In June 2024, BlueConduit launched a new product, Water Main Predictions. I spoke with CEO Lorne Groe to learn more about the new product and BlueConduit’s push beyond lead service lines.
What pushed you to look beyond lead service lines?
Lorne: At BlueConduit, we’re experts in AI and predictive modeling and we’ve always been focused on helping people make complex decisions in the face of uncertainty.
The first application of our work was around lead service lines. Jake and Eric, our co-founders, saw how difficult it was for Flint MI to make decisions about lead pipe replacement without a clear understanding of where the lead was and how much they had. They applied academically-grounded processes for data management and machine learning to this challenge, giving the city direction in their work. The processes they pioneered are now the backbone of service line material identification across the industry; their work even drove the EPA’s guidance for Lead Service Line inventories.
Our expertise is in predictive modeling and using data science to support local government decision makers. We never intended to stay focused on lead only. From the very beginning, we’ve always been looking for other opportunities to use our team, our processes, and our technology on behalf of local governments and, ultimately, the communities they serve.
How did you choose water mains for BlueConduit’s newest product?
Lorne: When we look at new product opportunities, we’re the data nerds. We listen closely to our customers’ needs and challenges. And then we do experiments to see if predictive modeling can work to help solve this particular problem.
We’ve had multiple customers talk to us about helping them predict likelihood of failure for their water mains. As our Data Science team started with the R&D process, it quickly became clear that we could apply a lot of our expertise from lead service lines to the challenge of predicting water main failure and leaks. We are already the industry leader in lead service line predictions, so this new product really lets us lean into our strengths and meet a critical need for our customers.
What is unique about BlueConduit’s approach to water main predictions?
Lorne: To me, I think the most unique aspects are the dynamism and the delivery via Esri.
The ecosystem surrounding water mains is always changing, from weather conditions to soil type to nearby construction, and this dynamic environment influences if and when a water main is likely to break. So our product takes all of this data, both internal water system data combined with external data like weather, and uses machine learning to dynamically generate likelihood of failure predictions for multiple timelines.
The more dynamic we can be, the more dynamically our customers can plan for condition assessments, maintenance, and replacements. And that’s exactly what we want – to equip our customers with the information and support they need to make the most efficient decisions on behalf of their community.
And, of course, we deliver our solutions directly via Esri. By meeting customers where they already work, we maintain their source-of-truth and increase their planning efficiency.
What’s coming next for BlueConduit?
Lorne: For water mains, we’re really excited about our current offering, as well as the roadmap of additional features and functionality we’re expecting to add in the coming months. You can expect to see a lot more options to help parse and interpret the data, as well as more support in the decision-making process itself.
Beyond water mains, we’ve always got a few new predictive modeling product ideas in R&D. We’re also working on some new tools to more explicitly provide recommendations and support utilities as they make replacement plans for lead service lines. So we’ll continue to look for opportunities where it makes the most sense to use predictive analytics and machine learning to help local governments and engineering firms make higher quality, customized decisions in the face of uncertainty.
