Predictive modeling helps reduce unknowns and avoid inflated replacement rates before the November 2027 deadline.

The Challenge

Utilities face a November 1, 2027 deadline for baseline service line inventories under the LCRI. Without identifying the unknown service lines, thousands of "unknowns" would count as lead for replacement calculations, even if most are likely non-lead.

The Solution

Machine learning predicts which unknowns are lead and non-lead based on data such as building records, geographic patterns, socio-economic data and installation dates. It's especially valuable if you have more than 5,000 unknowns and suspect most aren't lead.

Our experts Joanna Cummings, Kristin Epstein and Sandy Kutzing note that models typically cost around $200,000 for 50,000 service lines. At $500 per test pit, you break even after avoiding just 400 inspections, and most utilities save far more.

Start Now

With two years left, early action means better predictions and confident compliance. Ready to explore how machine learning fits your system?