How is Machine Learning used by the water sector?

BuntPlanet - Artificial Inteligence

We often hear about Artificial Intelligence, Machine Learning, Deep Learning… and they can seem like very complex topics. At BuntPlanet we are experts on the digitization of the water sector, our team of mathematicians, hydraulic experts and engineers live breath and dream about water supply data. So we wanted to shed some light on how these advances in technology are being applied to the water industry… to bring customers from all over the world a fresh glass of water.

On June 5th we had the pleasure of participating in the 2nd edition of Innovation Day organized by the Canal Foundation in Madrid. Our presentation was about the uses of Machine Learning in the water sector. We got a lot of great feedback from the event as people try to understand how these concepts apply to different industries. Here is a short summary of what we covered.

  • Artificial Intelligence(AI) is a science that encompasses different technologies by which a machine is capable of imitating the cognitive functions of a human: learning and solving problems.
  • Within AI we find Machine Learning, which is the set of techniques that help a machine learn from the data that is analyzed.
  • Going another level further, we find Deep Learning, this is the set of algorithms that are capable of generating responses and acting on them, it would be the closest thing to a human thinking.

What is this technology for?

It is not about replacing humans like we often see in the movies, but rather about helping humans. Humans have a valuable and irreplaceable tool: intuition. However, intuition is not infallible, and when making decisions, especially when dealing with huge amounts of data. AI can help us to confirm an intuition by studying all the variables and ingesting massive amounts of data that would be impossible for a human to cover in a short amount of time.

Machine Learning is based on analyzing large volumes of data, making predictions, classifying, and ultimately, optimizing decision making. Think of it like a spreadsheet, a tool that helps humans make calculated decisions in a short amount of time.

How is Machine Learning used in the water sector?

Thanks to predictive analysis we can better manage the diverse needs of our water supply networks. A water supply network is a complex network of pipes, pumps, sewers and storage facilities that connect us to fresh drinking water. The larger the water supply network, the more complex the system – Machine Learning is an invaluable tool that helps managers anticipate water supply flow when it comes to pumping needs, irrigation needs, or identifying when a pipe bursts in the network.

Intelligent data analysis allows utility companies classify different types of consumers, water end users, businesses or public fountains, it also helps to identify water consumption anomalies, which could mean a leak or a problem in the water supply.

Machine Learning is also an indispensable tool for optimizing management of sewage treatment plants, desalination plants or power supply.

All of these advances in technology allow water operators to answer important every day questions like: When should I renew a water meter? What type of meter is best suited for a particular consumer? Is someone stealing water? Does my network have a leak?

Are you interested in learning more about how AI and Machine learning can help to reduce water loss in the world? Do you manage a water supply network and would you like to better understand how Machine Learning and Hydraulic Simulation could reduce your leaks by half?

Get in touch! (+34 943 223 031 –