Machine learning helps UK water utility speed up leakage detection

The phrase machine learning may conjure up images to some of super smart robots taking over the world but now a water utility in the UK is using the artificial intelligence (AI) system to reduce water losses.

UK-based water company Severn Trent is the latest to use machine learning to help transform the way it approaches leaks.

The utility said it has so far collected “five billion records of flow and pressure data” which are being used to reduce leakage by over 16 percent.

In a nutshell, the process involves applying advanced analytics to data from the network of pipes collected through sensors to identify, locate and manage leaks.

French digital consulting company Capgemini teamed up with the water utility to deliver the pilot projects.

According to the utility: “In the areas the model has been used, time to find leaks has reduced by over half, engineers are able to pinpoint leaks more accurately, and overall the network in those pilot areas seen a reduction in leakage of over 16 percent.”

Rob Ryder, technology and lab data manager, said: “When using the model we’re getting the most out of our network data, giving us the upper-hand when it comes to locating and fixing leaks.”

Amit Ghosh, group account executive energy and utilities at Capgemini’s UK Business Unit, added: “We brought our deep expertise in data science and analytics, and our pre-built analytical models, which combined with Severn Trent’s subject matter expertise in leakage operations enabled them to fast track the roll out of the solution.”

Related content

• 3 reasons South Korea is becoming a smart water nation
• Five progressive digital water technologies to watch
• It’s time to drop the digital buzzwords in water
• Fatberg detecting 5G sewer ‘nervous system’ goes down under
• Explore all our coverage related to water treatment