Article Periodicals » Materials Evaluation » Article
Dichotomy of Fluorescent Penetrant Inspection Reliability Challenges versus Opportunities for Improved Performance and Capability

The detection of large-sized discontinuities in a component is required to determine probabilistic design and effective service life management. (Large-sized discontinuities are defined as those with sizes greater than the critical discontinuity size, >aNDE). Probabilistic design differs from classical design. Probability design operates by assuming a small probability of failures caused by undetected large-sized discontinuities, whereas classical design uses the safety factor. Therefore, the classification, detection, and sizing of disconti-nuities in a component are important. Among all nondestructive testing (NDT) methods, fluorescent penetrant inspection (FPI) is one of the most used, with a caveat that FPI suffers from such limitations as subjectivity. The subjectivity factor, related to human factors in terms of both errors and viola-tions, exclusively depends on the human involve-ment in processing, inspecting, and evaluating components. In this study, the authors have attempted to address some of these factors by first studying the probability of detection (POD) results for hit/miss (binary numbers) in a traditional manner—as in MIL STD 1823A—and then studying the total observed system variability based on measurement system analysis (MSA) for FPI. The MSA was performed using data for the measured lengths using a 0.2 mm (0.008 in.) pin gauge for sizing in 61 linear cracks in 17 FPI panels. The total observed system variability includes meas-urement errors, repeatability, reproducibility, and process variations. The process variations include processing and inspecting parts. This study shows that the total gauge variation (that is, repeatability and reproducibility combined) makes up 26.22% of the overall study variation and is marginally acceptable per general rules applied to decide the capability of a measurement system. This study further shows that there is large difference between operators, and in general all operators demonstrated positive bias (overestimate). The presence of large positive bias in this study can in part be attributed to several sources, including fluorescent penetrant, gauge pin, and human factors. Therefore, there is a need to find ways to mitigate these risks by minimizing subjectivity through reducing the effects of human factors on NDT inspection data.


ASTM, 2013, ASTM E855-08: Standard Test Methods for Bend Testing of Metallic Flat Materials for Spring Application Involving Static Loading, ASTM International, West Conshohocken, Pennsylvania.

ASTM, 2016, ASTM E1417/E1417M-16: Standard Practice for Liquid Penetrant Testing, ASTM International, West Conshohocken, Pennsylvania.

Contra Costa Health Services, 2011, “Section B: Chapter 2, Human Factors/Human Error.”

Drury, Colin G., 2001, “Human Factors Good Practices in Fluorescent Penetrant Inspection,” Prepared for The Federal Aviation Administration Office of Aviation Medicine and Flight Standards Service Under Contract Number Dtfa01-94-Y-01013, University of New York-Buffalo Department of Industrial Engineering, Buffalo, NY.

FAA AC23-13A, 2005, “Fatigue, Fail-Safe, and Damage Tolerance Evalua-tion of Metallic Structure for Normal, Utility, Acrobatic, and Commuter Category Airplanes,” pp. 1-76.

Gorelik, Michael, Alonso Peralta-Duran, Surendra Singh, Jonathan Moody, and Michael Enright, 2009, “Role of Quantitative NDE Techniques in Probabilistic Design and Life Management of Gas Turbine Components-Part II,” Proceedings of ASME Turbo Expo 2009: Power for Land, Sea, and Air GT, Orlando, Florida, USA, pp. 1-8.

National Aerospace Standard 410, NAS Certification and Qualification of Nondestructive Test Personnel. 

MIL-HDBK-1823A, 2009, Department of Defense Handbook: Nondestructive Evaluation System Reliability Assessment, US Department of Defense, Washington, DC.

Rasmussen, J., 1983, “Skills, Rules, and Knowledge; Signals, Signs, and Symbols, and Other Distinctions in Human Performance Models,” IEEE Transactions on Systems, Man, and Cybernetics, Vol. SMC-13, No. 3., pp. 257-266. 

Schoonahd, J. W., J. D. Gould, and L. A. Miller, 1973, “Studies of Visual Inspection,” Ergonomics, Vol. 16, No. 4, 1973, pp. 365–379.

Singh, Surendra, Dan Greving, Andy Kinney, Fred Vensel, Jim Ohm, and Mike Peeler, 2012, “Eddy Current Measurement System Valuation (MSA) for Corrosion Depth Determination on Cast Aluminum Aircraft Struc-ture,” Review of Progress in Quantitative Nondestructive Evaluation: volume 31, AIP Conference Proceedings, volume 1430, AIP Conference Proceedings, Vol. 1430, No. 1, pp.1789–1796.

Singh, Surendra, 2014, “NDE Reliability and Probability of Detection—Evolution and Paradigm Shift,” AIP Conference Proceedings, Vol. 1581, pp. 2079.

Spencer, Floyd W., 1996, “Visual Inspection Reliability of Transport Aircraft,” Proceedings of the SPIE, Vol. 2945, Sandia National Laboratories, FAA Airworthiness Assurance NDI Validation Center, New Mexico.

Usage Shares
Total Views
69 Page Views
Total Shares
0 Tweets
0 PDF Downloads
0 Facebook Shares
Total Usage