Examination of SAMEVAL: Software to calculate and evaluate sample sizes for national safeguards inspection in the Republic of Korea

Abstract

As the International Atomic Energy Agency (IAEA) implements the state level approach (SLA) as a safeguards approach, the role of a state system of accounting for and control (SSAC) becomes critically important. The Republic of Korea (ROK), as a member state under the SLA, is implementing an independent sampling plan to strengthen the technical capability of the SSAC. Regarding this sampling plan, this study proposes SAMEVAL: a GUI-based independent sample size calculation and evaluation software. The SAMEVAL software minimizes approximations in the IAEA's ''common inspection on-site software package (CIOSP)''. It can also evaluate the effectiveness of the calculated sample sizes for representative diversion scenarios: a novel capability not offered by CIOSP. This study then demonstrates the SAMEVAL software using 20 benchmark cases of an IAEA technical report (STR-261), and demonstrates the effectiveness of the calculated sample sizes using SAMEVAL for five diversion scenarios with different defect sizes. The effectiveness testing of SAMEVAL evaluates the non-detection probability (β) of the calculated sample sizes for a given diversion scenario based on the Monte Carlo calculation. It then compares the calculated β to the target non-detection probability (βc) of each case. Our results indicate that SAMEVAL reduces inspection burden by reducing the total sample size or sample sizes with more precise verification methods while maintaining the same level of non-detection probability. With these capabilities, SAMEVAL can be a tool for the sampling plan in national safeguards inspection in the ROK to strengthen the technical capability of the SSAC.

Keywords

• Nuclear safeguards
• National inspection
• Sample size calculation
• Diversion scenario
• Monte Carlo