Publication: Publication Date: 1 December 2022
What Is Probability of Detection?
Probability of detection (POD) evaluation is a widely accepted practice for quantifying the reliability of a nondestructive testing (NDT) technique. Inspections are often conceptualized and developed in laboratory environments, where factors affecting the inspection are highly controlled. However, when implemented in practice, NDT inspections suffer from many sources of variability, including changes from nominal geometry of the test piece, sensor variability, differences between operators, environmental effects on the sensor response including thermal and electromagnetic interference, and a myriad of other factors that are not present in the lab. Thus, to transition the NDT from the lab to production environments, engineers must have a quantified understanding of uncertainties. This is especially true for NDT systems that are implemented for safety-critical structures, where the life of the component may be managed with NDT. A fundamental question that must be answered in this context is: What is the largest discontinuity that could be missed when this NDT technique is implemented?
References
Agresti, Alan. 2007. An Introduction to Categorical Data Analysis. Hoboken, NJ: Wiley-Interscience. doi:10.1002/0470114754.
Annis, Charles, John C. Aldrin, and Harold A. Sabbagh. 2015a. “Profile Likelihood: What To Do When Maximum Probability of Detection Never Gets To One.” Materials Evaluation 73 (1): 96–99.
Annis, Charles, John C. Aldrin, and Harold A. Sabbagh. 2015b. “What is Missing in Nondestructive Testing Capability Evaluation?” Materials Evaluation 73 (1): 38–42.
Berens, Alan P. 1989. “NDE reliability data analysis.” In ASM Handbook, vol 17. Materials Park, OH: ASM International. 689-701.
Berens, Alan P., and Peter W. Hovey. 1981. Evaluation of NDE Reliability Characterization. vol. 1. Dayton, OH: University of Dayton Research Institute.
Brown, Jennifer. 2012. “It Takes More than a Statistician to Do Probability of Detection.” Materials Evaluation 70 (4): 421–426.
DeGroot, Morris H., and Mark J. Schervish. 2012. Probability and Statistics. Pearson Education.
Knott, Christine E., and C. Schubert Kabban. 2022a. “Modern Design and Analysis for Hit/Miss Probability of Detection Studies Using Profile Likelihood Ratio Confidence Intervals.” Materials Evaluation 80 (12): 32-49. https://doi.org/10.32548/2022.me-04272.
Knott, Christine E., and C. Schubert Kabban. 2022b. “Confidence Interval Comparisons for Probability of Detection on Hit/Miss Data,” Materials Evaluation 80 (12): 50-65. https://doi.org/10.32548/2022.me-04273.
Kutner, M.H., C.I. Nachtsheim, J. Neter, and W. Li. 2004. Applied Linear Statistical Models. 5th ed. New York: McGraw-Hill/Irwin.
Montgomery, Douglas C. 2017. Design and Analysis of Experiments. 9th ed. New York: John Wiley.
Rummel, Ward D. 1998. “Probability of detection as a quantitative measure of nondestructive testing end-to-end process capabilities.” Materials Evaluation 56 (1): 29-35.
Spencer, Floyd W. 1998 “Identifying sources of variation for reliability analysis of field inspections.” SAND-98-0980C; CONF-980552-ON, Sandia National Lab., Albuquerque, NM.
US DOD (Department of Defense). 2010. MIL-HDBK-1823A: Nondestructive Evaluation System Reliability Assessment, Department of Defense Handbook.
Metrics
| Usage | Shares |
|---|---|
Total Views 224 Page Views |
Total Shares 0 Tweets |
224 0 PDF Downloads |
0 0 Facebook Shares |
| Total Usage | |
| 224 |