Representation of Individual Decision Chapter 4

IMS 555

4
REPRESENTATION OF
INDIVIDUAL DECISIONS

CHAPTER OVERVIEW

REPRESENTATION OF INDIVIDUAL DECISIONS

Alternatives

Outcomes and States of Nature

Decision Matrices

Information about States of Nature

1|decision theory
 LEARNING OBJECTIVES:

After studying this chapter, you should be able to:

1. Describe the alternatives in decision making

2. Consider some possible outcomes and states of nature pertaining to
alternatives chosen
3. Transform alternatives and states of nature into matrix
4. Realize information about to which degree the various states of nature in that
matrix are supposed to obtain.

4.1 INTRODUCTION

This chapter discusses the methodology of presenting alternative or individual

decision into matrix and utility. Prior to this presentation, students should digest the
issues of effect or outcomes and states of nature of decision. With this digestion,
students will be able to easily transforming the individual alternative into decision
matrix and utilities that later will be used in making decision. This chapter also
discusses the natures of decision whether it is certainty, uncertainty, risk and
ignorance and the relationship of knowledge associated with those nature.

4.2 ALTERNATIVES

In decision making, alternatives are set of actions available for consideration that
lead to solving a problem. Decision makers choose an alternative among these sets
that he/she rank as best fit his preference.

Alternatives can be open or closed. In open decision, the alternative is unlimited

since a new alternative can be invented and discovered at any time along the
problem formulation. Open alternatives is associated with non structured or ill-
structured problem. In human daily life, the nature of problem is commonly both open
and closed alternatives.

However, in DT, the alternative is assumed to be closed. By closed alternative, no

new alternative can be added and introduced to the set of alternatives. A typical
example is decision to come to lecture either by car or motorcycle. The reason for
 Representation of Individual Decision Chapter 4
IMS 555

this is that closure makes decision problems much more accessible to theoretical
treatment. If the alternative set is open, a definitive solution to a decision problem is
not in general available.

Furthermore, the alternatives are commonly assumed to be mutually exclusive, i.e,

such that no two of them can both be realized. The reason for this can be seen from
the following scenario. A person could not decide the brand of handphone to buy
since all are good. He/she likes Brand A but Brand B is also interesting in some
features. He/she opts to have one brand only, without realizing that he/she has a
mutually exclusive option that is buying both brands. The three alternatives in this
scenario are mutually exclusive, since no two of them can be realized. This way of
representing the situation is more elaborate and more clear, and is preferred in
decision theory. Hence, in decision theory it is commonly assumed that the set of
alternatives is closed and that its elements are mutually exclusive.

4.2 OUTCOMES AND STATES OF NATURE

Any decision that decision maker made has an effect or outcome. The effect yielded
much depends on the (1) choice and method of dealing with it how we carry it
through, and (2) external factors beyond decision-maker's control. The external
factors can be known and unknown. The known factors are background information
that the decision-maker has. The unknown depend on reflection of others and
environment.

To illustrate the outcome and known/unknown factors, let consider the example of a
decision “vacation to Cameron Highland”. The outcome of this holiday trip would be
satisfaction and relaxation since the ultimate goal of any vacation is mind and body
rest. However, satisfaction level of this “vacation” will depend on what will happen
during the holiday trip. If the weather is good, hotel room is available (environmental
factors) and treatment from hotel staff is excellent (human factor), then the
satisfaction and relaxation would be gained. This can go vice versa.

In DT, common practice is to summarize the variety of external and extraneous

unknown factors into a number of cases known as state of nature. In the given
example above, state of nature can be further explained. While satisfaction and
relaxation are the outcomes, state of nature can be varied. The satisfaction is much
depend on weather is good, hotel rooms are available and good treatment. However,
the weather can be worse; hotel can be fully occupied and bad treatment from staffs.
The cases of good weather or worse weather, room available or not available and

3|decision theory
 good treatment or bad treatment are all states of nature in a decision-theoretical
treatment. DT defines (possible) outcomes of a decision as the combined effect of a
chosen alternatives and state of nature that comes together with that alternative.

