A new simulation method based on the RVR principle for the rare event K - network reliability problem

Abstract

In this paper we consider the evaluation of a well Known K-network unreliability parameter by means of a new RVR Monte-Carlo method. It is based on seres-parellel reductions and a conditioning procedure using pathsets and cutsets for recursively changing the original problem into the unreliability problem for a smaller network. We illustrate by experimental results that the proposed method has good behavior in rare event cases and offers significant speed-ups over other state-of-the art variance-reduction techniques.