Chapter 5 Component Importance

Marvin Rausand
Department of Production and Quality Engineering Norwegian University of Science and Technology marvin.rausand@ntnu.no

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 1/19

Measures Covered
The following component importance measures are dened and discussed in this chapter:
I I I I I I

Introduction G Measures Covered G Importance Depends On Importance Measures

Birnbaums measure (and some variants) The improvement potential measure (and some variants) Risk achievement worth Risk reduction worth The criticality importance measure Fussell-Veselys measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 2/19

Importance Depends On
The various measures are based on slightly different interpretations of the concept component importance. Intuitively, the importance of a component should depend on two factors:
I I

Introduction G Measures Covered G Importance Depends On Importance Measures

The location of the component in the system The reliability of the component in question

and, perhaps, also the uncertainty in our estimate of the component reliability.

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 3/19

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Importance Measures

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 4/19

Birnbaums Measure (1)

Birnbaum (1969) proposed the following measure of the reliability importance of a component: Birnbaums measure of importance of component i at time t is h(p(t)) I (i | t) = pi (t)
B

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Birnbaums measure is thus obtained by partial differentiation of the system reliability with respect to pi (t). This approach is well known from classical sensitivity analysis. If IB (i | t) is large, a small change in the reliability of component i will result in a comparatively large change in the system reliability at time t

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 5/19

Birnbaums Measure (2)

By fault tree notation, Birnbaums measure may be written as IB (i | t) = where qi (t) = 1 pi (t) Q0 (t) = 1 pS (t) = 1 h(p(t)) Birnbaums measure is named after the Hungarian-American professor Zygmund William Birnbaum (1903-2000) Q0 (t) qi (t)

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 6/19

Birnbaums Measure (3)

By pivotal decomposition we have h(p(t)) = pi (t) h(1i , p(t)) + (1 pi (t)) h(0i , p(t)) = pi (t) [h(1i , p(t)) h(0i , p(t))] h(0i , p(t))

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Birnbaums measure can therefore we written as h(p(t)) I (i | t) = = h(1i , p(t)) h(0i , p(t)) pi (t)
B

Note that Birnbaums measure IB (i | t) of component i only depends on the structure of the system and the reliabilities of the other components. IB (i | t) is independent of the actual reliability pi (t) of component i. This may be regarded as a weakness of Birnbaums measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 7/19

Birnbaums Measure (4)

Since h(i , p(t)) = E[(i , X(t)] we can write
Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

IB (i | t) = E[(1i , X(t)] E[(0i , X(t)] = E[(1i , X(t) (0i , X(t)] When the structure is coherent [(1i , X(t)) (0i , X(t))] can only take on the values 0 and 1. Therefore IB (i | t) = Pr((1i , X(t)) (0i , X(t)) = 1) This is to say that IB (i | t) is equal to the probability that (1i , X(t)) is a critical path vector for component i at time t Birnbaums measure is therefore the probability that the system is such a state at time t that component i is critical for the system

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 8/19

Birnbaums Measure (5)

Assume that component i has failure rate i . In some situations we may be interested in measuring how much the system reliability will change by making a small change to i . The sensitivity of the system reliability with respect to changes in i can obviously be measured by h(p(t)) h(p(t)) pi (t) pi (t) = = IB (i | t) i pi (t) i i A similar measure can be used for all parameters related to the component reliability pi (t), for i = 1, 2, . . . , n. In some cases, several components in a system will have the same failure rate . To nd the sensitivity of the system reliability with respect to changes in , we can still use h(p(t))/

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 9/19

Birnbaums Measure (6)

Consider a system where component i has reliability pi (t) that is a function of a parameter i . The parameter i may be the failure rate, the repair rate, or the test frequency, of component i. To improve the system reliability, we may want to change the parameter i (by buying a higher quality component or changing the maintenance strategy). Assume that we are able to determine the cost of the improvement as a function of i , that is, ci = c(i ), and that this function is strictly increasing or decreasing such that we can nd its inverse function. The effect of an extra investment related to component i may now be measured by h(p(t)) h(p(t)) i pi (t) i = = IB (i | t) ci i ci i ci

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 10/19

Birnbaums Measure (7)

In a practical reliability study of a complex system, one of the most time-consuming tasks is to nd adequate estimates for the input parameters (failure rates, repair rates, etc.). In some cases, we may start with rather rough estimates, calculate Birnbaums measure of importance for the various components, or the parameter sensitivities, and then spend the most time nding high-quality data for the most important components. Components with a very low value of Birnbaums measure will have a negligible effect on the system reliability, and extra efforts nding high-quality data for such components may be considered a waste of time

