Statistical analysis of condition assessment data and prediction of future performance of PCCP

December 30, 2011
Publication: Pipelines 2011: A Sound Conduit for Sharing Solutions - Proceedings of the Pipelines 2011 Conference, ASCE p 160-169
Author(s): Mehdi Zarghamee Graham Cranston Fongemie, Roger Wittas, Daniel

Abstract: This paper presents the benefits of a well-managed, long-term asset management program in determining the current condition and predicting future deterioration of prestressed concrete cylinder pipe (PCCP). More than 10 years of inspection and monitoring data were analyzed, accumulated from Palo Verde Nuclear Generating Station (PVNGS), operated by the Arizona Public Service Company (APS). Pipe segments were inspected every 3 years on average. These data were analyzed to determine the current condition of the PCCP assets, and to develop statistical models of future deterioration of the pipeline and future deterioration rates of distressed pipes in different repair priorities by tracking individual deterioration zones throughout the inspection history. Using established risk and uncertainty analysis techniques, future repair priorities are forecast. The frequency of inspections and quality of the data collected allowed us to develop meaningful predictive statistical models of deterioration in PCCPassets. Starting from the detailed data on distress levels in the pipelines, the models forecast future progression of distress, including the total number of distressed pipes, total number of highly distressed pipes requiring repair or replacement, areas of severe distress, areas of rapid deterioration, and specific pipe segments reaching severe distress level in a specific future time period. The inspection data and results of this study were incorporated into a Geographic Information System (GIS). Such information is of paramount importance for effective PCCP asset management.

Keywords: PCCP Pipeline