Uncertainty propagation in fault tree analysis in systems subject to common cause faults

In Probability Safety Analysis (APS) of nuclear power plants, risk quantification involves the assessment of the accident scenario, its frequency of occurrence and the severity of the consequences. In the quantitative risk assessment, there are uncertainties in each of these elements. The frequency and unavailability values of critical systems for safety are, in most cases, evaluated in APS as fault tree top events, from data failure rates and repair (or probabilities of failures) of subsystems, equipment or components (basic events). The uncertainties of basic events are usually expressed in terms of confidence intervals of failure probability values or probability distributions. Estimating the uncertainties of the top event is an important step in nuclear power plant risk assessment, given that many accident scenarios are rare events and any decision-making processes should include uncertainty analyzes associated with their occurrence. Among the methods of analysis of the propagation of uncertainties in fault trees, this is, the obtention of the uncertainty of the top event from the uncertainties of the basic events, we can highlight the Moment Method. In fault tree analysis, one should consider the existence of Common Cause Failures (CCFs). CCFs can be caused by a variety of factors such as sabotage, human error, environmental conditions, design or manufacturing failure. Among the methods for estimating CCF probabilities, in the quantitative evaluation of fault trees, the Beta Factor Model can be highlighted. The objective of this work is to apply a method of uncertainty propagation in a system susceptible to CCFs and compare the results obtained with results available in the literature