Reliability Based Inspection For Corroded Pipelines

Pipeline Integrity Management is a growth industry. Much of the material published in the last few years has been directed to optimising maintenance cost by accepting and managing some level of risk. Risk is typically established by complex statistical calculations based on available data. The method used in the present study works well when both the accuracy and confidence of the data used in the calculations are very high. This method works less well when the data is less accurate and can lead to underestimating both the degree of risk and its associated cost. This project is a general review of industry practice for determining risk and its cost. The project reviews also other specifications (accuracy and confidence) of in-line-inspection tools with regard to the use of data as a basis for complex calculations (based on Monte Carlo Simulation) to determine the acceptable risk and the optimum inspection interval. Currently the facilities protection department of the Sonatrach’s transportation branch has no risk-based models that consider both the risk of failure along the pipeline network and the cost of inspection as a basis for management decisions on ILI (in-line) inspection scheduling and repair activities. Furthermore the ILI inspection decision is generally based on a fully subjective assessment. The facilities protection department does a good job of maintaining the pipelines, but the decision as to where to allocate resources and when in some cases may be a reactive measure. The present study applies risk-based model to show that the accepted risk has a quantifiable cost commensurate with the accuracy of the data used in the risk assessment process. The proper use of risk assessment / risk management principles and tools in the present study can help the pipeline operator to maintain the flow of pipeline integrity data and the analysis of this data. In the other hand the risk can be estimated and its attributes can be defined through an algorithm (series of relationships and mathematical simulations). This risk-based model may aid in the rational, prioritisation of resources and identification of improvement opportunities.