Cluster Monte Carlo for System Unavailability Evaluation
Liu Ruilan · Acta Simulata Systematica Sinica · 2001
In simulation of highly reliable system, crude Monte Carlo methods are extremely wasteful of computer time. This paper presents a new application of “cluster sampling”of discrete distribution for calculating the system unavailability of system represented by a coherent fault tree or by a reliability block diagram. Since two uniform random numbers generate a trial, cluster Monte Carlo method appreciably reduces computation time, and hence a large number of trials become possible for highly reliable system simulation.