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.

Constructing a decision tree 145<br />

because each branch emanating from this node presents an option, the<br />

decision maker can choose which branch to follow. A circle, on the<br />

other hand, is used to represent a chance node. The branches which<br />

stem from this sort of node represent the possible outcomes of a given<br />

course of action and the branch which is followed will be determined,<br />

not by the decision maker, but by circumstances which lie beyond his or<br />

her control. The branches emanating from a circle are therefore labeled<br />

with probabilities which represent the decision maker’s estimate of the<br />

probability that a particular branch will be followed. Obviously, it is not<br />

sensible to attach probabilities to the branches which stem from a square.<br />

The following example will be used to demonstrate how a decision<br />

tree can be used in the analysis of a multi-stage problem. An engineer<br />

who works for a company which produces equipment for the foodprocessing<br />

industry has been asked to consider the development of a<br />

new type of processor and to make a recommendation to the company’s<br />

board. Two alternative power sources could be used for the processor,<br />

namely gas and electricity, but for technical reasons each power source<br />

would require a fundamentally different design. Resource constraints<br />

mean that the company will only be able to pursue one of the designs,<br />

and because the processor would be more advanced than others which<br />

have been developed it is by no means certain that either design would<br />

be a success. The engineer estimates that there is a 75% chance that the<br />

electricity-powered design would be successful and only a 60% chance<br />

that the gas-powered design would be a success.<br />

Figure 6.1 shows an initial decision tree for the problem with estimated<br />

payoffs in millions of dollars. After considering this tree the engineer<br />

realizes that if either design failed then the company would still consider<br />

modifying the design, though this would involve more investment and<br />

would still not guarantee success. He estimates that the probability that<br />

the electrical design could be successfully modified is only 30%, though<br />

the gas design would have an 80% chance of being modified successfully.<br />

This leads to the new tree which is shown in Figure 6.2. Note that the<br />

decision problem is now perceived to have two stages. At stage one<br />

a decision has to be made between the designs or not developing the<br />

problem at all. At stage two a decision may have to be made on whether<br />

thedesignshouldbemodified.<br />

After some reflection, the engineer agrees that this tree is a satisfactory<br />

representation of the options facing the company. Other alternatives<br />

such as switching to the development of a gas-powered design if the<br />

electrical design is not successful are not considered to be feasible, given<br />

the resources available to the company.

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

Saved successfully!

Ooh no, something went wrong!