4.3 DECISION MATRICES

The representation of evaluation-choice routine in individual DT is commonly using

decision matrix. Decision matrix is a table or chart that allows decision maker to
systematically identify, analyze, and rate the strength of relationships between sets of
information. In decision matrix, all alternatives available to decision maker are
tabulated against the possible state of nature. Those alternatives and states of nature
are placed in rows and columns respectively. The evaluation-choice routine of
decision “vacation to Cameroon Highland” is as below;

Criterion Weight Alternative A Alternative B Alternative C

Transportation 3 Private Public Both
Accommodation 3 Tent Hotel Guest House
Activities 2 On the Ground Hotel Package Nil
Food 2 Cook Hotel Package Restaurant
Duration 1 On the Ground Hotel Package Nil

The textual values of alternatives can now converted into utilities as below;

Criterion Weight Alternative A Alternative B Alternative C

Transportation 3 3 (x3) = 9 2 (x3) = 6 1 (x3) = 3
Accommodation 3 1 (x3) = 3 3 (x3) = 9 2 (x3) = 6
Activities 2 3 (x2) = 6 2 (x2) = 4 1 (x2) = 2
Food 2 1 (x2) = 2 3 (x2) = 6 2 (x2) = 4
Duration 1 3 (x1) = 3 2 (x1) = 2 1 (x1) = 1
Total Rating 23 27 16

Notice that for each alternative in the rows and states of nature in the column, the
decision matrix assigns at least one outcome, if not many. The decision matrix helps
in decision analysis. However, to use decision matrix to help in decision analysis,
additional information is needed. On top of the matrix itself, decision maker needs

(1) information about how the outcomes [satisfaction and relaxation] are
valued, and
 Representation of Individual Decision Chapter 4
IMS 555

(2) information about to which of the states of nature [transportation services,

treatment from hotel, types and availability of activities and taste of food] will
be realized.

For ease of analysis, the values of outcomes in decision matrix are converted to
numerical values or utilities. In ordinary words, this exercise is said as “to assign
utilities to them (outcomes)”. Textual values of outcomes now being replaced by
utility values as above. Converting or assigning utility matrices as expression to the
problem is considerably exclusive in mainstream DT. This is due to that modern
decision-theoretic methods require numerical information that helps in computation.

4.4 STEPS IN DEVELOPING DECISION MATRIX

Development of a decision matrix should be following steps or processes;

4.5.1 Identify alternatives

Based on nature of problem, a (group of) decision maker identify all possible
alternatives and prepare list the alternatives across the top of the matrix.

4.5.2 Selection criteria

These key criteria may come from a previously prepared affinity diagram or
from a brainstorming activity. Make sure that members have a clear and
common understanding of what the criteria mean. Also ensure that the criteria
are written so that a high score for each criterion represents a favorable result
and a low score represents an unfavorable result. List the criteria down the
left side of the matrix.

4.5.3 Assign weights

If some decision criteria are equally important or more important than others,
review and agree on appropriate weights to assign (e.g., 1, 2, 3).

4.5.4 Design scoring system

Before rating the alternatives, the team must agree on a scoring system.
Determine the scoring range (e.g., 1 to 5 or 1, 3, 5) and ensure that team
members have a common understanding of what high, medium, and low
scores represent.

5|decision theory
 4.5.5 Rate the alternatives

For each alternative, assign a consensus rating for each decision criterion.
The team may average the scores from individual team members or may
develop scores through a consensus-building activity.

4.5.6 Total the scores

Multiply the score for each decision criterion by its weighting factor. Then total
the scores for each alternative being considered and analyze the results.

4.5 INFORMATION ABOUT STATES OF NATURE

To make DT more meaningful, utility matrices are combined together with various
types of states of nature information. By this combination, DM will be able to know
which state of nature will be best considered. In the given example, if the visitor to
Cameron highland definitely knows that “weather is good”, then the decision of
committing to the vacation trip becomes very simple. This kind of decision is called
“decision making under certainty” for DM knows and confident of the outcome of the
alternative. If the circumstance is vice versa, then the decision is uncertainty.

Uncertainty or non-certainty is usually divided into further categories, such as risk,

uncertainty, and ignorance.

4.6.1 Risk

Risk if each action leads to one of a set of possible specific outcome, each
outcome occurring with a known probability. The probabilities are assumed to
be known to the decision maker.

4.6.2 Uncertainty

Uncertainty if either action or both has as its consequence a set of possible

specific outcomes, but where the probabilities of these outcomes are
completely unknown or are not even meaningful."

