ENIQ TGR Technical Document - Probability of Detection Curves: derivation and uses for Risk-Informed In-Service Inspection Applications

In the application of a non-destructive evaluation (NDE) method there are several factors that will influence whether or not the inspection will result in the correct decision as to the presence or absence of a flaw. In general, NDE involves the application of a stimulus to a structure and the subsequent interpretation of the response to the stimulus. Repeated inspections of a specific flaw can produce different magnitudes of the stimulus response because of very small variations in setup and calibration. This variability is inherent in the process. Different flaws of the same size can produce different response magnitudes because of differences in the material properties, flaw geometry and flaw orientation. Further, the interpretation of the response can be influenced by the capability of the interpreter (manual or automatic) and by the mental acuity of the inspector (in turn, dependent on many factors such as fatigue, emotional outlook, ease of access, environment, etc.). Much of the modern literature on inspection reliability constantly refers to a small set of seminal papers, derived from studies mainly carried out in the aeronautical industry. Most of the issues involved are of course very similar. One notable exception is possibly the fact that in the nuclear industry, for the very nature of the components being inspected, the sample sizes of inspected cracks tend to be much lower. In Europe, the ENIQ methodology for inspection qualification was specifically developed in the early 1990s because of the difficulty and cost of procuring or manufacturing representative flaws in test pieces in a high enough number to draw quantitative (statistical) conclusions on the capability of the NDE system being investigated. Rather, the fundament of the ENIQ methodology is the Technical Justification, a document assembling evidence and reasoning providing assurance that the NDE system is capable of finding the flaws which it is designed to detect with a high enough reliability. This assurance is qualitative, and comes usually in the form of statements such as: "Sufficient experimental verification of the procedure has been performed, on representative defects in test blocks with the correct geometry, to be confident that the procedure and equipment will find all defects which conform to the detection criteria and to specifications within the range of plausible defects". The purpose of this document, aimed mostly at NDE engineers and practitioners, is threefold: (1) to provide a brief literature review of some important papers and reports; (2) to review in a simple and structured way the statistical models that have been proposed to quantify inspection reliability and to point out problems and pitfalls which may occur, and (3) to describe and recommend statistical best practices for producing POD vs size curves from either hit/miss data, or â vs. a data.JRC.DDG.F.5-Safety of present nuclear reactor