Decision analysis for management judgement by Goodwin & Wright
Chapter 2 - How people make decisions involving multiple objectives
Bounded rationality: the fact that the limitations of the human mind mean that people have to use
‘approximate methods’ to deal with most decision problems and, as a result
they seek to identify satisfactory, rather than optimal, courses of actions
- Approximate methods are seen as heuristics
- Fast and frugal heuristics: quick ways of making decisions, especially when time is
limited on the basis of your knowledge
- Decision makers’ have their own mental toolbox
Heuristics used for decisions involving multiple options
When a decision maker has multiple objectives the heuristic used will either be compensatory or
non-compensatory:
Compensatory strategy an option’s poor performance on one attribute is
compensated by good performance on others
o Involve more cognitive effort because the decision maker has the difficult task of making
trade-offs between improved performance on some attributes and reduced performance
on others
The recognition heuristic is used where people have to choose between two options. If one is
recognized and the other not, the recognized one is chosen
o Recognition heuristic can be useful when choices have to be made on how to rank
objects on some criterion
The minimalist strategy the decision maker first applies the recognition heuristic, but if
neither option is recognized the person will simply guess which the
best option is
o Choose a random attribute and decide, when that is not possible choose a second one
Take the last the same as the minimalist heuristic except that, rather than picking a
random attribute, people recall the attribute that enabled them to reach a
decision last time when they had a similar choice to make
o If none of the previous attributes work, they choose a random one
The lexicographic strategy identifying the most important attribute and selecting the alternative
which is considered to be best on that attribute
o Like the earlier heuristics, the lexicographic strategy involves little information processing
o Lexicographic strategy is non-compensatory; with deeper reflections, a decision maker
might have preferred an option that performed less well on the most important attribute
because of its good performance on other attributes
The semi-lexicographic strategy if the performance of alternatives on an attribute is similar,
the decision maker considers them to be tied and moves on
to the next attribute
o It violates a fundamental axiom of decision analysis that is known as transitivity
o Transitivity states that if you prefer A to B and B to C then you should also prefer
A to C
o In this strategy it is contra-dictionary because C is preferred if you compare them directly
1
,Elimination by aspect (EBA) the most important attribute is identified and a cut-off point, which
defines the boundary of acceptable performance on this attribute is
then established. Any alternative which has a performance falling
outside this boundary is eliminated
o The process continues with the second most important attribute and so on
o The major flaw is its failure to ensure that the alternatives retained are in fact, superior to
those which are eliminated
o The strategy is non-compensatory; the decision maker’s focus is thus on a single attribute
at a time rather than possible trade-offs between attributes
The strategies we have outlined so far are theories intended to describe how people make a decision
when they are face with a simultaneous choice between alternatives. The coming ones don’t.
Sequential decision making: satisficing decision makers stop searching as soon as they find an
alternative that is satisfactory
o This satisfactory options may not be the best available
o The key aspect of satisficing is the aspiration level of the decision maker which
characterizes whether an alternative is acceptable or not.
o Decision makers’ asperation levels may change during the search process as they develop
a better idea of what they can reasonably achieve
o Non-compensatory strategy
o Also useful when all alternatives are available simultaneously, but where you can’t
overlook them
Reason-based choice when faced with the need to choose, decision makers often seek and
construct reasons in order to resolve the conflict and justify their choice to
themselves and others
o Can lead to some unexpected violations of principles of rational decision making
o Choice should be invariant to the way decision is framed, but according to this theory it
does (which candidate to select, which candidate to reject
o Another principle of rational decision making is of independence of irrelevant
alternatives; you are comparing two things and when a third thing comes up you don’t
change to also compare that one, in this theory the third has an effect on the decision
o Final consequence is that if an option has some features that are only weakly in its favour,
or irrelevant, this can actually deter people from selecting that option
Factors affecting which strategies people employ
1. Time available to make the decision
2. The effort that a given strategy will involve
3. The decision maker’s knowledge about the environment
4. The importance of making an accurate decision
5. Whether or not the decision maker has to justify his or her choice to others
6. A desire to minimize conflict
Decision makers choose their strategies to balance the effort involved in making the decision against
the accuracy that they wish to achieve effort – accuracy framework
Other characteristics of decision making involving multiple objectives
- Decoy effects by creating a situation where one option is clearly better than another
option, the decision maker is presented with an easy comparison
o Asymmetric dominance; one of the deals clearly dominates the decoy, but the other does
not
2
, o Decoy is the attribute which causes the effect that you chose (the decoy one you don’t)
o Phantom decoys: these options which asymmetrically dominate a particular option but
then turn out to be unavailable
- Choosing by unique attributes the degree to which attributes are shared across alternatives
influences which alternatives are preferred
o Decision makers tend to place more importance on attributes which possess unique
rather than shared information
o Attribute-salience effect
- Emotion and choice apart from the effect of ambient mood on choice, the emotional or
affective impressions attached to particular choice alternatives have
also been shown to influence choice
- Justifying already-made choices
o When asked to make a choice, decision makers may be unsure what their preferences are
and their preferences can change over time
o Coherence shifts; where attribute evaluations change to provide support for the choices
that we have made over time
- Partitioning the total cost of an item changes preferences
o Display the shipping cost differently from the product cost for example
Summary on page 28
Chapter 3 - Decision involving multiple objectives: SMART
By splitting the problem into small parts and focusing on each part separately, the decision maker is
likely to acquire a better understanding of the problem than that which would be achieved by taking
a holistic view.
