Order a plagiarism free paper now

Our professional writers are ready to do this paper for you

  

2500 words on given case

Due 15th of Jan.

BBA312 DECISION ANALYSIS FINAL ASSESSMENT Task brief & rubrics

Task

 Description: Individual task
 Submission format: Report
 Tasks you are required to perform

You are asked to develop and write a report to assess the following case study from El Mejicano

The directors of fast food chain of restaurants based in Barcelona named El Mejicano were considering whether to begin the promotion for their new
line of menus than originally planned. “I think we should go ahead with the price cuts,” Pedro Morales said. “After all, it couldn’t hurt! At the very worst, we’ll sell
these menus cheap for a little longer than we had planned, and on the other side we could beat Panchito to the punch. Panchito is the most important
competitor of El Mejicano in Barcelona and they are also a fast food chain specialized in Mexican food. “That’s really the question, isn’t it?” replied Ana Perales,
the marketing manager of the firm. “If Panchito really is planning their own menus promotion, and we start our promotion now, we would beat them to the
punch. On the other hand, we might provoke a price war. And you know what a price war with that company means. We spend a lot of money and resources
fighting with each other. There’s no real winner. We both just end up with less profit.”

Manuel Pereira, the finance manager for El Mejicano, piped up, “The consumer wins in a menus price war. They get to buy things cheaper for a while.
We ought to be able to make something out of that.”

Antonia Lozano, CEO for El Mejicano, looked at the finance manager thoughtfully. “You’ve shown good horse sense in situations like these, Manuel. How
do you see it?”

Manuel hesitated. He didn’t like being put on the spot like this. “You all know what the sales projections of menus are for the 4-week promotion as
planned. Ana Perales tells us to expect sales of 2 million euros. The objective is to gain at least 2 percentage points of market share of the fast food market in
Barcelona, but our actual gain could be anywhere from nothing to three points. Profits during the promotion are expected to be down by 10%, but after the
promotion ends, our increased market share should result in more sales and more profits.”

Pedro Morales, the sales manager, broke in. “That’s assuming Panchito doesn’t come back with their own promotion in reaction to ours. And you know
what our report is from Laura Valencia. She says that he figures Panchito is up to something.”

“Yes, Laura did say that. But you have to remember that Laura works for our advertising agent. Her incentive is to sell advertising. And if she thinks she
can talk us into spending more money, she will. Furthermore, you know, she isn’t always right. Last time she told us that Panchito was going to start a major
campaign, he had the dates right, but it was for a different menu line altogether.”

Antonia wouldn’t let Manuel off the hook. “But Manuel, if Panchito does react to our promotion, would we be better off starting it early?”

Manuel thought for a bit. If he were working at Panchito and saw an unexpected promotion begin, how would he react? Would he want to cut prices to
match the competition? Would he try to stick with the original plans? Finally he said, “Look, we have to believe that Panchito also has some horse sense. They
would not want to get involved in a price war if they could avoid it. At the same time, they aren’t going to let us walk away with the market. I think that if we
move early, there’s about a 30 % chance that they will react immediately, and we’ll be in a price war before we know it.”

“We don’t have to react to their reaction, you know,” replied Antonia.

“You mean,” asked Ana Perales, “we have another meeting like this to decide what to do if they do react?”

“Right.” “So,” Manuel said, “I guess our immediate options are to start our promotion early or to start it later as planned. If we start it now, we risk a
strong reaction from Panchito. If they do react, then we can decide at that point whether we want to cut our prices further.”

Ana Perales spoke up. “But if Panchito reacts strongly and we don’t, we would probably end up just spending our money for nothing. We would gain no
market share at all. We might even lose some market share. If we were to cut prices further, it might hurt profits, but at least we would be able to preserve what
market share gains we had made before Panchito’s initial reaction.”

At this point, several people began to argue among themselves. Sensing that no resolution was immediately forthcoming, Antonia Lozano adjourned the
meeting, asking everyone to sleep on the problem and to call her with any suggestions or insights they had.

Please answer the following questions:

1) Would you add a different alternative to the given decision- making alternatives? From your point of view, what other decision variables would you also
take into account in this important decision- making for el Mejicano?

2) Develop and draw a decision tree and an influence diagram for this decision analysis. What roles do the two diagrams play in helping to understand and
communicate the framework and structure of this decision- making to the team of directors?

3) Based on the information in the case, what are El Mejicano’s objectives in this decision-making?
4) Describe and assess the different risks associated to each alternative of the decision making and select your risk preference or tendency to choose a risky

or less risky option.

Your analysis should be thorough and include concepts, theories and methodologies studied in class

Formalities:

 Wordcount: between 2500- 3000 words
 Font: Arial 12,5 pts.
 Text alignment: Justified.
 The in-text References and the Bibliography have to be in Harvard’s citation style.

Submission: Week 13 – Via Moodle (Turnitin). Final assessment will be due on week 13 by the 17 th of January at 23.59

Weight: This task is a 60% of your total grade for this subject.

