Spatial effects in risk-based design and maintenance of pipelines

Pipelines are to a large extent spatially continuous systems having a system-component relationship that is not as clearly articulated as for other structural systems. Reliability-based design methods for pipelines often provide conflicting views about the spatial extent of limit states, the effect of spatial correlation, the applicability of target risks and target reliabilities (for instance on a per unit length basis), the link with lifecycle cost methods, and risk acceptability in general. The present paper first reviews probabilistic design and assessment approaches for pipelines, ranging from partial factors and limit state design, to reliability based and consequence-based methods. Subsequently we identify the various types of limit states from the point of view of their spatial characteristics. The paper also reviews the possible approaches to target risks and target reliabilities in view of the different spatial extent of the limit states. The role of spatial correlation as it impacts on different kind of pipeline limit states and on the risk acceptance process is discussed. The role of inspection, repair and maintenance can easily be included in many of the reliability-based pipeline design and assessment approaches as the lifetime costs of mitigative actions are fairly well defined, together with the spatially distributed consequences of failure, but they do add some additional challenges to the spatial modeling of the system.