Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Water Management Ai topic
No spam. Unsubscribe anytime.
Committee approves Vanderbilt-Metro pilot to use AI on water system data to detect anomalies
Summary
The Transportation Infrastructure Committee voted 9-0 to recommend approval of a pilot project with Vanderbilt University and Metro Water Services to apply anomaly‑detection tools to existing water system data as an early warning detection system for valves, leaks and other operational anomalies.
The committee recommended approval, 9-0, of BL 2025-912, a pilot agreement between Vanderbilt University and the Metropolitan Government of Nashville and Davidson County, acting through Metro Water Services, to test early‑warning anomaly detection tools for the water management system.
Why it matters: The pilot seeks to analyze pressure, flow and work‑order data already collected by the utility to identify anomalies — such as valves that are unexpectedly open or closed or signs of leaks — and to provide earlier alerts to operations staff. Committee members asked whether the pilot would require new physical sensors; Metro Water Services staff said it would not.
Mary Peoples, Metro Water Services council liaison, introduced the item and the department called Walter Ashford to answer technical questions. Ashford explained the pilot’s scope: "We have a lot of data that we take in, pressure data, flow data, some work order data. And what we're gonna do is we're going to look at how AI can kind of look through this data to provide kind of an early warning system for anomalies in the system," he said. He added the work relies on analysis of existing data rather than installing new hardware.
Outcome and next steps: The committee recommended approval 9-0 and the pilot will proceed under the terms of the agreement; Metro Water Services will oversee implementation and provide periodic updates to the committee.
Ending: The pilot is intended to test whether advanced analytics can reduce response times for leaks and valve issues by giving operations earlier, data‑driven alerts.
