Measuring Differences Among Probability of Detection Curves
Probability of Detection (POD) curves are compared by two statistical methods to quantify system-to-system differences. The first method assesses performance among a group of inspection systems through an adaptation of statistical analysis of variance (ANOVA). The second method uses a chi-squared statistic to test for a difference between two systems. Examples using eddy current data are given for each technique.
Unable to display preview. Download preview PDF.
- 1.Thompson and Chimcnti, editors, (1982–1986) — Review of Progress in Quantitative Nondestructive Evaluation, Vol. I (1982), Vols. 2A,B (1983), Vols. 3A,B (1984), Vols. 4A,B (1985), Vols. 5A,B (1986), Plenum Press, New York.Google Scholar
- 2.Annis, C. et.al. (1979) 1st Executive Review “Concept Definition-Retirement-for— Cause of F100 Rotor Components”, 1 August 1979.Google Scholar
- 3.Berens, A.P. and P.W. Hovey (1983), “Flaw Detection Reliability Criteria”, Final Report, UDR-TR-83–137, — University of Dayton Research Institute.Google Scholar
- 4.Annis, C. and K. Erland (1987), “Estimating the Probability of Crack Detection from Data with Right- and Left-Censored Observations”, submitted for publication, and offered as a working document for the Air Force NDE Reliability Team; to be presented at the American Statistical Association Annual Meeting, New Orleans, August 1988.Google Scholar
- 6.Kendall and Stuart (1961) — The Advanced Theory of Statistics, Vol. 2: Inference and Relationship, London: Charles Griffin.Google Scholar
- 7.Cramer, H. (1946) — Mathematical Methods of Statistics, Princeton, New Jersey: Princeton University Press.Google Scholar
- 8.Bury, K.V. (1975) — Statistical Methods in Applied Science, New York: John Wiley.Google Scholar
- 9.Cheng and lies (1983) — “Confidence Bands for Cumulative Distribution Functions of Continuous Random Variables”, Technometrics, Vol. 25, No. 1, February, 1983.Google Scholar