Optimizing Crack Repair Policies using Value of Information Analysis considering Imperfect Inspection of Pipelines

Abstract

This study presents an approach for optimizing policies on crack repair, which may take place after pipeline inspection, depending on the possible failure risk (i.e., the severity of crack) and the repair cost. The approach is built on the basis of Value of Information (VoI) analysis, and it seeks the optimal repair criterion by minimizing pre-posterior cost for re-inspection. For this purpose, a tool is developed for VoI analysis on a high-performance computing platform, which allows parallel computing to improve computational efficiency. The tool is specifically designed for inspection of a single critical crack in this study. The effectiveness of the presented approach in increasing VoI, i.e., taking better advantage of inspection, is demonstrated through its application to a hypothetical pipeline example. Different failure-to-repair cost ratios and prior uncertainties are also considered to reveal their effect on the optimal repair policies.