Influence of sample size and other factors on hit/miss probability of detection curves : ENIQ TGR technical document

The purposes of this document, aimed mostly at NDT engineers and practitioners, are threefold: (1) to extend the conclusions of an earlier report (ENIQ report No 41: "Probability of Detection Curves: Statistical Best-Practices" ), (2) to justify the Rule-of-Thumb that a valid Probability of Detection (POD) vs. size curve requires a minimum of 60 targets for binary response (hit/miss) data, (3) to provide guidelines for the NDE practitioner in designing a study to assess the effectiveness of a binary response inspection system using POD vs. size curves.