Chapter 15. Probabilistic Approaches for Analysis of Inspection Data of Pipelines

Inspection of pipeline to assess the extent of corrosion is performed using in-line inspection (ILI) tools, called intelligent pigs. The two mains ILI tools are magnetic flux leakage (MFL) and ultrasonics. MFL tools are the most widely used because they are applicable for both gas and liquid pipelines. Ultrasonic tools are generally limited to liquid pipelines. These inspection tools, like all inspection tools, suffer from three main uncertainties: Probability of detection (POD), probability of false call POFC) and sizing accuracy. Therefore, the decision regarding the integrity of the pipeline and its remaining safe operating life must account for the influence of these uncertainties. In this paper, two probabilistic approaches are presented. The first approach is based on assessing the extent of corrosion damage based on historical measurement of matching defects and thus it mainly considers the sizing accuracy of the inspection tool. The second approach does not depend on defect matching but accounts for defect population along a pipeline section subjected to similar environmental conditions. This population-based approach accounts for POD, POFC and sizing accuracy. The paper includes example analyses demonstrating the two approaches.