Guideline 1.2-- Clarity Test.

DOE-DP-STD-3023-98

Discussion. Comprehensiveness ensures that important considerations or points of

view will not be ignored. Relevance ensures that evaluation criteria will reflect

appropriate program scope. Mutual exclusiveness means that the decision objectives

do not overlap, and double counting is prevented. Independence of decision

objectives (in the sense that the value or importance of achieving any one objective

does not depend on the degree to which other decision objectives are achieved)

promotes mathematically simpler RBP models that are easier to understand.

Sometimes, the statements of decision objectives should be limited for pragmatic

reasons and should be addressed by decision makers outside the RBP process. A

minimum number of decision objectives should reduce the difficulty of implementation.

9.1.2

Guideline 1.2-- Clarity Test. Performance measures should be as unambiguous as

possible, such that given sufficient data, it would be possible to specify a level of

performance for each performance measure and for each decision option.

Discussion. For illustration, consider the risk measure " umber of people receiving

excessive exposure." This measure fails to pass the clarity test. The terms

"excessive" and "exposure" would have to be clarified before " eaningful estimates"

could be associated with specific alternatives. A proper measure could be " umber of

people receiving a whole body radiation dose in excess of 200 rem committed

effective dose equivalent."

9.1.3

Guideline 1.3-- Consistency with Principles of Rationality. The prioritization logic

should produce a ranking of decision options that does not violate any basic principles

of rationality. In particular,

a. Activities with identical benefits and costs should be ranked identically.

b. If the activity is changed for the better, the ranking should not decline.

c. The ranking of an activity should be insensitive to the addition or removal of

independent activities.

d. The ranking of an activity should be insensitive to previously committed costs

(sunk costs).

Discussion. Some prioritization models, while appearing at first glance to be perfectly

logical, can produce results that are inconsistent with what might be considered basic

axioms for rational decision making. Therefore, it is important to verify that the chosen

model conforms with the basic principles of rationality. As an example, suppose

activities have been labeled and ranked A, B, C, D, and E, and sufficient funds are

available to conduct only the top three ranked activities. Assume that A, B, and C turn

out to be the top-ranked activities, and activities D and E are determined to fall below

the cutoff line. If the availability of activity D were determined to be irrelevant to the

choice of the top three activities (i.e., Guideline 1.4 was followed), removing activity D

from the prioritization should not alter the relative ranking of the remaining activities.

A system that did not ensure this result would lack credibility even if it was never used

in situations where activities were added or subtracted.

Guideline 1.4-- Independence of Decision Options. Decision options should be

9.1.4

defined in a manner that maximizes their independence. Dependencies occur if the

cost, risk, or benefit of performing any decision option depends on whether or not any

other alternative is conducted.

Discussion. If two decision options proposed for evaluation are determined to be

dependent, they should be combined into a single independent composite decision

option to facilitate prioritization. For example, if decision options A and C share the

use of capital investments or each affect overlapping risks, the option of selecting both