02.08.2013 Views

[Sample B: Approval/Signature Sheet] - George Mason University

[Sample B: Approval/Signature Sheet] - George Mason University

[Sample B: Approval/Signature Sheet] - George Mason 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.

The next step was to verify the additive independence assumptions. It was done<br />

through assessment of the attributes under each fundamental objective in order to define<br />

the form of the utility function (see section 3.7). The assessment procedure was<br />

conducted following the methodology proposed by Keeney and Raiffa. 107<br />

Since the additive independence implies utility independence, the assessment<br />

started verifying the former. If additive independence did not hold, then utility<br />

independence was assessed.<br />

As an example, the procedure used to assess the independence conditions between<br />

the attributes A3.3 Range (y) and A3.4 Payload Weight (z) will be described. The<br />

attributes Y and Z are additive independent if and only if the lotteries in figure 27 are<br />

indifferent for all amounts of y, z given a specific y‘, z‘.<br />

In this example: y = 5, y‘= 0, z = 250, z‘=0<br />

0.5 (5, 250)<br />

0.5<br />

L1 ≡ and L2 ≡<br />

0.5 (0, 0)<br />

0.5<br />

Figure 27. Additive Independence Assessment.<br />

Practically speaking, for any of the pairs taken from figure 26, the DM should be<br />

indifferent between lotteries L1 and L2 for additive independence to hold. In this specific<br />

107 Keeney and Raiffa, Decisions with Multiple Objectives.<br />

60<br />

(0, 250)<br />

(5, 0)

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

Saved successfully!

Ooh no, something went wrong!