In pipeline integrity threat management, the comparison of In-Line Inspection (ILI) and field results (trending) is typically used as an effective approach to determine the performance of ILI tool. ILI and field trending typically influences the integrity threat management of unmitigated features. Hence, pipeline operators should ensure collecting an adequate number of excavated ILI features from the field which are statistically relevant and represent the entire population of ILI data. The number of the trended data points is considered a sample from the entire population of ILI reported features. Nevertheless, it is considered a biased sample since it is drawn from the ILI population based on selective threat mitigation criteria. This paper describes several sampling methodologies with a focus on advantages and disadvantages of each approach. A Bayesian approach is introduced as an appropriate approach to the integrity sampling problem.

This content is only available via PDF.
You do not currently have access to this content.