Monte Carlo games especially designed for probabilistic dynamics problems have been shown to be very efficient for the calculation of a pressurized water reactor (PWR) pressurizer unreliability. However, these methods rely on the implicit hypothesis that transition rates and probabilities, as well as control means response times, can be precisely determined. But reality is different: Because of a lack of knowledge, only crude estimations of transition rates can be obtained. Moreover, control device response times cannot be assumed constant for rapid transients. Therefore, considering distributed values of these quantities seems more realistic. Several attempts to take into account these uncertainties, while conserving as many advantages as possible of the high-efficiency simulation techniques, are summarized. The PWR pressurizer application is used to compare the capabilities of the proposed methods.