The planning problem: Water utilities track patterns in consumption data to discover unusual events such as leaks. Unlocking this valuable information is a slow and labor-intensive process. An early-warning system can save time and water, but would require real-time automated analysis. The uncertainty: smart water meters send events (consumption rates), which aggregate various water-consuming processes ― which processes are happening?

To help utilities enable new services that save drinking water, Seita is developing a pattern detection and classification toolbox to provide real-time automated analysis for monitoring platforms.

The toolbox will be used by MeterInsight to deliver labelled history reports to businesses with smart water meters. This project is partly funded by the Province of Noord-Holland.