Updated Guidance on Sensitivity/Uncertainty Methods Use for Nuclear Criticality Safety Validation

The use of sensitivity/uncertainty (S/U) methods in nuclear criticality safety (NCS) was established more than 25 years ago and has been increasing ever since. Recent interest in commercial applications involving uranium enrichments above 5 wt%²³⁵U has increased focus on demonstrating the applicability of existing critical experiments to this enrichment range. S/U techniques can also be particularly useful in assessing the similarity of new fuel forms in relation to those documented in existing published evaluations of benchmark critical experiments. Publishing updated recommendations for the use of these methods is thus important to ensure that new S/U practitioners can benefit from the years of development and application invested in generating a set of best practices to maximize the utility of these tools. This paper summarizes the main conclusions of each section of a NUREG/CR document that was published in April 2025 by the US Nuclear Regulatory Commission specifically addressing this need for updated guidance on S/U method use in NCS validation. Although the document focuses on implementations within the SCALE code system, the same principles are applicable to other code systems which have implemented a subset of the techniques available within the TSUNAMI suite. The first sections of the NUREG/CR serve as review and introduction to validation and sensitivity/uncertainty analysis. Following that, a set of three case studies demonstrate the generation of sensitivity coefficients and their use for similarity assessment. The final section discusses advanced tools including the generalized linear least squares (GLLS) method for data adjustment.