It assesses the following learning outcomes:

 Outcome 3: To apply the framework of effective decision- making within the business context of Risk Management
 Outcome 4: To critically appreciate the framework of problem analysis with the context of decision-making

Rubrics

Exceptional 90-100 Good 80-89 Fair 70-79 Marginal fail 60-69
Knowledge &

Understanding
(20%)

Student demonstrates
excellent understanding of
key concepts and uses
vocabulary in an entirely
appropriate manner.

Student demonstrates
good understanding of the
task and mentions some
relevant concepts and
demonstrates use of the
relevant vocabulary.

Student understands the
task and provides minimum
theory and/or some use of
vocabulary.

Student understands the task
and attempts to answer the
question but does not
mention key concepts or uses
minimum amount of relevant
vocabulary.

Application (30%) Student applies fully
relevant knowledge from
the topics delivered in
class.

Student applies mostly
relevant knowledge from
the topics delivered in
class.

Student applies some
relevant knowledge from
the topics delivered in
class. Misunderstanding
may be evident.

Student applies little relevant
knowledge from the topics
delivered in class.
Misunderstands are evident.

Critical Thinking
(30%)

Student critically assesses
in excellent ways, drawing
outstanding conclusions
from relevant authors.

Student critically assesses
in good ways, drawing
conclusions from relevant
authors and references.

Student provides some
insights but stays on the
surface of the topic.
References may not be
relevant.

Student makes little or none
critical thinking insights, does
not quote appropriate
authors, and does not
provide valid sources.

Communication
(20%)

Student communicates
their ideas extremely
clearly and concisely,
respecting word count,
grammar and spellcheck

Student communicates
their ideas clearly and
concisely, respecting word
count, grammar and
spellcheck

Student communicates
their ideas with some
clarity and concision. It
may be slightly over or
under the wordcount limit.
Some misspelling errors
may be evident.

Student communicates their
ideas in a somewhat unclear
and unconcise way. Does not
reach or does exceed
wordcount excessively and
misspelling errors are
evident.

Decision Trees and Influence
Diagrams

BBA312 – Decision Analysis; Fall Semester 2020/21
Paulo Azevedo; [email protected]

Week 6: 9 November – 15 November

CONTENT

• Revision of Week 5
• Decision Trees
• Influence Diagrams

2

WEEK 6 – COURSE CONTENT

This unit discusses how decision trees and influence diagrams
can be extremely useful in helping people to gain an
understanding of the structure of the problems which
confront them.
 Introduction to normative algebraic techniques in decision-

making;
 Overview of the use of statistical regularities to assist human

decision-making.

3

INTRODUCTION

4

Decision trees (DT) and influence diagrams (ID) allow for a better
understanding of the structure of complex problems – how are
decisions, variables or uncertainties, and outcomes connected.

DECISION TREES

5

• Useful to model multi-stage problems (decomposition)
• Multi-stage problems – a certain decision may lead to the need to

make another decision (often a future decision). For example, if a
company decides to build a new warehouse a new set of problems
will appear, such as location, size, etc. All of which should be taken
into account.

• DT can help to unveil the structure of the problem and important
interconnections, to promote critical thinking and new alternatives,
and to highlight the need for new information

BUILDING DECISION TREES

6

• Building – iterative process – original structure is likely to be
modified – not a clean sequential process

• Very large trees may be counterproductive – too complex models
are often useless

• Two main symbols:

➔Decision node
➔Chance node

(uncertainty)
Source:
Decision
Analysis for
Management
Judgment

BUILDING DECISION TREES

7

Example from Decision Analysis for
Management Judgment:
Develop a new product (food processing
machine) that can either be electric or
gas-powered. Initial estimates of the
probabilities of success and resulting
profits or losses are made. The first
version of the DT can be drawn.

BUILDING DECISION TREES

8

What about the possibility of
successfully modifying a
failed design?

BUILDING DECISION TREES

9

What about the possibility of
successfully modifying a
failed design?
A second decision stage is
added – this would only
come into play in case we
have a failed initial design.

DECISION TREES – OPTIMAL POLICY

10

The DT represents a set of policies or plans of action. What are these
possible plans of action in this food processor example?

11

The DT represents a set of policies or plans of action. What are these
possible plans of action in this food processor example? These are
simple plans of action, based on two sequential decisions at most:
• Electric machine / If it fails, modify design
• Electric machine / If it fails, abandon project
• Gas machine / If it fails, modify design
• Gas machine / If it fails, abandon project
• No development of new product.

DECISION TREES – OPTIMAL POLICY

12

Example: assuming profit/loss is the only relevant attribute, and that
attitude towards risk is neutral (one many projects of the company, for
example) – expected monetary value is appropriate criterion.
Rollback method is used to determine optimal plan of action (set of
decisions): branch by branch comparison of decisions from right to
left.

