Detecting cracks in fan blades using Sonic IR

We describe the analysis of data from sonic IR inspections on turbine blades. Separate but similar analyses were done for two different purposes. In both analyses, we fit statistical models with random effects to describe the crack-to-crack variability and the effect that the experimental variables have on the responses. In the first analysis, the purpose of the study was to find Sonic IR equipment settings that will provide good crack detection capability over the population of similar cracks in the particular kind of jet engine turbine blades that were inspected. Then, the fitted model was used to determine the test conditions where the probability of detection is expected to be high and probability of alarm is expected to be low. In our second analysis, crack size information was added and a similar model was fit. This model provides an estimate of POD as a function of crack size for specified test conditions.