Predictive modeling helps water systems classify service line materials as lead or non-lead with minimal digging. But how does it work, exactly? In the infographic below, BlueConduit’s expert team breaks down the process from unknown to classification of pipe materials using predictive modeling using these 6 steps:
Step 1: Identify and quantify your unknowns
Step 2: Collect a representative, randomized sample for field verification
Step 3: Generate predictions using AI and machine learning models and an enriched dataset
Step 4: Set thresholds for classifying predictions as lead/GRR, non-lead, or unknown
Step 5: Update your newly classified lines into your state inventory template
Step 6: Submit your post-project report with methodology details and other state requirements, per state guidelines.
