Network s-t reliability bounds using a 2-dimensional reliability polynomial

Computing the reliability of a generic network is computationally difficult, and an exact solution is unattainable for many problems. Faced with this computational difficulty, two compromises are usually made: (1) be content with reliability bounds instead of exact reliability; and (2) simplify analysis by assuming that all but 1 component-type are perfectly reliable. This paper relaxes the assumption of perfect components and computes the s-t reliability of a network whose operation is a function of the reliability of 2 component types (nodes and links.) The extension of the reliability polynomial to 2 dimensions removes the need to assume that links or nodes are perfectly reliable.>