Skip to the content

Follow us @UKAuthority

What’s needed to create a digital twin?

28/01/19

Mark Say Managing Editor

A new element to the vision for future public services involves the emergence of a multitude of digital twins.

Abstract of twins dancing

It is a concept that is beginning to build momentum, with plenty of people asking questions about what it can do and a few pioneer projects aimed at demonstrating its potential for the public sector and national infrastructure.

In short, it refers to a digital replica of physical assets, processes, people, places and systems, taking in all of the relevant data points. It can be used for monitoring, diagnostics, prognostics and intelligent maintenance and, as they become more sophisticated, predictive modelling.

The concept has made progress in manufacturing and healthcare – in which pioneers are building digital twins of patients to assess what can help their wellbeing – and there is a growing recognition of its potential in planning and infrastructure projects.

Singapore has begun to use a model developed by Dassault Systèmes to look at factors such as energy consumption, airflows and the movements of pedestrians around the city; and Newcastle University and Northumbrian Water are aiming to build one for real time resilience testing of the local infrastructure.

The most ambitious project so far for the UK is the National Infrastructure Commission’s (NIC) plan for a national twin of infrastructure assets, involving simulations of the four main elements – transport, energy, water and telecommunications networks. It is envisaged as a federation rather than one definitive model, and the NIC is looking to support pilot projects to test the concept and ultimately work towards it having a predictive capability.

New ground

What is needed to create a digital twin came under the spotlight at an event staged by IT industry association techUK last week. It pointed to issues that are familiar in digital transformation, but also highlighted that this is new ground on which nobody can expect rapid progress.

One of the main points was that the full range of benefits will not become visible until the model is further developed. Sarah Hayes, senior regulatory adviser at the NIC, said it is difficult to identify all at the moment and that over time some will emerge that for now cannot be envisaged.

In turn, this makes it difficult to assess everything needed. Peter Vale, engineering information manager at Tideway – which is building the Thames super sewer – pointed out that different projects are at different levels of maturity and investment, and that it will take a few years of going through the process to work out what can and cannot be achieved.

But most organisations will need a decent indication of benefits before they are ready to invest. It will help if there is a campaign to widely share any examples as they emerge, and James Kidner, director of partnerships at tech start-up Improbable, said it “needs a good story on how people’s lives have been transformed – something that captures the public imagination”. This could make it easier for board level executives to see what could be achieved.

On a practical basis, it would also help if any digital twin has a strong, high profile anchor customer that others would be ready to follow.

Data practicalities

The practicalities of handling mountains of data, some of its from outdated and messy legacy structures, are going to create another big challenge. One element is in obtaining the data: some of the potential contributors in the private sector are likely to be reluctant to share, especially if they are monetising it.

Some evangelising, routed in the economic benefits, is needed to overcome this and create a mindset that says their long term interest is in sharing rather than embedding it in a silo.

Given the large number of sources it also needs clear guidance on data sharing and standards. As a digital twin pulls data from different organisations there are bound to be problems over compatibility, and it will take some serious resources to ensure it works effectively within the model.

Along with this are the familiar issues of data quality and security, ensuring that it is reliable and tamper proof. Any irregularities could undermine the workings of a digital twin and threaten significant mistakes.

Vale said there have to be people who are directly responsible for this and, a familiar refrain in large organisations, that the workforce has to be trained and made aware of the importance of secure, high quality data.

Gemini factor

Most of the data issues are being addressed by the Digital Framework Task Group within the Centre for Digital Built Britain, a partnership between the Department of Business, Energy & Industrial Strategy and the University of Cambridge. It is working on an information management framework to support the developments and published the Gemini Principles to support the management of infrastructure.

Its deputy director Alexandra Bolton outlined five core streams of its roadmap for the interoperability of data: finding an overall approach to delivering benefits; creating good governance structures and processes; developing guidance, specifications and standards; identifying potential enablers and blockers; and encouraging industry to adopt the framework. It is aimed at a more coherent effort that makes it easier to share the lessons from different projects.

Beneath all that is the need for a common language. As with most concepts or technologies in their early days, people are creating their own terms for same components or ideas, and there is a danger of misunderstandings. Peter Vale said there is a need for a standard language that encourages understanding and can accelerate developments.

Is all this a long way in future. Sarah Hayes said a federation of digital models that shows the current situation is not very far away, but it will take much longer, with advances in machine learning and artificial intelligence, to develop the predictive capability.

Small steps

“Our vision is long term and ambitious, but there are small steps we can take at the moment around companies building their own digital twins for existing operations,” she said. “We want to bring all those together so they are interoperable and we see there is greater value in the whole.”

It is clear that there is still plenty to learn, but the digital twin appears to be a logical progression in how public authorities can use data, especially in planning to create smarter, sustainable environments.

Hayes summed up its potential: “These digital twins will be able do things that we haven’t set out and measured yet, so we need to keep an open mind about what they can do and what the value will be.”

Image: Sérgio Valle Duarte, CC BY 3.0 from Wikimedia Commons

 

Register: Library & Alerts

Keep informed - Get the latest news about the use of technology, digital & data for the public good in your inbox from UKAuthority.