Decision Making

1
10th ed.

• Decision Making

A General Overview
2
Why study decision making?

• -It is the most fundamental task performed by
  managers.
• -It is the underlying theme of the course. The
  techniques and tools employed in the text are
  intended to improve our decision making skills as
  managers.
• -It is assumed that a better understanding of
  decision making will lead to better decision by
  managers.

3
The Decision Process (Descriptive Model)

• Identify the Problem
• Specify objectives and the criteria for choosing
  a solution
• Develop alternatives
• Analyze and compare alternatives

• Select the bestalternative
• Implement the chosen alternative
• Monitor the results

4
(No Transcript)
5
Pareto Phenomenon

A vital few things are important for reaching
an objective or solving a problem.

80/20 Rule - 80 of problems are caused by
20 of the activities.
How do we identify the vital few?
6
Models

Mathematical
What are the pros and cons of models?
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
Systems Approach
The whole is greater than the sum of the parts.
Suboptimization
11
Causes of Poor Decisions (Contd)

• Suboptimization
• The result of different departments each
  attempting to reach a solution that is optimum
  for that department

12
Causes of Poor Decisions

• Bounded Rationality
• The limitations on decision making caused by
  costs, human abilities, time, technology, and
  availability of information

13

• Some Very Basic Approaches to
• Decision Making
• Based on
• Decision Theory

14
Decision Theory
Decision Theory represents a general approach to
decision making which is suitable for a wide
range of operations management decisions,
including
15
Decision Making Conditions

• Certainty Condition in which relevant
  parameters have known values
• Uncertainty Condition in which it is impossible
  to assess the likelihood of various future
  events
• Risk Condition in which certain future events
  have probable outcomes

16
Decision Making Under Certainty
Examples of a criterion
Break Even Analysis Decide to produce items if
they lead to a profit

17
Example BE analysis
18
Decision Making under Uncertainty
Examples of three criteria

• Maximin - Choose the alternative with the best of
  the worst possible payoffs
• Maximax - Choose the alternative with the best
  possible payoff
• Laplace - Choose the alternative with the best
  average payoff of any of the alternatives

19
Try to do homework problem 1(a,b,c) on page 229
in the textbook
20
Decision Making under Risk
Examples of two criteria

• Expected monetary value calculate the expected
  value (average or mean) of each alternative,
  choosing the alternative with the best expected
  value
• Expected value of perfect information calculate
  the difference between the expected payoff under
  certainty and the expected payoff under risk

21
Format of a Decision Tree
22
Try to do homework problems 2(b,a,c), 10, and 11
on pp. 229-232.
23
Solutions to Homework

• Partial solutions are on the slides that
• follow. Be sure to try the problems before
• you look at the solutions.

24
Solutions
1.
1. a. Maximax Expand 80 is the highest payoff Expand 80 is the highest payoff Expand 80 is the highest payoff
b. Maximin Worst payoffs Worst payoffs Worst payoffs
Do Nothing 50 best of the worst payoffs
Expand 20
Subcontract 40
c. Laplace Average Payoff
Do Nothing 55 Indifferent between Do Nothing
Expand 50 And Subcontract
Subcontract 55

Comments on the above Part (a) is simply the
highest return of all possibilities in the
table. Part (b) shows 50, 20, and 40 as the
worst return for each of the three alternatives.
Hence, the best of the worst is 50, and the
answer is Do Nothing. Note that Part (c)
simply finds the average for each alternative,
l. Leading to a tie between Do Nothing and
Subcontract. In a sentence or two, could you
find a method to break the tie?
25
Now that we have probabilities for High (.7) and
Low (.3) demand, our problem changes from
decision making under uncertainty (as in Problem
1) to decision making under risk (which allows
us to build a decision tree as shown below).
2.
Low
High
2 b.
Low
High
Low
act
High
2a. The following gives the Expected Values
(or Expected Payoffs) of the three alternatives
for the above decision tree (this
allows us to solve the above decision tree)
EV(Do Nothing) EVDN (.3) (50) (.7) (60)
57 EV(Expand) EVEXP (.3) (20) (.7)
(80) 62 EV(Sub-Contract) EVSC (.3) (40)
(.7) (70) 61Since EVEXP is the largest, we
cut the branches for Do Nothing and Subcontract,
and oursolution is to EXPAND. Note that EVEXP
is also our Expected Value (EVr) under
conditions of risk, which we will use in the
problem for finding the EV of perfect
information.
26
2.
2 b.
act
Note that we had three probability distributions
on the tree in the previous slide, and replaced
each distribution with its expected value (or
average), allowing us to simplify the tree as
shown above. Hence, as stated in the previous
slide, our solution to the tree was to Expand
since its payoff of 62 is larger than the
payoffs of the other two alternatives. Note the
cuts that we made on the above tree used the
symbol .
27
50
2c
Low
Do Nothing
.3
EVc (.3) (50) (.7) (80) 71
High
.7
Expand
We draw the above tree which
represents what we would do if we knew
for certain that demand was to be Low (upper
branch) or High (lower branch). The EV
of the above tree would have to be our Expected
Value under conditions of certainty.
Hence, EVPI EVc -
EVr 71 - 62 9, Or
Expected Value under conditions of
certainty EVc (.3) (50) (.7) (80) 71
Expected Value under conditions of risk EVr

-62 EVPI (Expected Value of Perfect
Information)
9
80
28
10.
.70 high
29
11.
EV5
30
In-Class Assignment (You will be asked in class
how this problem could be solved. However, you
need NOT solve this problem for class. We will
do it together during the lecture. All you
should do, at the minimum, is to read this
problem carefully, and go to the next slide.)
It is suggested that you make a hardcopy of this
slide and the next slide, and bring the hardcopy
to the next class.
31
(a) This is the structure of the decision tree
for the problem above. Try to complete this
tree (i.e., print out this slide and add the
payoffs and the probabilities).(b) For a
greater challenge, try to solve this tree using
the monetaryexpected value criterion, and
writea brief statement on what Joe shoulddo.
(c) Based on your answer in part (b), What is
the probability that Joe could go bankrupt if
bankruptcy occurs from a loss of 75,000? What
is the probabilityof bankruptcy for a loss of
150,000? For even a greater challenge, and
assuming that you were able to answer the above
questions, write a sentenceor two on how Joe
could do something unethical in order to get a
better return, if he later foundthe market test
to be unfavorable.
32
Discussion Problem Do not try to solve this
problem. Simply read through it afew times.
This problem will be solved in class as a
computer application. Please make a hardcopy of
this slide, and bring it to class. .
33
End of Session. See you during our next class
meeting.