PUB's Digital Twin and Smart Sensors — Singapore Cuts Water-Pipe Leaks to 5 per 100 km a Year

PUB, Singapore's National Water Agency, has spent nearly a decade layering sensor networks and analytics onto its roughly 6,000 km drinking-water pipe network to cut losses in a country with no natural aquifers to fall back on. From 2017, PUB trialled 120 EchoShore-TX acoustic sensors made by Canadian firm Echologics along about 70 km of large DN700 steel water mains, with data streamed to a cloud server where algorithms searched continuously for the noise signatures of developing leaks.
The sensors proved their worth in 2018, when the system flagged a leak on a DN700 main that field crews confirmed as a 13mm corrosion hole within 500mm of the location the software had predicted — catching it before it grew into a burst. On the back of that result, PUB extended acoustic and ball-shaped in-pipe sensors to survey roughly 500 km of mains over five years, and has worked with digital-twin software (including Bentley's OpenFlows platform) to combine machine-learning anomaly detection with hydraulic simulation models across its network.
PUB now reports about 5 leaks per 100 km of pipe a year, a figure widely cited as among the lowest non-revenue-water rates of any water utility in the world. The reporting on this figure comes largely from trade press and vendor case studies rather than a PUB-published evaluation, and it reflects the cumulative effect of decades of asset renewal, not the sensor-and-AI layer alone — so the specific contribution of the AI components is not independently isolated in the available sources.

Read the full analysis: https://smartwatermagazine.com/news/pub-singapores-national-water-agency/pub-adopts-smart-tools-detect-and-pre-empt-pipe-leaks