Risk-Based Considerations in Developing Strategies to Ensure Pipeline Integrity—Part II: Applications

Abstract

This is the second in series of two papers generated from a recent study on risk-based analysis for developing strategies to ensure pipeline integrity. This paper (Part II—Applications) focuses on the applications of the proposed deterministic and probabilistic models presented in the first paper (Part I—Theory) (Leis and Rahman, 1994) for stochastic pipe fracture evaluations. Using these models, numerical predictions are made for line-pipe steel typically used in gas transmission pipelines and are compared with the available test data. Thereafter, the paper explores the significance of the random variables related to serviceability in pipelines subjected to flaw growth in service. The results are discussed in the light of a hydrotest-based approach to ensure pipeline integrity. It is concluded that analysis of hydrotest strategies to optimize safety for such populations (e.g., Leis and Brust, 1992) should be based on a probabilistic analysis that permits risk assessments associated with pipeline operating decisions and the type and frequency of hydrotests done to ensure continued safe operation of the line. This same probabilistic framework could be used to assess the operating and safety implications for flaw populations characterized by in-line inspection.