Also, by requiring a commitment of time and effort, analysis encourages the decision maker to
think deeply about the problem enabling a rationale, which is explicit and defensible, to be
developed
As a result the decision maker should be better able to explain and justify why a particular option
is favoured
Axioms: a set of generally accepted propositions or ‘a formalization of common sense’.
- If the decision maker accepts the axioms then it follows that the results of the analysis will
indicate how he or she should behave if the decision is to be made in a rational way
The analysis is therefore normative or prescriptive; it shows which alternative should be chosen if
the decision maker acts consistently with his or her stated preferences
The method explained here is normally applied in situations where a particular course of action is
regarded as certain or virtually certain to lead to a given outcome, so that uncertainty is not a major
concern of the analysis (there are exceptions)
The aim of our analysis is to enable the decision maker to gain an increased understanding of his or
her decision problem not one single best course of action
Terminology
Objective an indication of the preferred direction of movement (stating objectives include,
minimize and maximize)
Attribute used to measure performance in relation to an objective
Proxy attribute not directly related to the objective
For each course of action facing the decision maker we will be deriving a numerical score to measure
its attractiveness to him.
3
, - If the decision involves no element of risk and uncertainty we will refer to this score as the value
of the course of action
- Alternatively where the decision involves risk and uncertainty, we will refer to this score as the
utility of the course of action
Simple multi-attribute rating technique = SMART
Although SMART may not always capture all the detail and complexities of a decision, it can be an
excellent methods for illuminating the important aspects of the problem and how they relate to each
other. Which often is sufficient for a decision to be made.
Stage 1: identify the decision maker (decision makers)
Stage 2: identify the alternative courses of action
Stage 3: identify the attributes which are relevant to the decision problem (value tree useful!)
Stage 4: assign values to measure the performance of the alternatives on that attribute
Stage 5: determine a weight for each attribute
Stage 6: for each alternative, take a weighted average of the values assigned to that
alternative
Stage 7: make a provisional decision
Stage 8: perform sensitivity analysis
Further looking into some stages
Stage 3: identify the attributes which are relevant to the decision problem (value tree useful!)
The aim is to arrive at a set of attributes that will allow us to measure performance on a numeric
scale. However, the initial attributes elicited from the decision maker may be vague, and they may
therefore need to be broken down into more specific attributes before measurement can take place:
Value tree:
- Start constructing the tree by addressing the attributes which represent the general concerns of
the decision maker
- Break down the attributes chosen in more specific attributes that will make it easier to compare
- Sometimes you need to specify again
How do we judge whether a value tree is accurate and a useful representation of the decision maker’s
concern?
a. Completeness all the attributes which are of concern to the decision maker will have
been included
b. Operationality when all the lowest-level attributes in the tree are specific enough for
the decision maker to evaluate and compare them for the different
options (could also be a proxy attribute when you cannot find a way
to numerically measure an attribute)
c. Decomposability the attractiveness of an option on one attribute can be assessed
independently of its attractiveness on other attributes
d. Absence of redundancy when two attributes represent the same thing, one is
redundant and you need to get that one out
e. Minimum size attributes should not be decomposed beyond the level where they
can be evaluated and attributes which have the same result in all
options can be left out.
Stage 4: assign values to measure the performance of the alternatives on that attribute
In measuring these attributes our task will be made easier if we can identify variables to represent
the attributes. But for some attributes it is not easy to do that, there are two ways how this can be
done:
Direct rating about attributes which cannot be represented by easily quantifiable variables
4
