A pragmatic approach to modeling common cause failures in multi-unit PSA for nuclear power plant sites with a large number of units

•A pragmatic approach to model common cause failures in multi-unit PSA is provided.•This approach is applicable to nuclear power plant sites with a large number of reactor units.•This approach can also be applied to cases where non-identical units are included.•Its effectiveness is demonstrated by applying several cases with different number of units.•Results are compared with those obtained from existing CCF modeling approaches. One of the major issues in multi-unit probabilistic safety assessment (MUPSA) is how to deal with inter-unit common cause failures (CCFs). Most existing studies on MUPSA have focused on two-unit nuclear power plant (NPP) sites, where it is often not difficult to extend currently available CCF modeling approaches, such as the Alpha Factor and Beta Factor models, to address inter-unit CCFs. However, when considering an NPP site with three or more units, these approaches can be inapplicable or yield overly conservative results. This paper proposes a pragmatic approach to modeling CCFs for application to MUPSA involving a large number of NPP units. Provided here are the criteria for selecting CCF groups for which inter-unit CCFs are considered, as well as the methods for modeling CCF combinations and estimating their probabilities. The effectiveness of the proposed approach is then demonstrated by application to cases with different numbers of identical units as well as to cases where non-identical units are included in MUPSA. Results are also compared with those obtained from existing approaches.