skip to content

Centre for Digital Built Britain completed its five-year mission and closed its doors at the end of September 2022

This website remains as a legacy of the achievements of our five-year foundational journey towards a digital built Britain
 
Sarah Hayes

DT Hub Blog September 2019

Sarah Hayes, author of the National Infrastructure Commission’s ‘Data for the Public Good’ (2017), reflects on progress and possibilities on the development of a National Digital Twin following her keynote at the recently held National Digital Twin Day. At National Digital Twin Day, CDBB and the Institution of Civil Engineers (ICE) brought together digital pioneers from government and industry to provide an update on digital twins, the NDT Programme and to celebrate the launch of the CDBB’s Digital Twin Hub (DT Hub).

How did you get to work today? How was your journey? Could you have taken an alternative route or tried a different mode of transport that might have been quicker? What would have happened if there had been a power outage or a security alert during your journey? Would it have made you late or even prevented you reaching your destination altogether?

Many of us use Google Maps to help us figure out how to get where we need to be and, with familiar routes, we might know what delays to expect. I think this is what a digital twin model could do – it would take Google Maps, which is a kind of digital twin model, and combine it with transport models and wider knowledge and probabilities of disruption to offer options on how best to complete your journey. If the situation were to change, the model would advise the next best alternative.

Now, you might say Google Maps already does this. But when driving using a satnav or route- planning app, how many of us actually follow directions when a change of route is suggested? I live in Cornwall and know that if Google Maps offers a new route it will invariably send me down a country lane, where I will also inevitably meet a camper van requiring me to reverse back up a windy hill, possibly scraping the side of my car during the manoeuvre and definitely increasing my stress levels.

I don’t always follow a change-of-route direction because the satnav cannot foresee the potential negative consequences I have come to know from experience. The technology lacks complete information to be able to fully advise me on what alternative route to take.

Launching an innovation requires time for development and learning. Tests are completed in a safe environment and feedback secured to inform improvements. But it’s difficult to experiment with infrastructure because it involves such large costs, equipment and land, which is why digital twin models will be particularly useful.

Imagine building a new, fast, long train line. How could we actually think through all the permutations of what that train line could look like, or what the consequences of that line, both positive and negative, could be? As humans, I don’t think we can, but with the aide of machines I think we could make a better job of it.

Artificial intelligence involves the development of machines that can think. Models able to learn from data can make more informed predictions about the future. So, it could be possible to develop models which can predict solutions to future problems and also help to identify what future problems we will face and what we need to do about them.

I am excited about future possibilities to predict what kind of infrastructure we will need to meet climate change targets. In particular, to increase our resilience in the face of extreme weather events or security threats so that we can simulate and plan suitable responses that minimise damage. When I talk about the predictive capability of digital twins, some people’s eyes glaze over, thinking ‘we haven’t even got a database of where all our pipes and wires are, so how are we going to start predicting what we need in the future?’

It takes some imagination and optimism to believe that a National Digital Twin can be achieved. It requires government funding and support and skills, of course. More crucially, it requires people to make it work. It’s not so much about the technology but the people behind the technology – the data scientists and the AI specialists and the enablers who want to make it work because they see what the benefits of a National Digital Twin might be.

I think the main barriers to the development of digital twins will be human rather than technical – a lack of belief and imagination. Similarly, the catalyst for the development of digital twins is people working hard to make digital twins work for us all. It will be collaboration between research institutions, universities and the private sector which will accelerate our progress towards the National Digital Twin.

It has been exciting to have made a recommendation just a year and a half ago in the Commission’s Data for the Public Good report to move towards a National Digital Twin, to help us understand, manage and plan our infrastructure better, and then to subsequently attend the CDBB’s National Digital Twin Day earlier this month (September). It shows that with hard work and creativity the National Digital Twin is an achievable vision.

Hard work has to be done to ensure we’re all heading in the same direction and speaking the same language, and we need to get on with it quickly if we are to lead the way with our concentration of AI capability here in the UK. A long-term strategy outlining where we’re heading with the National Digital Twin is necessary because this is a long-game, not just a quick win.

We need to continue to work hard. Like learning an instrument, if we put this project down for a while or neglect it due to other priorities, we won’t make progress and will lose out on discovering new ways of doing things. This process is about the journey and the destination. On the way there will be milestones where we find digital twins can deliver benefits for individual companies.

The next milestone might be whole industries benefitting from shared information, perhaps advances in healthcare from simulating the effects of different treatments. Then there could be benefits for whole cities and counties as digital twins become interconnected to minimise energy or water use across entire regions.

The National Digital Twin will provide a bird’s eye view of a city, town or region to help us improve the way we use our infrastructure in that area or to plan and predict what is needed to improve our quality of life. That said, I think it’s important to note these models are tools to guide our decision-making – they will not make the decisions themselves.

We can still be the ones to decide how we want to get to work in the morning.