Type I and type II error associated with verification and confirmation measurements

Conference · Fri Jun 01 04:00:00 EDT 2001

OSTI ID:975620

Scott, B G ^[1]

---

1. Brian G.

DOE M 474.1 states that 'For Category I and II items, the acceptance/rejection criteria for verification measurements and, where possible, for confirmation measurements must be based on the standard deviation for the measurement method under operating conditions.' Determination of the acceptance/rejection criteria for confirmation and verification measurements may involve null and alternative hypothesis testing. The specific values used for computing the appropriate test statistics are dependent on the desired type I and II errors. If hypothesis testing is applied to a verification measurement (such as verification measurement signal interpretation equals nuclear material book value), then the successful application of hypothesis testing and accurate determination of type I and type II error is dependent on the existing statistical assumptions. This paper discusses some common statistical assumptions that may be used in support of confirmation and verification measurements. The implications of these assumptions on type I and II error are briefly analyzed. Suggestions for validating the desired type I and II error for confirmation and verification measurements are provided. This paper, by providing an introduction to some common statistical assumptions and the associated implications on error, aids safeguards professionals in their ability to determine the validity on the statistical models that support their verification and confirmation measurement programs.

• Research Organization: Los Alamos National Laboratory
• Sponsoring Organization: DOE
• OSTI ID: 975620
• Report Number(s): LA-UR-01-3666
• Country of Publication: United States
• Language: English

Related Subjects

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS
HYPOTHESIS
NUCLEAR MATERIALS MANAGEMENT
SAFEGUARDS
STATISTICAL MODELS
STATISTICS
TESTING
VERIFICATION