Anda di halaman 1dari 4

Narsee Monjee Institute of Management Studies

Decision Analysis and Modeling


Title: Decision Theory: I (Risk) MBA Core: Trimester III 2018

1. Suppose a grocer is faced with a problem of how many cases of milk to stock to meet tomorrow’s demand.
All the cases of milk left at the end of the day are worthless. Each case of milk is sold for Rs 8 and is
purchased for Rs 5. Hence, each case sold brings a profit of Rs 3 but if it is not sold at the end of the day,
then it must be discarded resulting in a loss of Rs 5. The historical record of the number of cases of milk
demanded is as follows:
Number of 13 14 15 16 17 18
cases of milk
demanded
Number of 5 10 20 30 25 10
times
demanded
What should be the optimal decision of the grocer concerning the number of cases of milk to stock?
Assuming that the grocer has a perfect knowledge, then what would be his expected profit?

2. A newspaper boy has the probabilities of selling a magazine as shown in table below:
Number of copies sold Probability
9 0.05
10 0.1
11 0.15
12 0.3
13 0.25
14 0.15
The cost of copy sold is Rs 30 and the sale price of the magazine is Rs 40. The unsold copies fetch a
salvage value of Rs 5 in the second-sale market. How many copies should be ordered?

3. As a fund-raiser for a student organization, some students have decided to sell individual pizzas outside
the Union on Friday. Each pizza will sell for $1.75 and costs the organization $0.75. Historical sales
indicated that between 55 and 60 dozen pizzas be sold with the probability distribution given below:

Dozen of pizzas 55 56 57 58 59 60
Probability 0.15 0.20 0.10 0.35 0.15 0.05
To maximize the profit contribution, how many pizzas should be ordered? Assume pizzas must be ordered by
dozen. What is the expected value of perfect information in this problem? What is the maximum amount the
organization would be willing to pay for perfect information?

4. A newsstand receives its weekly order of “Weekly News” magazine on Monday & cannot reorder. Each
copy costs Rs 15 and sells for Rs 20. Unsold copies may be returned the following week for Rs 4. When
the news stand runs out of copies and cannot supply a customer wanting, it estimates its “good-will” loss
at Rs 6 in future profits, figuring that the customer will take his business elsewhere for a couple of weeks,
on the average. Demand distribution is given below:
---------------------------------------------------------------------------------
Demand (No. of Copies) 10 15 20 25
---------------------------------------------------------------------------------
Probability .28 .35 .17 .20
---------------------------------------------------------------------------------
(a) Construct a payoff table and use it to determine the optimal number of copies to stock and the expected profit.
(b) Computed Expected Value of Perfect Information (EVPI).
5. Harry's Hardware does a brisk business during the year, but during Christmas, Harry's Hardware sells
Christmas trees for a substantial profit. Unfortunately, any trees not sold at the end of the season are totally
worthless. Thus, the number of trees that are stocked for a given season is a very important decision. The
following table reveals the demand for Christmas trees.
DEMAND FOR CHRISTMAS TREES PROBABILITY
50 0.05
75 0.1
100 0.2
125 0.3
150 0.2
175 0.1
200 0.05
Harry sells trees for $15 each, but his cost is only $6.
(a) How many trees should Harry stock at his hardware store?
(b) If the cost increased to $12 per tree and Harry continues to sell trees for $15 each, how many trees should
Harry stock?
(c) Harry is thinking about increasing the price to $18 per tree. Assume that the cost per tree is $6. It is
expected that the probability of selling 50, 75, 100, or 125 trees will be 0.25 each. Harry does not expect
to sell more than 125 trees with this price increase. What do you recommend?

6. Technico Ltd has installed a machine costing Rs 4 Lacs and is in the process of deciding on appropriate
number of spare parts required for repairs. The spare parts cost Rs 4000 each but are available only if they
are ordered now. In case the machine fails and no spares are available, the cost to the company of mending
the part is Rs 18000. The plant has estimated life of 8 years and the probability distribution of failures
during this time, based on experience of similar machines is as follows.
No of 0 1 2 3 4 5 6+
failures
Probability .1 .2 .3 .2 .1 .1 0
Ignoring any discounting for time value for money, determine following:
a) The cost table and the optimal choice on the basis of expected cost principle.
b) The regret table and the optimal choice on the basis of expected regret criterion
c) Find EVPI

7. Center City motors Sales has recently incorporated. Its chief asset is franchise to sell automobiles of a
major American manufacturer. CCMS’s general manager is planning the staffing of the leadership’s
garage facilities. From information provided by the manufacturer and from the nearby dealerships, he has
estimated the number of annual mechanic hours that the garage will be likely to need.

Hours 10000 12000 14000 16000


Probability 0.2 0.3 0.4 0.1
The manager plans to pay each mechanics $9.00 per hour and to charge customer $16.00. The mechanics will
work for 40 hour week and get annual 2 week vacation. Determine how many mechanics Center City should
hire? How much should Center City pay to get perfect information about the number of mechanics it needs?

