INSTITUTE OF ACCOUNTANCY ARUSHA
MODULE: OPERATIONS RESEARCH : AFU7437 / BMU07424 / SSU07404
TOPIC: DECISION ANALYSIS
EXERCISE FOR STUDENTS
1. Your company is testing a site for drilling for oil. You may hit a dry well, a small
oil well, or a large oil well. Consider the following payoff table (in $ thousands) for
your situation:
Dry Well Small Oil Large Oil
Well Well
Drill -5,000 1,000 6,000
Do not 0 0 0
Drill
a) What are the decision alternatives? What are the states of nature?
b) What would be the best decision under each of the following criteria?
i. Maximax
ii. Maximin
c) Consider now that you have the following expectations based on your analysis
of the site. You believe that there is a 50% chance of hitting a small well and a
20% chance of hitting a large well.Calculate the following:
i. Expected returns for both decisions (which one is better?)
ii. Expected return with perfect information.
iii. Expected value of perfect information
2. A man owns a pizza shop in a down town mi-mall. It is Saturday morning, and he
is trying to decide how many pizzas to make to meet today’s lunch hour demand.
Based upon his experience with Saturdays, he thinks that the probability of being
able to sell 20 pizzas is 0.2, of being able to sell 40 pizzas is 0.3, and of being
able to sell 50 pizzas is 0.5.
Suppose a pizza sells for TZS1,000 and has an incremental cost of TZS425. If
he has leftover pizzas, he can sell them to homeless shelter for TZS125 each. If
demand exceeds the number of pizzas he has prepared, every disappointed
customer costs him TZS25 worth of lost customer goodwill.
Required:
i. Define the decisions, alternatives, states of nature, outcomes, revenues, costs,
and payoffs for this problem. Then construct the payoff table.
ii. What type of the decision making situation based on the knowledge about the
state of nature?
iii. Assume he did not have the probability knowledge. Identify the decision he
would make using each of the following criteria: Maximax, Maximin, Realism
and Minimax Regret.
1
iv. Find the expected value (EV) for each alternative. Based upon the EV
criterion, what is the best decision? Also find the EV with perfect information
(or EV with PI) and the EV of perfect information (or EVPI).
v. Draw the decision tree that this problem calls forth and solve it.
3. General Motors (GM) is planning their production strategy for their next model.
Three alternatives are being considered for their model MX: 30,000, 20,000 and
12,000. GM decides to categorize the demand for MX for the next year as either
High (H) or Low (L). The payoffs measured in millions of dollars and probabilities
of states of nature are presented in the table below.
States of Nature
Decision Alternatives High (H) Low (L)
Produce 30,000 29 -12
Produce 20,000 18 8
Produce 12,000 3 11
Probabilities 0.62 0.38
For this problem, if we want to do a decision tree,
i. How many decision nodes are required?
ii. How many branches come out of each decision node?
iii. How many chance nodes are required?
iv. How many branches come out of each chance node?
v. Draw the decision tree. Label each branch completely including probabilities
and payoffs.
vi. Solve the decision tree and find the best production strategy.
4. John Alex, a business owner, is considering opening a second location of his
business and has secured a property at the Dar es Salaam city centre. He has
determined his options for leasing property as: commit to a two year lease at
TZS50,000,000 per year or commit to a one year lease at TZS60,000,000. At the
beginning of the second year, he can cancel the two year lease at a cost of
TZS20,000,000 or continue with the lease. If he only signed a one year lease
initially, he has the option of renewing the lease for another year at
TZS60,000,000.
Based on previous experience in the business and his research on the new
location, John Alex estimates there is a 40% chance that sales (net of cost of
goods sold) in the first year will be considered “good” and will amount to
TZS100,000,000; alternatively, they will only amount to TZS30,000,000. If sales
in year one were “good”, the chance of them being “good” in year two as well is
80%. However, if sales in year one were “bad”, the probability of them being
“bad” in year two as well is 90%. Good sales in year two also amount to
TZS100,000,000; and bad sales in year two also amount to TZS30,000,000.
Required:
Solve the decision tree and recommend to John Alex what he should do.
