Using Machine Learning to Plan Water Main Replacement

Water Infrastructure is Aging

Water utilities are facing a massive, invisible problem – over a million miles of pipes that are expected to reach their end of life in the next 30 years. This “pipeaggedon” is estimated to cost $1 trillion.

Current methods for planning water main replacement are either very expensive, inaccurate or both. When replacement decisions are not optimized it can impact operational costs and water quality, resulting in headaches and pain for utilities, cities and tax payers alike.

Accurately Predict Likelihood of Failure

What was a subjective, complicated and time consuming process is now streamlined. Fracta improves and optimizes planning in three simple phases:

1. Data Assessment and Cleaning

2. Machine Learning Analysis

3. Using Likelihood of Failure in Planning

Optimizing water main replacement plans saves millions

If your current water main replacement plans rely heavily on break history or pipe age, you are likely replacing pipes that still have decades of useful life left in them. Using a machine learning based solution can drive optimized replacement plans that can save millions by deferring replacement of good water mains.

By focusing on replacing the pipes with the highest likelihood of failure, you can drive down operational costs, prioritize resources and reduce main break rates.

What our clients say

“Fracta created a ranking system in a couple of months which we took 20 years to build ourselves…”