Reliability-based management of underground pipeline network using genetic algorithm

When the residual ultimate strength of buried pipeline has exceeded the limit due to external loadings, failure becomes imminent and overall reliability is reduced. Therefore, the concerned industry has to plan how to operate and maintain the system within budget constraints. This paper is concerned with estimating reliability and deciding when and how interventions are needed to prevent unexpected failures of flexible buried metal pipelines subject to externally applied loadings and corrosion at minimal cost. The probability of failure due to corrosion induced excessive deflection with respect to varying time has been estimated in this study. Then intervention year for maintenance is determined and the most appropriate renewal solution is identified by minimising failure risk and whole life cycle cost using genetic algorithm. An example is presented to validate the proposed method with a view to prevent unexpected failure by prioritising maintenance based on failure severity and pipeline reliability.