This article is part of our guest contributor series, featuring insights from Oldcastle Infrastructure. We’re excited to share their perspective with our readers.
The core of any country’s drinking water system is the network of pipes connecting local processing plants with millions of homes and businesses. In the U.S., that network is 2 million miles long. It’s vital, but it’s also vulnerable because much of it is antiquated. Some sections are more than 100 years old.
And it fails, repeatedly. On a typical day, more than 700 pipes break somewhere in the U.S. - that’s roughly one every two minutes. As a result, municipalities lose about 6 billion gallons of abstracted, treated and distributed water daily to leakage from the system. This is waste we can no longer afford. But fixing it won’t be cheap: the Environmental Protection Agency estimates that repairs and replacement over the next 20 years are likely to require an investment of about $625 billion.
Many factors contribute to the complexity of the problem. Engineers working on asset management have to factor in the size of the pipes, the chemical characteristics of the materials used, and the composition of the soil supporting the network.
For example, a 2023 study conducted by Utah State University (USU), Water Main Break Rates in the USA and Canada, discovered a loose correlation between soil corrosion index and the number of breaks for cast iron pipe, and a tight correlation for ductile (flexible) iron pipe. Those are significant factors when repairing or replacing pipes in certain soils.
Although the analysis from the USU study is invaluable in terms of understanding the scale and complexity of the situation, it is an aggregate view that cannot provide local direction to planners and maintenance teams on the ground. And with limited resources, approaches that have been adequate historically are no longer sufficient.
The old techniques of finding an underground leak by hit-or-miss digging through water flowing to the surface are inefficient and expensive, and while teams are searching for the source of the leak the disruption goes on in the form of water shortages and shutoffs. If the flooding knocks out utilities or essential services, the inconvenience and damage get worse.
Pioneering approaches such as Oldcastle Infrastructure’s CivilSense™ offer a better way. After using VODA.ai’s daVinci AI to analyze network data and identify areas of the network that are at high risk of leaks, Oldcastle Infrastructure’s field teams deploy acoustic sensors to gather real-time data. When a leak is detected the AI tells the engineers the size and exact location of the leak, thus facilitating the repairs - accurately and efficiently.
See case studies from utilities using AI in the field →
In short, thanks to new technologies water asset management has an opportunity to evolve from inefficient, reactive response to leaks of unknown origin to proactive prevention, using real-world data to direct activity and drive better decision making.
Even better, the new technologies will move water asset management toward the day when we can replace pipes before leaks actually happen.
This article was originally published on the Oldcastle Infrastructure blog and is republished here with their permission.
Register for our webinar
Join us at 14:00 ET on October 8 for our upcoming webinar, “How utilities can use AI to improve water asset management”.
This article is part of our AI for Utilities series – where we break down what AI really means, how it works, and how it helps utilities move from reactive fixes to proactive prevention. We’re proud to feature guest perspectives, bringing field insights into the bigger picture of smarter water asset management.
🔔 Subscribe to our blog to hear what’s working – and what’s next.