Partially unbiased estimators for unavailability calculations

The analogy between system reliability and neutron transport is well-known. It has been a major source of inspiration for the development of Monte Carlo (MC) algorithms for the estimation of reliability characteristics. Strictly speaking, a formal analogy is only valid for reliability calculations, because in this case a system failure can be interpreted as the absorption of a particle. Hence availability estimations require an extension of the transport formalism, which is briefly presented for the Markovian assumption. Several sets of estimators of the system unavailability are deduced from this framework, each of these sets giving unbiased estimations, whatever the starting conditions. The respective performances of these sets are compared on a numerical example.