Reliability Importance Analysis
Reliability importance analysis detects design weaknesses and component failures that are critical to the proper functioning of the system. It can assist in identifying the event that, if improved, is most likely to yield the greatest improvement in system performance. In the FTA module, you can treat any gate as the top gate for performing importance analysis. Importance analysis can help you to determine if additional resources should be allocated or redundancy added to improve the overall reliability of the system.
If the goal of the analysis is merely to calculate reliabilities and availabilities at different time points, importance analysis is not necessary. Basic event importance can be defined with respect to the top event probability at specified times.
The following table describes the most commonly used importance measures and indicates when you should use them for making various decisions. For more information, see Reliability Importance Measure Calculations.
Importance Measure
Description
Birnbaum
Determines the maximum increase in risk when component A is failed compared to when component A is operating. Use Birnbaum importance measures if you can decrease the unavailability of each event a specified amount with the same effort.
Criticality
Given that the top event occurs, determines the probability that the failure is a result of the failure of component A. Use criticality importance measures if improvements can be made only to events that have high unavailabilities. (It may be impossible to decrease the unavailability of events after a certain limit, and the unavailability of some events may already be very low.) Additionally, use criticality importance measures if the object is to prioritize maintenance efforts.
Fussell-Vesely
Given that the system has failed, determines the probability that component A has contributed to the system failure. Use Fussell-Vesely importance measures if the objective is to minimize the individual contributions of basic events.
Fussell-Vesely Alternative
Given that the system has failed, provides a relative indication on how much a primary event is contributing to the top-level event.
‘ith’ Cut Set
Provides a percentage of cut set failure probability regarding the failure probability for the top-level event.