02.03.2013 Views

Downloadable - About University

Downloadable - About University

Downloadable - About University

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

144 Decision trees and influence diagrams<br />

Decision trees can also help a decision maker to judge the nature of the<br />

information which needs to be gathered in order to tackle a problem and,<br />

because they are generally easy to understand, they can be an excellent<br />

medium for communicating one person’s perception of a problem to<br />

other individuals.<br />

The process of constructing a decision tree is usually iterative, with<br />

many changes being made to the original structure as the decision<br />

maker’s understanding of the problem develops. Because the intention<br />

is to help the decision maker to think about the problem, very large<br />

and complex trees, which are designed to represent every possible scenario<br />

which can occur, can be counterproductive in many circumstances.<br />

Decision trees are models, and as such are simplifications of the real<br />

problem. The simplification is the very strength of the modeling process<br />

because it fosters the understanding and insight which would be<br />

obscured by detail and complexity. Nevertheless, in rare circumstances<br />

highly complex trees may be appropriate and software developments<br />

mean that their structuring and analysis can now be facilitated with<br />

relative ease. For example, Dunning et al. 1 used software to apply a<br />

decision tree with over 200 million paths to a ten-year scheduling problem<br />

faced by the New York Power Authority. Similarly, Beccue 2 used<br />

a tree with around half-a-million scenarios to help a pharmaceutical<br />

company to make decisions relating to the development and marketing<br />

of a new drug.<br />

Influence diagrams offer an alternative way of structuring a complex<br />

decision problem and some analysts find that people relate to them much<br />

more easily. Indeed Howard 3 has called them: ‘The greatest advance I<br />

have seen in the communication, elicitation and detailed representation<br />

of human knowledge ... the best tool I know of for crossing the bridge<br />

from the original opaque situation in the person’s mind to a clear and<br />

crisp decision basis.’ As we shall show later, influence diagrams can<br />

be converted to decision trees and we will therefore regard them in<br />

this chapter as a method for eliciting decision trees. However, some<br />

computer programs now exist which use complex algorithms to enable<br />

the influence diagram to be used not just as an initial elicitation tool but<br />

as a means for identifying the best sequence of decisions.<br />

Constructing a decision tree<br />

You may recall from earlier chapters that two symbols are used in<br />

decision trees. A square is used to represent a decision node and,

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!