Leaks from water distribution systems are significant. Around 20% of treated, potable water is lost after entry into the distribution network https://www.ofwat.gov.uk/reducing-leakage/. Data (well-defined, accessible, usable, etc.) is needed to provide accurate water demand predictions, early warnings for possible system failures and of potential supply shortages. Data is also critical for effective AI tools that are the future of proactive leakage management in water distribution networks. Real-time sensor data on water flow and pressure is a priority whilst data on the weather, soil characteristics, demographics, pipe conditions, repair logs, etc. can improve accuracy of AI insights.
Methods for AI are already emerging but water authorities must overcome barriers to data collection, data quality, data definition, and data sharing before insights from AI and computational models can contribute to decision making and operations.
The Water Systems Leakage (WSL) project will contribute to Data Infrastructure for National Infrastructure (DINI) by highlighting the data barriers and opportunities, providing potential data descriptions (covering quality, semantics, and logic of data), and showing the potential for data infrastructure to support water infrastructure by reducing water leakage.
WSL are keen to hear from water authorities regarding barriers to data collection, data quality, data definition, and data sharing. Please contact l.varga@ucl.ac.uk if this project is of interest to you.