A mid-sized Mid-Atlantic utility with a reputation for taking a proactive and focused approach to continuously improving service reliability for their 270,000 customers was facing an all too common situation. The more than 1,000 miles of water mains across their system, had an average age of about 50 years. This had led to an increase in water main breaks, and so they started seeking innovative strategies that would improve service reliability while minimizing repair and replacement costs.
With water main breaks increasing, the customers served by the utility were challenged with unpredictable service outages and costly repairs, as well as highly disruptive road closures. They desired to take a more proactive approach to prevent main breaks and improve their customer level of service (LoS) by focusing on the pipes that needed the greatest attention. Previous experience in working with Xylem to manage their PCCP (prestressed concrete cylinder pipe) inventory led the utility to seek out a better replacement prioritization strategy than traditional techniques such as age and break history.
“Our goal is to leverage machine learning to identify variables that could lead to pipeline failure and, as an outcome, support our capital improvement planning.”
Project Manager, Department of Public Works
In 2014, the utility partnered with Xylem to develop and pilot a data-driven and actionable risk model to evaluate the health of their water system. Since then, the growing partnership has led to one of the industry’s most comprehensive AI-based pipeline analysis models, while also providing improved condition and event data collection techniques.
The utility and Xylem worked together to implement a quantitative risk model combining probability of failure (when a pipe will most likely fail) with the consequence of failure (the social, financial and environmental costs of the failure). Xylem’s machine learning solution forecasts the probability of when each water main in the system might fail using multiple data inputs including results from prior condition assessments. As more data is collected over time and changes are made to the system, the GIS and machine learning algorithms are updated to give a continuous understanding of the overall health of the system. This data-driven approach provides substantial advantages over traditional subjectively scored models where outputs often remain static even when inputs are updated. Xylem’s unique AI-based risk model is able to update results on-demand as new information is collected from the system, including break, condition and operational data. Results allow clients to prioritize and stage pipeline replacement, lowering costs and reducing customer impacts by targeting the most critical and deteriorated pipes.
Additionally, Xylem provided a mobile field event tracking application (capturing information on pipe breaks) for the utility’s field operators. This value-added feature not only increased the accuracy of break data records, it also reduced the overall labor time required to update their CMMS and GIS, and improved pipe failure predictions.
The success of the pilot program will allow the utility to develop focused and cost-effective pipeline renewal strategies for the County. Once the machine learning approach is implemented for the entire distribution system, this model can help the utility lower their annual costs related to pipeline replacement from $90 million to just $20 million while achieving a dramatic four-fold reduction in failures.