The calculation of the reliability of large systems represented by the minimal cut set of componental failure events formulated in the space of basic uncertainty variables is proposed by using first-order reliability techniques. Since the computational effort can grow enormously with the size and complexity of the system, it is proposed to replace subsystems by equivalent components representing as far as possible the reliability characteristics of the subsystem. Then, simpler, computationally amenable systems are obtained at the higher levels in the system hierarchy. It is shown that cut set representations of subsystems are much more suitable for the derivation of equivalent failure surfaces than tie set representations. Two importance measures for basic uncertainty variables, components or subsystems are defined in the context of first-order reliability.
[1]
Richard E. Barlow,et al.
Statistical Theory of Reliability and Life Testing: Probability Models
,
1976
.
[2]
G. C. Corynen.
Stop: a fast procedure for the exact computation of the performance of complex probabilistic systems
,
1982
.
[3]
R. Rackwitz,et al.
First-order concepts in system reliability
,
1982
.
[4]
R. Rackwitz,et al.
Non-Normal Dependent Vectors in Structural Safety
,
1981
.
[5]
R. Rackwitz,et al.
Quadratic Limit States in Structural Reliability
,
1979
.
[6]
O. Ditlevsen.
Narrow Reliability Bounds for Structural Systems
,
1979
.
[7]
O. Ditlevsen.
Generalized Second Moment Reliability Index
,
1979
.