Statistical Analysis of In-Line Inspection Performance with Gamma Distribution

The performances of two or more In-Line Inspection (ILI) systems in terms of defect sizing accuracy are typically compared using standard statistical methods. Each system produces a performance specification, computed using a standard set of defects, using three quantities: Tolerance, Certainty and Confidence. In order to compute the three quantities that comprise the performance specification, an underlying statistical distribution is assumed. The most common choice of such distribution is Gaussian distribution.

To assess the specified performance of a system, a different approach is typically used. API 1163 [1] specifies use of binomial distribution based method to verify the claimed performance of a system. References [2], [3] describe the method in greater details.

It is observed that the procedures used for performance specification and verification use different approaches. Also there are some aspects of the procedures that are non-intuitive and theoretically inadequate. In this paper, author proposes a new and uniform approach for performance specification and verification using Gamma distribution.

References:

1. API 1163, “In-Line inspection Systems Qualification Standard”, American Petroleum Institute, First Edition, 2005.
2. Guy Desjardins, Mike Reed and Randy Nickle, “ILI Performance Verification and Assessment Using Statistical Hypothesis Testing”, 6^th International Pipeline Conference, 2006.
3. Roger McCann, Rick McNealy and Ming Gao, “In-Line Inspection Performance Verification”, NACE International Corrosion Conference and Expo, 2007.