DECISION TREES – OPTIMAL POLICY

13

Rollback method – compare expected results of later decisions first:
• Electric / Modify => Expected profit = 0.3×6 – 0.7×7 = -$3.1m
• Electric / Abandon => Expected profit = -$3m

DECISION TREES – OPTIMAL POLICY

14

Rollback method – compare expected results of later decisions first:
• Electric / Modify => Expected profit = 0.3×6 – 0.7×7 = -$3.1m
• Electric / Abandon => Expected profit = -$3m

DECISION TREES – OPTIMAL POLICY

15

Rollback method – expected value corresponding to preceding
decision can now be computed:
• Electric => Expected profit = 0.75×10 – 0.25×3 = $6.75m

DECISION TREES – OPTIMAL POLICY

Irrelevant

16

Solution =>
P2 > P1
P3 > P4
P3 > P2 > P5

Couldn’t we have applied
the EMV criterion for all
possible plans of action at
once?

DECISION TREES – OPTIMAL POLICY

Plan 1

Plan 2

Plan 5

Plan 3

Plan 4

17

• After determining optimal solution, the analysis is not finished –
sensitive analysis, for instance, should be carried out.

• DT reflects the available information at the time of its conception –
new information and results may lead changes in perception and
estimates (probabilities and values).

DECISION TREES – OPTIMAL POLICY

18

The same procedure can be applied for a utility-based decision – in
this case expected utilities are computed and compared instead of
expected profits:

DECISION TREES – UTILITY

19

Rollback: a different,
less risky decision is
suggested.
What is this plan of
action?

DECISION TREES – UTILITY

20

When the number of possible (numerical) outcomes is infinite and can
be described by a continuous probability distribution, one can
approximate it by a discrete distribution (loss of information). An
important discretization method is called Extended Pearson-Tukey
approximation (EPT). This method is based on a rule that has been
found to approximate a wide range of distributions rather well:
• Point of the distribution which has a 95% chance of being exceeded – probability

of 0.185.
• Point of the distribution which has a 50% chance of being exceeded – probability

of 0.63.
• Point of the distribution which has a 5% chance of being exceeded – probability of

0.185.

CONTINUOUS PROBABILITY DISTRIBUTIONS

21

CONTINUOUS PROBABILITY DISTRIBUTIONS

95% chance of being exceeded – probability of 0.185 / 50% chance of being
exceeded – probability of 0.63 / 5% chance of being exceeded – probability of
0.185.

Source:
Decision
Analysis for
Management
Judgment

22

CONTINUOUS PROBABILITY DISTRIBUTIONS
Probability density function –
Height of NBA players (cm) Height of NBA players considering

a mean of 200 cm and a standard
deviation of 9 cm.

185.2 cm => p = 0.185
200 cm => p = 0.63
214.8 cm => p = 0.185

23

The use of EPT method should be limited to relatively symmetric
distributions with one peak (normal or close to normal, for example).
Why?

CONTINUOUS PROBABILITY DISTRIBUTIONS

INFLUENCE DIAGRAMS

24

• ID are a great tool for a proper definition of complex problems –
often used as first analysis step, intended to transform a “messy”,
unclear situation into a well defined problem, with clear goals,
decisions and variables.

• Often used as a method for eliciting DT: more intuitive graphical
representation, easier to revise and modify and less structurally
“rigid”.

INFLUENCE DIAGRAMS

25

Source:
Handbook of
Decision
Analysis

Source:
Decision
Analysis for
Management
Judgment

INFLUENCE DIAGRAMS TO DECISION TREES

26

An ID may be represented as a DT if there are no loops (“arrow
circles”). Example from Decision Analysis for Management Judgment:

Imagine that you are a businessman and you are considering making
electronic calculators. Your factory can be equipped to manufacture
them and you recognize that other companies have profited from
producing them. However, equipping the factory for production will be
very expensive and you have seen the price of calculators dropping
steadily. What should you do?

INFLUENCE DIAGRAMS TO DECISION TREES

27

Example:

INFLUENCE DIAGRAMS TO DECISION TREES

28

Example (partially represented DT):

INFLUENCE DIAGRAMS TO DECISION TREES

29

Another example:

Source:
Handbook
of Decision
Analysis

INFLUENCE DIAGRAMS TO DECISION TREES

30

Another example:

Source:
Handbook
of Decision
Analysis

INFLUENCE DIAGRAMS TO DECISION TREES

31

Another example:

Source: Handbook of Decision Analysis

WEEK 6 – RECOMMENDED READING

• Handbook of Decision Analysis, Parnell, Bresnick, Tani and
Johnson; John Wiley & Sons, Chapter 11

• Decision Analysis for Management Judgment, 4th ed., Goodwin
& Wright; John Wiley & Sons, Chapter 7

32

NEXT WEEK

Review.

33

Slide 1
Slide 2
Slide 3
Slide 4
Slide 5
Slide 6
Slide 7
Slide 8
Slide 9
Slide 10
Slide 11
Slide 12
Slide 13
Slide 14
Slide 15
Slide 16
Slide 17
Slide 18
Slide 19
Slide 20
Slide 21
Slide 22
Slide 23
Slide 24
Slide 25
Slide 26
Slide 27
Slide 28
Slide 29
Slide 30
Slide 31
Slide 32
Slide 33

Order your essay today and save 10% with the discount code ESSAYHELP