Abstract
The Maintenance Rule (MR) is a requirement established by the U. S. Nuclear Regulatory Commission (NRC) to check the effectiveness of maintenance carried out in nuclear plants. MR was consolidated in the U.S. in 1996. MR classifies Structures, System or Components (SSC) into two categories: Category (a)(2), the SSCs that reach the intended performance demonstrate that preventive maintenance is being appropriately performed, and category (a)(1), which stands for the SSCs that do not fulfill category (a)(1), and must have established goals, so that the discrepancies can be revised and then return to Category (a)(2).
The purpose of this paper is to present a model that can be implemented in a personal computer, which allows the reliability analysis of nuclear power plant safety systems to be performed, starting from a state transition diagram, typically employed in Markovian reliability analyses. The classical Markovian model will not apply if at least one of the transition rates in the state transition diagram is time-dependent, due to equipment aging. To overcome this difficulty, the model is recast into a Markovian model by means of supplementary variables. The algorithm developed for this purpose also estimates the necessary parameters for the reliability analysis, starting from crude failure information available from the plant and using goodness of fit statistical tests for the most commonly employed probability distributions.
