Sta 3113 Revision Questions
Sta 3113 Revision Questions
Sta 3113 Revision Questions
The Management of Unga Ltd is faced with the decision whether to build a new plant,
expand the existing plant or subcontract some of the operations of the company in
order to meet the increasing demand of the company’s product. To build a new plant
will cost 20 million shillings, expanding the existing plant will cost 10 million shillings
while subcontracting some of the operations of the company will cost 7 million shillings.
The future demand of the company’s products is expected to be high, moderate or low
with probabilities of 0.2, 0.5 and 0.3 respectively. The projected sales revenue under
the various options is given in millions in the table below;
State of Demand
Decision High Moderate Low
Build 70 40 15
Expand 40 30 25
Subcontract 30 20 25
Before making the decision, Unga Ltd may decide to contract a market
consultant to conduct a market survey on the future prospects of the company’s
products. The market survey will cost the company 2 million shillings, and the
outcome of the survey will either be acceptable or unacceptable with
probabilities of 0.4 and 0.6 respectively. If the outcome of the survey is
acceptable, the probabilities of high, moderate or low demand would be 0.7, 0.2
and 0.1, respectively. If the outcome of the survey is unacceptable, the
probabilities of high, moderate or low demand would be 0.1, 0.3 and 0.6
respectively, and Unga ltd will not build a new plant.
Draw a decision tree for this problem, and calculate the expected monetary value
of each decision. Hence, recommend an appropriate decision to Unga ltd
(10
Marks)
Question 7 (See example 12.1-4 Experimental Data in Decisions Under Risk)
Kaluworks Ltd produces aluminium sufurias in lots of fixed sizes. Occasionally,
malfunctions in the production process occur. As a result, bad lots with an unacceptable
number of defectives may be produced. Past experience indicates that the likelihood of
producing a bad lot is 0.04. The Operations manager has been considering this
probability as sufficiently small to warrant a random selection of the lots prior to
delivering to customers. However, he received a complaint recently from one of the
customers that some of the lots of sufurias he received had some defective sufurias. He
has now asked the Quality Control Department to handle the complaint and ensure that
it does not occur again. The Quality Control Department decides to employ a method
whereby they will be sampling some sufurias from lots ready for delivery and test them.
As a first step, they decide to be testing a sample of two items from each lot. Suppose
the defectives in a good lot is 5%, while a bad lot has 16% defective items;
(i) Work out the posterior probabilities and comment on these probabilities.
(10 Marks)
(ii) Suppose Kaluworks Ltd ships sufurias to two customers, Tuskys Supermarket
and Uchumi Supermarket. The contracts specify that the percentage of
defectives for Tuskys and Uchumi should not exceed 5 and 8, respectively. A
penalty of KShs. 8,000 is incurred per percentage point above the maximum
limit. On the other hand, supplying better quality lots will cost the manufacturer
KShs. 6,400 per percentage point. How should the manufacturer decide where to
deliver an inspected lot? (10 Marks)
Question 8 ( See example 12.4.2 & Qn 7 b Study Text)
a) Distinguish between a pure strategy game and a mixed strategy game (2 Marks)
b) Consider the following game;
B
Question 9 (To answer this qn, you need to understand example 12.1-1 notes)
a) The JKUAT preventive maintenance policy requires making decisions about when a
machine should be serviced on a regular basis in order to minimize the cost of
sudden breakdown. The time horizon is specified in terms of equal time periods, and
thus, the decision entails determining the optimal number of periods between two
successive maintenances. If preventive maintenance is applied too frequently, the
maintenance cost will increase while the cost of sudden break down will decrease.
Therefore, a compromise between the two cases calls for balancing the costs of
preventive maintenance and sudden breakdown.
The decision situation can be summarized as follows; A machine in a group of n
machines is serviced when it breaks down. At the end of T periods, preventive
maintenance is performed by servicing all n machines. The decision problem is to
determine the optimum T that minimizes the total cost per period of servicing
broken machines and applying preventive maintenance.
Let Pt be the probability that a machine would breakdown in period t, and nt be the
random variable representing the number of broken machines in the same period.
Further, assume c1 is the cost of repairing a broken machine and c2 the preventive
maintenance cost per machine. Suppose c1=1000/= and c2=100/=, and n=50. The
probabilities are tabulated below;
T 1 2 3 4 5
Pt 0.05 0.07 0.10 0.13 0.18
Determine, using the expected value-variance criterion, the optimal number of time
periods after which preventive maintenance should be applied (10 Marks)
Dr. Joel Chelule is considering investing in farm machinery. There are two models
available in the market i.e. Model A and model B. The following details relate to
each of the models;
Model A
Model B
i) Dr. Chelule can commission a survey in the next one year in order to
obtain more information on the viability of the investment. The outcome
of the survey would lead to either undertaking or abandoning the
investment with a probability of 0.6 and 0.4, respectively.
ii) The cost of purchase of each model and the corresponding resultant
profitability remain unchanged to the time of completion of the survey but
the revised probabilities of high or low demand would be 0.8 and 0.2,
respectively, irrespective of the model purchased.
iii) The estimated useful life of both models is 5 years from now with an
estimated present value at disposal of 700, 000/= for model A and 150,
000/= for model B.