​​​​​​​​​​​​​​Data-driven ​Water Pipe ​Failure ​Prediction

A project by scientists from CSIRO's Date61 group that helps to predict which water pipes need maintenance before they burst has won the Excellence in Data Science category at this year’s Eureka Prizes.

by Stern Curator

The Eureka Prizes have been awarded annually since 1990 by the Australian Museum in partnership with a number of sponsors and supporters. They are given for excellence in the fields of research & innovation, leadership, science engagement, and school science.

Image source: Video Print Screen (Find the Video Below)

According to Data 61, Australian water utilities spend $1.4 billion per year on reactive repairs and maintenance, and predictive techniques to enable their efforts to be focussed on preventative repairs have the potential to save the water industry $700 million annually.

Data61 says it has developed an advanced failure prediction tool using non-parametric machine learning techniques, with which it expects to double the precision of prediction within this one percent of the network that is inspected.

Data61 has developed the tool by analysing data from a massive number of pipe failures. It has worked with water utilities and research organisations worldwide, including the United Kingdom Water Industry Research (UKWIR) to analyse data from 27 utilities around the world, nine million pipes spanning 525,000 kilometres, and 700,000 failure records.

Read full article: WhaTech

Attached link

Media

Taxonomy

• Technology
• Leakage Detection
• Leakage
• Infrastructure
• Leakage Reduction
• Integrated Infrastructure
• Urban Water Infrastructure
• Infrastructure Management
• Utility Pipe Network
• Pipeline
• Water Software
• green infrastructure
• Critical Infrastructure Protection
• Pipes and Pipelines