Fionn Boyle, strategic innovation lead, Anglian Water, looks at how a digital twin and AI will help keep the taps flowing in the future.

While you might expect a telecoms provider to be a big user of AI, automation, and data science, many might be surprised if a water company did likewise.

In fact, a project centred on Anglian Water, which is being done in partnership with a wide range of water utilities, universities and technology companies, is setting the standard for how smart infrastructure based on AI and digital twin technology can be applied in the water industry and beyond.

The impetus for this project comes from how the water industry recognises it needs to improve its water delivery infrastructure. Despite significant work being done, the UK’s network of water mains is leaking three billion litres of water a day, or the equivalent of 1,200 Olympic sized swimming pools. When the complex changes around climate change and population growth are added to the mix, there is a real risk that parts of the UK could run out of water within the next 20 years if nothing is done.

The challenge for water companies like Anglian Water is how they find themselves firefighting issues due to a limited visibility of the larger picture across their complex systems. There is a pressure to understand where there is the greatest need for investment in their networks. The answer is a more systematic data-led approach to understanding, planning, and managing the water supply infrastructure between the water treatment works and customers’ properties.

To achieve this, the innovation team at Anglian Water is mid-way through a partnership project called Safe Smart Systems to understand how data analytics, automation and AI could enhance the resiliency of the water supply service that it provides, address the risk of future water shortages, and help protect the environment. The cathedral city of Ely and the surrounding area in Cambridgeshire is the test bed for the project.

A major component of Safe Smart Systems is how Anglian Water, and its partners are developing an AI decision engine which will be used to control and automate the supply of water for Anglian Water customers in the Ely area. This will involve the engine detecting an anomaly and autonomously opening and closing valves that are currently controlled manually. The engine would take advantage of data gathered and analysed from an array of connected sensors and control systems linked to its physical assets.

While the project’s potential for intelligent automation captures the imagination, the real engine for change within Safe Smart Systems is the building of a comprehensive and super-accurate digital twin for water supply operations. Specifically, the team is building a water information management landscape to accurately structure and maintain its data flows in simulating the entire water supply infrastructure.

By developing an information management landscape, the project will enable Anglian Water to conduct effective information management across its entire built environment throughout its lifecycle. Given how water supply is a prime example of critical national infrastructure, another intrinsic part of the project is securing these computer-controlled and automated systems and processes from cyberattacks.

With secure, resilient interoperability of data, the digital twin of Anglian Water’s water supply operations accurately mirrors the real-world infrastructure. The significance of the digital twin is it is part of a national UK digital framework initiative originally set up by the Centre for Digital Built Britain. Now successfully wrapped up, this was a partnership between the Department of Business, Energy & Industrial Strategy, and the University of Cambridge to understand how the construction and infrastructure sectors could use a digital approach to better design, build, operate, and integrate the built environment. The Anglian Water digital twin will be providing fundamental learning for digital twins in all sectors and many of the biggest water utilities in the UK and internationally are collaborating with Anglian Water on the project.

So, what could be the potential benefits of deploying the digital twin and Safe Smart Systems overall?

At a fundamental level, it can give water companies like Anglian Water a better situational awareness through mapping the water supply infrastructure, its physical assets, and the connectivity between them. The next level of maturity is where the digital world accurately represents the physical through near real time automated data. This would allow engineers to use the digital model to understand how the physical system is performing and simulate how it might respond to an intervention such as fixing a leak. This playbook tested in the digital twin can then be applied to its physical twin.

Where the project becomes most impactful is when there is a bidirectional flow of data between the physical and the real world. This creates a feedback loop through automated data flows so that the digital twin can interpret the physical network’s state in real time and make decisions about water supply that can drive faster actions than would be possible with human intervention.

Anglian Water and its partners envisage three major ways Safe Smart Systems will change how the water supply infrastructure is operated.

Firstly, it will enable proactive maintenance that predicts where repairs need to be done ahead of time. Secondly, there is the possibility of instant response that uses AI and machine learning to automate how Anglian Water detects leaks or a water quality event.

The final transformation would be the most ground-breaking. The level of autonomous operations could be so high that there is the potential for Safe Smart Systems to support self-healing water supply networks. Having automatically sensed a leak that has happened in real time, the network would automatically reconfigure itself to either completely mitigate or minimise its effect by diverting water supply to prevent wastage. This has never been done before at this scale across the world, but the Safe Smart Systems project is a critical building block to achieving this goal, enabling others to follow suit.

Safe Smart Systems is well on course to be completed in 2025. The lessons being learnt in its design development and deployment are already becoming instructional for how the water industry and other sectors embrace digital transformation in how they design, build, and run infrastructure in the UK and beyond.