malfunctioning. This malfunctioning had a two-
fold impact - not only would the company incur a huge repair-expenditure, but also lose a large amount of precious
work hours. This was something Digitron could not afford to tolerate.

To identify the origin of the problem Mr. Olann decided to conduct a pilot survey. A simple random sample of 100 PCs
was selected without replacements. Technicians were deputed to examine the conditions of the PCs selected, and to
calculate the cost of needed repairs. The results of this inspection were as follows:
• sample average repair cost = Rs. 830
• sample standard deviation of repair cost = Rs. 110.

Mr. Olann, not satisfied with just a mere 100 PCs, wanted some additional PCs to be sampled and examined. Find this
extra number such that the estimate of the total cost of repairing all PCs (formed by pooling the observations in the
pilot and the planned sample) is to be within Rs. 100000 of the true total cost with probability 90%?

Case 3 – Forlorn Foreman (Quality Control)

Perfume bottles were coming down the conveyor belt at the far end of the factory. The foreman was standing at this
corner, looking rather tired and intensely bored with the monotony of his work. He was trying to assess the percentage
of the number of defective bottles and hoping that the number would be a small one!
He had taken some bottles (called a lot) off the conveyor belt, checked them (very lethargically) for any detectable
defects, and to his utter dismay, found that there were four defective bottles in the lot. Feeling quite apprehensive, he
proceeded to select four more lots, which gave him devastating results – five, one, three and two ‘defectives’!
He immediately reported the matter to his supervisor, who decided to stop the production and check for any technical
malfunctioning of the three machines responsible for the manufacturing of these bottles. A bottle can be manufactured
by any one of the machines.
(a) Usually the number of defects in a lot is quite a small no. What probability distribution is the no. of defects
likely to follow? Why?
(b) A bottle was found to be defective. The foreman wonders which machine the defective bottle had come from.
What technique of probability theory would help him answer his query? Give justification for your answer.

Case 4- Platinum Card (Market Segmentation)

♦ Mr. Prashant, the vice president of Citibank is busy discussing the feasibility of the Platinum Credit Card.

1
♦ The discussion obviously hovered around different aspects of the affluent people of the society - the whims of their
behaviour, their attitude towards spending, frequency of ‘big’ purchases, et cetera.
♦ Among the stentorian male voices, a mild female voice was heard - it was Ms. Minu, the resident statistician.
♦ Ms. Minu: “Are we really clear on the definition of ‘affluent’? Well, who is ‘poorest’ among these rich people?”
♦ A survey done last year for launching the ‘Elite Card’ revealed the display of income among the rich of the society.
It was decided that not more than 0.57 % of these ‘rich’ people should be considered for the ‘Platinum Card’;
under the apprehension that too many astronomical transactions would be difficult to handle immediately. The
number of such cards issued may be increased later, depending on their acceptability and the capacity of the bank.

Case 5– Recovery of loans (Finance)

As accounts manager in your company, you classify 75% of your customers as "good credit" and the rest as "risky
credit" depending on their credit rating. Customers in the "risky" category allow their accounts to go overdue 50% of
the time on average, whereas those in the "good" category allow their accounts to become overdue only 10% of the
time.
(a) What percentage of overdue accounts is held by customers in the "risky credit" category?

Case 6 - CD Craze (Sales Forecasting)

Astounded by the overwhelming sales of CDs of ‘Soft Rock’, the manager of a renowned music store realised that he
need to do something about the pressing demand of his ‘Soft Rock’ loving customers.
Procrastination in taking a decision about the number of such CDs he must immediately order, may lead to disgruntled
customers, who may eventually start patronising his competitor across the road.
After all it is the customers’ world! He quickly assembled data on the daily sales (in units) of ‘Soft Rock’ CDs sold
over the last few days.
(a) Over the next few days, while this surge is on, what is the minimum number of CDs he should keep in
store so that on any day no customer returns empty handed?
(b) He considers selling less than 200 CDs a day as a ‘bleak’ sale (too optimistic!). How many days in the next
month, would be a witness to these ‘bleak’ sales?