DSM reliability concerns — Impact on safety assessment

□ Constant failure rate hypothesis used for all Safety studies is strongly put into question □ Methodologies to deal with non-constant failure rates are not handled in the regulation, which even recommend to “contact a specialist” □ Methods allowing to deal with non-constant failure rates (Weibull law, Monte Carlo simulation) are quite complex □ Finally, we have to answer these three questions : □ First, the question of the compliance to Safety requirements (which margin do we have ?); □ Second, the question of the position of the regulation relating to non-constant failure rates; □ Third, the question of the way to deal with non-constant failures from a mathematical and methodological point of view.