Five More Years of Data: Predicting the Performance of Prestressed Concrete Cylinder Pipes; An Updated Case Study

In 2017, the Lake Huron Primary Water Supply System (LHPWSS) undertook a study to examine and predict the remaining useful life of its 30 mi (48-in.) PCCP pipeline. This pipeline was previously assessed using electromagnetic technology to determine broken wire wraps as well as continuously monitored for breakage with an acoustic fiber optic (AFO) cable along the length of the pipeline. The original study in 2017 focused on utilizing this inspection and monitoring data to develop a predictive model to forecast when each of the nearly 10,000 pipes will reach their respective "limit state" based on structural analysis of the pipeline. The inspection and monitoring data acquired from the pipeline was used as a case study to calculate deterioration rates across the length of the pipeline and provide the Water Supply System with information regarding when individual pipes may reach their failure limits so that long-term management strategies could be evaluated. Since the initial modelling in 2017, five more years of data have been collected along this pipeline from the AFO system. Additionally, the LHPWSS has collected high-resolution transient pressure data from points along the pipeline to determine the potential effects that the pressure changes may have on (1) pipeline damage; and (2) changes in pumping configurations. This paper is a comparative study of the initial analysis and the updated study based on the new data collected. In particular, the new analysis highlights how sections of the pipeline which resulted in high probability of failure in the 2017 study reacted to operational changes that modified the pressures in these areas in the intervening five years. The LHPWSS' data shows that its transmission mains (like others in North America) have very low rates of damage. Regardless of any stated design life, these assets, like the Golden Gate Bridge, are not assets to be replaced after a set number of years but can be maintained virtually in perpetuity. To this end, in the near-term, the updated predictive modelling informs the urgency associated with an individual pipe segment that is showing signs of deterioration such that the LHPWSS can plan how quickly it needs to intervene and replace a pipe. In the long term, LHPWSS uses the predictive modelling information to forecast the number of pipe segments that are expected to be replaced over the capital planning period and incorporate this information into the asset management planning programme and budget development process.