8. Emily Scott, head of a small business consulting firm, must decide how many M.B.A.s to hire as full-time
consultants for the next year. (Emily has decided that she will not bother with any part-time employees.)
Emily knows from experience that the probability distribution on the number of consulting jobs her firm
will get each year is as follows:

Consulting jobs 24 27 30 33
Probability 0.3 0.2 0.4 0.1
Emily also knows that each M.B.A. hired will be able to handle exactly three consulting jobs per year. The
salary of each M.B.A. is $60,000. Each consulting job is worth $30,000 to Emily’s firm. Each consulting
job that the firm is awarded but cannot complete costs the firm $10,000 in future business lost.
(a) How many M.B.A.s should Emily hire?
(b) What is the expected value of perfect information to Emily?

9. The Writer’s Workbench operates a chain of word-processing franchises in college towns. For an hourly
fee of $8.00, Writer’s Workbench provides access to a personal computer, word-processing software, and
a printer to students who need to prepare papers for their classes. Paper is provided at no additional cost.
The firm estimates that its hourly variable cost per machine (principally due to paper, ribbons, electricity,
and wear and tear on the computers and printers) is about 85¢. Deborah Rubin is considering opening a
Writer’s Workbench franchise in Ames, Iowa. A preliminary market survey has resulted in the following
probability distribution of the number of machines demanded per hour during the hours she plans to
operate:
Number of machines 22 23 24 25 26 27
Probability 0.12 0.16 0.22 0.27 0.18 0.05
If she wishes to maximize her profit contribution, how many machines should Deborah plan to have? What is
the hourly expected value of perfect information in this situation? Even if Deborah could obtain a perfectly
accurate forecast of the demand for each and every hour, why wouldn’t she be willing to pay up to the EVPI for
that information in this situation?

10. A multinational company has installed a beverages vending machine in its premises. The imported
machine requires costly spare parts during repairs and maintenance. The spare parts cost Rs 4,000 each
but are available only when the machine is ordered. In case the vending machine has a breakdown and
the spares are not available, it will cost at least Rs 20,000 for repairs including downtime, and for alternate
arrangements. The vending machine has an estimated life of five years and the probability distribution of
failures every year over the five year period is as given in table below:

Number of 0 1 2 3 4 5
failures per
year
Probability 0.1 0.2 0.3 0.25 0.10 0.05
of failure
Determine (a) Decision on the basis of the EMV method (b) The regret table and the optimal choice on the
basis of the lowest expected regret criterion (EOL matrix) (c) EVPI

11. A TV dealer finds that the cost of holding a TV in stock for a week is Rs 50. Customers who cannot obtain
new TV sets immediately tend to go to the other dealers and he estimates that for every customer who
cannot get immediate delivery he loses an average of Rs.200. For one particular model of TV the
probabilities of demand of 0, 1, 2,3,4 and 5 TV sets in a week are 0.05, 0.10, 0.20, 0.30, 0.20 and 0.15,
respectively.
(a) How many televisions per week should the dealer order?
(b) Compute EVPI.
(c) The dealer is thinking of spending on a small market survey to obtain additional information regarding
the demand levels. How much should he be willing to spend on such a survey?

12. A manufacturer of designer jeans must decide whether to build a large factory or a small factory in a
particular location. The profit per pair of jeans manufactured is estimated at Rs 10. A small factory will
have an amortized annual cost of Rs 2,00,000 with a production capacity of 50,000 jeans per year. A large
factory will have an amortized annualized cost of Rs 4,00,000 with a production capacity of 1,00,000
jeans per year. Four levels of manufacturing demand are considered equally likely, namely, 10,000,
20,000, 50,000 and 1,00,000 pairs of jeans per year.
(a) Set up a payoff table for building a small factory and a large factory for manufacturing designer jeans.
(b) Compute the EMV for building a small factory and a large factory.
(c) What is the EVPI?
(d) Which would you choose to build, a small factory or a large factory? Why?

13. After buying a computer system, Sim Thomas must decide whether to purchase (1) a complete
maintenance (or service) policy at a cost of $ 500, which would cover all maintenance costs; (2) a partial
maintenance policy at a cost of $300, which would cover some of the costs of any maintenance; or (3) no
maintenance policy. The consequences, costs, and probabilities are given in the following table:

maintenance required not required

No Service Agreement $3,000 $0

Partial Service Agreement $1,500 $300

Complete Service Agreement $ 500 $ 500

Probabilities 0.2 0.8

(a) What do you recommend?


(b) If the probability of needing maintenance is 0.8 (instead of 0.2) and the probability of not needing
maintenance is 0.2, how does this change Sim’s decision?

Anda mungkin juga menyukai