A Quantitative Approach to Risk-Informed Safety Significance Categorization with an Early Expert Participatory

摘要

A quantitative approach to RISSC with an early expert participatory process is proposed. The relative importance of each of the decision factors is quantified using the AHP and the safety significance in each of the decision factors is, also, quantified in a probability form using BBN. These measures, finally, are integrated into P(S_T=Hi) which is the safety significance measure for the RISSC in this work. In the applications, there were differing opinions among the experts, as shown in DF-8 in Table 3. Expert-1 and expert-3 evaluated that CC-V073 fell in c_2 in DF-8, while expert-2 evaluated that the component fell in c_3. The P(S_8=Hi) was, however, updated into P(S_8=Hi) through the BBN process on the basis of all the expert evaluations. In addition, as the probabilities P(S_i) were updated, the decision factors in which the target component was safety significant were clearly distinguished, as shown in Table 3. This merit is thought to be useful for the presentation of the bases of the final decisions to the relevant stakeholders including regulatory authorities. Consequently, it is concluded that the proposed method can overcome the demerits of the conventional methods and allow us to effectively arrive at a final categorization. The probabilistic and deterministic insights used in this work are, however, based only on level 1 and 2 internal event PSAs and normal operating conditions. The study on other conditions, therefore, should be followed in order to perform the RISSC based on all the conditions of an NPP.