4.6.3 Ignorance

Ignorance if decision is described even more parsimoniously by the set of all

its (foreseeable) consequences. The problem of ranking decisions amounts
therefore to a problem of ranking sets of these consequences.
 Representation of Individual Decision Chapter 4
IMS 555

However, in DT, decision problems are considerably fall between the categories of
risk and uncertainty, leaving out the certainty. In the given example, notice that DM’s
decision about visit and shopping much depend on the weather condition that is
uncertainty or unknown to the DM. The DM does not know the probability of weather
to get well as DM does also not know the probability of weather to get worst. The
probability of good and bad weathers can be 50-50 or 30-70 or 70-30 or any possible
ratio. In this case, there is common to consider the “uncertainty” to cover the situation
with partial knowledge of the probabilities. The relationship between information
about the states of nature and availability of knowledge situation is as following;

Certainty - deterministic knowledge

Risk - complete probabilistic knowledge
Uncertainty - partial probabilistic knowledge
Ignorance - no probabilistic knowledge

Based on this relationship, it is common to have phrases such as “decision making

under certainty”, “decision making under risk”, “decision making under uncertainty”
and “decision making under ignorance” in decision making realm and its literature

4.6 CONCLUSION

In evaluating an alternative of decision, DM should consider the outcomes or effect

that associated with that alternative. To ease the evaluation, the standard
representation of a decision consists of (1) information about various states of nature,
that later can be converted into (2) a utility matrices. For the case of decision-making
under risk, the standard representation includes a probability assignment to each of
the states of nature (i.e., to each column in the matrix).

POINT TO PONDER

Students should think about

 A decision matrix evaluates and prioritizes a list of options. The team first establishes
a list of weighted criteria and then evaluates each option against those criteria.

 Although differ in concept, states of nature sometime can also be thought as

outcome

 Decision matrix is a way decisions can be made easily, quickly and objectively. It
demonstrates how decision is reached and as reference point for checking why
certain decision was made.

7|decision theory
  A decision matrix evaluates and prioritizes a list of options. The decision maker
establishes a list of weighted criteria and then evaluates each option against those
criteria.

ACTIVITIES

Student should further explore;

1. Alternatives of a decision
2. Example of set of alternatives based on a problem
3. The reason why alternative is considered closed
4. The term states of nature with regard to decision making
5. The relationship between and impact of states of nature to outcome of a decision
6. Conceptual definition and concept of Decision Matrix (DM)
7. Criteria for selecting a solution
8. The relationship between states of nature and the type of decision that include certainty,
uncertainty, risk and ignorance
9. Categorical knowledge associated with decision type

STUDY QUESTIONS
PART A: DEFINITION
Please define the following terms:

1. Alternative/Set of alternatives
2. Closed and open alternatives
3. Nature/State of nature
4. Outcome/decision outcome
5. Decision matrix
6. Decision criteria
7. Solution
8. Certainty decision
9. Uncertainty decision
10. Decision under ignorance
11. Decision under risk

PART B: SHORT ANSWER

Answer the following questions:
1. Write out a problem. Trace all possible alternatives for the problem.
2. Present the outcomes and states of nature for all derived alternatives in question #1.
 Representation of Individual Decision Chapter 4
IMS 555

3. Discuss the decision matrix. Assign the alternatives and states of nature as in question
#2into decision matrix.
4. Assign utility to states of nature in question #3. How decision maker uses utilities in
making decision.
5. Discuss the decision making under certainty, uncertainty, risk and ignorance with
examples and highlight the major differences among them

FURTHER READING

Gravel, N., Marchant, T. & Sen A. (2009) Uniform Expected Utility Criteria for Decision
Making under Ignorance or Objective Ambiguity. Unpublished manuscript, IDEP,
Ghent University and Indian Statistical Institute. Online. http://decision.greghec.com/
downloads/gmsApril09V2.pdf
Hansson, S. O. (1994). Decision Theory : A Brief Introduction. Department of Philosophy
and the History of Technology, Royal Institute of Technology (KTH). Stockholm.
Liu, X. W. (2004). On the methods of decision making under uncertainty with probability
information, International Journal of Intelligent System, Vol. 19, pp.1217-1238.
Luce, R. D. & Raiffa, H. (1957) Games and Decisions: Introduction and Critical Survey New
York: Wiley.
Ragsdale, C. T. (2007) Spreadsheet Modeling & Decision Analysis: A Practical Introduction
to Management Science, 5th Edition, USA: Thompson South-Western
Tague, N. R. (2004) The Quality Toolbox, 2nd Edition, Milwaukee: ASQ Quality Press

9|decision theory