Evaluation of risk impact of changes to surveillance requirements addressing model and parameter uncertainties

Ageing PSA incorporating effectiveness of maintenance and testing

“…Nowadays, PSA is an efficient tool for evaluation of risk impact of changes to licensing basis [4] and in particular to surveillance requirements of NPP technical specifications within the framework of the Risk Informed Decision Making (RIDM) according to R.G. 1.174 principles [5,6].…”

“…This APSA could be used to support risk-informed decisions on the effectiveness of maintenance programs and technical specification requirements of critical equipment of Nuclear Power Plants (NPP), for example adopting the methodology proposed in Refs. [5,6].…”

“…monthly or quarterly, with positive and negative effects [5]. The positive effect of testing is its capability to detect hidden failures and this way limiting the risk of undetected downtimes of the safety component, i.e.…”

“…Using a level 1 PSA, the corresponding risk metric for the evaluation of ageing is the annual average baseline CDF considering the effect component ageing, which can be re-formulated for a single component i using the law of the total probability as follows: On the other hand, RG 1.174 establishes two risk metrics for evaluating the risk impact of whatever change to the licensing basis [5], which are herein adapted to address the risk impact of equipment ageing in the example of application. Using a level 1 PSA, the required risk metrics for the evaluation of risk impact of ageing would be the annual average baseline CDF before ageing (named herein CDFb) and the change in the annual average baseline CDF due to equipment ageing, which can be formulated as follows:…”

Sensitivity analysis with MC simulation for the failure rate evaluation and reliability assessment

“…In addition to the example of data from field above mentioned, in a complex system as a process or plant other sources of uncertainty affect input data can be considered such as measurement errors for sensor wear or fault, poor information about equipment installation and work environment, environmental condition of use, maintenance activities, partial understanding of the driving forces and mechanisms, and so on [13][14][15][16][17][18][19]. Considering the incomplete knowledge of data, often acquired by fieldfield data of service as, for instance, maintenance information or failure informationor collected in database, and different sources of uncertainty, it is fundamental to identify model inputs that cause significant uncertainty in the output.…”