Comparison between failure sequence and failure path for brittle systems

Abstract

For classes of structures with discrete members, such as trusses and frames, system failure is generally expressed in terms of component failures. The components are presumed to be ‘two-state’ members i.e. either fail or safe. Furthermore, for such systems, it is convenient to use a failure tree to represent the various combinations of component failures that will lead to system failure. Search strategies are then developed to search the failure tree for the dominant modes of the system. As part of strategy the probability of occurrence of a failure mode is evaluated. The exact method of evaluating such probabilities is called the failure sequence formulation. Alternatively, an approximate method called the failure path formulation has been widely used. A comparison between the failure path formulation to the failure sequence formulation is considered. The implications of the differences of the two formulations on the result of search strategies are discussed.