EP 1130-2-500
27 Dec 96
Step 1. Based upon the reliability index (see paragraph f, Engineering Considerations)
calculated for the current physical condition, select the probability of unsatisfactory performance
for each feature, or component, from Table D-1 of Appendix D. If the probability of
unsatisfactory performance is due to a combination of events, provide the method used to
determine these probabilities. Both the probability of unsatisfactory performance of a feature
and the probability of occurrence of an event which results in load conditions causing the
unsatisfactory performance shall be explicitly discussed and displayed. Reporting requirements
to support the reliability analysis are addressed in Appendix D.
Step 2. Based on the existing physical condition of, and the current and forecasted
demands on the features, estimate the frequency of service disruption and the physical
consequences resulting over the planning period. Frequencies and consequences should be
expressed in terms which are unambiguous and which facilitate analysis. For example, estimate
the percent chance of disruption per year (annual probability) or probability of disruption per
event (per event probability).
Step 3. Develop an event tree. A useful way of presenting information of alternative
future pathways is an event tree diagram. The event tree is used to display the possible outcomes
from some initiating event. Figure C-1 is an event tree for a hydroelectric generating facility.
Step 4. Estimate all costs necessary to correct the service disruption. The repair should
be the least cost fix necessary (as considered reasonable for the circumstances) to continue
service.
Step 5. Estimate the economic cost for each disruption. (See Appendix E)
Step 6. Combine the frequency of service disruption with the consequences of disruption.
Monte Carlo simulation is one technique for combining risks and determining expected values.
This technique is especially useful when the arithmetic of the expected value calculation is
highly complex or intractable. Under some, perhaps many situations, the standard statistical
procedure of summing the products of the probabilities and corresponding consequences is
sufficient. That is, calculating the value analytically may be more expedient and transparent than
estimating by simulation. An advantage of the Monte Carlo approach is that it yields both the
expected value and the variance. The fundamental point of the analysis however, is to explicitly
consider the likelihoods and consequences of the base condition. See Appendix F for further
consideration of this approach.
(3) With Rehabilitation Condition.
(a) General. As previously stated, the base condition should@ describe an immediate
or certain failure. Nor is the only project alternative immediate and full scheduled rehabilitation.
There are a variety of intermediate strategies that should be evaluated. In addition, the
rehabilitation decision must give consideration to the choice of timing and extent of
rehabilitation. Therefore, the approach is to develop alternatives to solve the problems. This
does not predetermine that one major rehabilitation scenario is the only alternative.
(b) Alternatives Considered. Discuss the alternatives considered. The narrative should
address the level of detail developed for each alternative, the data available, assumptions made
and the level of reliability y, risk and uncertainty associated with the alternative. Present the
B-5
Previous Page