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 11/19

Improvement Potential (1)

The improvement potential of component i at time t is IIP (i | t) = h(1i , p(t)) h(p(t)) The improvement potential may be expressed as IIP (i | t) = IB (i | t) (1 pi (t)) or, by using the fault tree notation IIP (i | t) = IB (i | t) qi (t)

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 12/19

Improvement Potential (2)

IIP (i | t) is the difference between the system reliability with a perfect component i, and the system reliability with the actual component i. In practice, it is not possible to improve the reliability pi (t) of component i to 100% reliability. Let us assume that it is possible to improve pi (t) to new value (n) pi (t) representing, for example, the state of the art for this type of components. We may then calculate the realistic or credible improvement potential (CIP) of component i at time t, dened by ICIP (i | t) = h(pi (t), p(t)) h(p(t)) where h(pi (t), p(t)) denotes the system reliability when component i is replaced by a new component with reliability (n) pi (t).
(n) (n)

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 13/19

Risk Achievement Worth

The importance measure risk achievement worth (RAW) of component i at time t is IRAW (i | t) = 1 h(0i , p(t)) 1 h(p(t))

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

The RAW is the ratio of the (conditional) system unreliability if component i is not present (or if component i is always failed) with the actual system unreliability. The RAW presents a measure of the worth of component i in achieving the present level of system reliability and indicates the importance of maintaining the current level of reliability for the component.

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 14/19

Risk Reduction Worth

The importance measure risk reduction worth (RRW) of component i at time t is IRRW (i | t) = 1 h(p(t)) 1 h(1i , p(t))

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

The RRW is the ratio of the actual system unreliability with the (conditional) system unreliability if component i is replaced by a perfect component with pi (t) 1. In some applications, failure of a component may be an operator error or some external event. If such components can be removed from the system, for example, by canceling an operator intervention, this may be regarded as replacement with a perfect component.

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 15/19

Criticality Importance (1)

The component importance measure criticality importance ICR (i | t) of component i at time t is the probability that component i is critical for the system and is failed at time t, when we know that the system is failed at time t. I
CR

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

IB (i | t) (1 pi (t)) (i | t) = 1 h(p(t))

By using the fault tree notation, ICR (i | t) may be written as I

CR

IB (i | t) qi (t) (i | t) = Q0 (t)

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 16/19

Criticality Importance (2)

Let C(1i , X(t)) denote the event that the system at time t is in a state where component i is critical. We know that
Pr(C(1i , X(t)))

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

= IB (i | t)

The probability that component i is critical for the system and at the same time is failed at time t is
Pr(C(1i , X(t)) (Xi (t)

= 0)) = IB (i | t) (1 pi (t))

When we know that the system is in a failed state at time t, then

Pr(C(1i , X(t)) (Xi (t)

= 0) | (X(t)) = 0)

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 17/19

Criticality Importance (3)

Since the event C(1i , X(t)) (X(t) = 0) implies that (X(t)) = 0), we get
Pr(C(1i , X(t)) (Xi (t) Pr((X(t))

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

= 0))

= 0)

IB (i | t) (1 pi (t)) = 1 h(p(t))

ICR (i | t) is therefore the probability that component i has caused system failure, when we know that the system is failed at time t. For component i to cause system failure, component i must be critical, and then fail. When component i is repaired, the system will start functioning again. This is why the criticality importance measure may be used to prioritize maintenance actions in complex systems

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 18/19

Fussell-Veselys Measure
Fussell-Veselys measure of importance, IFV (i | t) is the probability that at least one minimal cut set that contains component i is failed at time t, given that the system is failed at time t. Fussell-Veselys measure can be approximated by IFV (i | t) 1
mi j=1 (1

Introduction Importance Measures G Birnbaums Measure (1) G Birnbaums Measure (2) G Birnbaums Measure (3) G Birnbaums Measure (4) G Birnbaums Measure (5) G Birnbaums Measure (6) G Birnbaums Measure (7) G Improvement Potential (1) G Improvement Potential (2) G Risk Achievement Worth G Risk Reduction Worth G Criticality Importance (1) G Criticality Importance (2) G Criticality Importance (3) G Fussell-Veselys Measure

i (t)) (Q j

Q0 (t)

mi i j=1 Qj (t)

Q0 (t)

i (t)) denotes the probability that minimal cut set j where Q j among those containing component i is failed at time t

Marvin Rausand, March 19, 2004

System Reliability Theory (2nd ed.), Wiley, 2003 p. 19/19