How do you diagnose and solve the problem of a failing physical asset which is located 200,000 miles away in outer space?
Thirty or so years before the term ‘digital twin’ was officially coined, NASA’s solution for fixing the damaged Apollo 13 spacecraft is thought to be one of the earliest examples of digital twin usage.
A digital twin is a virtual representation that serves as a real time replica of a physical object or process. It comprises of three elements: the physical item in a real space, the digital twin in software form and the data that links these first two elements.
The twin could mirror a factory, a mining shaft, a human heart, a city, a group of cities or an entire planet.
Back in April 1970 to salvage the Apollo 13 mission and bring its crew safely back to Earth, the digital twin was focussed around the 15 simulators which were used to train NASA astronauts and controllers in every aspect of the mission, including failure scenarios.
The simulators themselves were not the digital twin, but the way the NASA mission controllers used them to rapidly adapt and modify the simulations, to match conditions on the real-life crippled spacecraft, so that they could research, reject, and perfect the strategies required to bring the astronauts home, makes it the archetypal digital twin use case.
“What this means is that a digital twinning experience is not derived from one source of data, but from a variety of data that mirrors and maps and helps us understand what has been experienced by the digital twin.”
Speaking at IoT Tech Expo Europe a couple of months ago tech entrepreneur and British Computer Society fellow Joe Calder, explained during a session on digital twin deployments why he thought that this very first example of digital twinning was useful for examining some of the fundamentals:
“We learned that digital twins are most useful for physical devices that are out of direct reach at least part of the time; We learned that connectivity is important – digital twins rely on constant feedback of data and the situation of the physical asset; We learned that adaptability is important – digital twins need to be able to be updated to mirror changes in the physical asset,” he said.
According to Calder, this first use case was also a masterclass in responsiveness: “It showed us how digital twinning can be designed to scale for rapid adoption and re-engineering – in the case of Apollo 13, this all occurred over just three days.”
The threading of data from different sources during the Apollo 13 rescue mission was also instrumental – and Calder argues that true digital twins are comprised of “multiple interactive models that can be combined to bring together an overall view”.
He added: “What this means is that a digital twinning experience is not derived from one source of data, but from a variety of data that mirrors and maps and helps us understand what has been experienced by the digital twin.”
Most modern digital twins involve a remote physical asset connected to a digital model through a continuous stream of data which might be derived from IoT, AI or big data sources.
For many industrial use cases, data usually comes from operational technology (OT) sources – that is data from the hardware and software, monitors and sensors that keep assets like factories and power plants running.
One of the current challenges in digital twin deployment is the interoperability of all these data sources, according to Dan Issacs, CTO of the Digital Twin Consortium, who also spoke about digital twin deployments at IoT Expo.
“Different areas of a businesses all have their own data silos, all installing sensors running AI against predictive models and other tools to solve their problems. it’s estimated that over 50% of the project costs [for a digital twin] are from the integration alone,” he says.
Calder adds that for businesses looking to build a successful digital twin, being able to find ways to combine their IT and OT data sources is crucial.
“IT and OT generally speaking are not usually combined but combining and aligning them will be invaluable for organisations developing and deploying digital twinning experiences” he says.
Calder claimed that modern digital twins also need to be able to find ways of connecting to larger Internet of Things experiences and opportunities.
“IoT is so much more prevalent now than it was for example, five years ago, there are tremendous opportunities for metric data to be gathered and for us to leverage those experiences,” he says.
Despite these challenges, there are plenty of examples of digital twins being used today across a broad range of markets. For instance, every Tesla vehicle sold has digital twin, which is designed to provide a continuous update on the state of the vehicle and each driver habit and its environment.
As well as the automotive market, healthcare industry is starting to embrace digital twins to improve personalised medicine, healthcare organisation performance, and new medicines and devices.
One large scale example that involves a number of stakeholders coming together is The Living Heart Project – launched in 2014 to crowdsource a virtual twin of the human heart.
The project has evolved as an open-source collaboration among medical researchers, surgeons, medical device manufacturers, and drug companies to serve as a common technology base for education and training, medical device design, testing, clinical diagnosis, and regulatory science.
The company also has a ‘Living Brain’ project which is guiding epilepsy treatment and tracking the progression of neurodegenerative diseases.
Another current example of digital twinning is in the running or smart cities. One of the launch members of the Digital Twin Consortium, Leeds-based start-up Slingshot Simulations, is currently leading a UK-government backed project that is using digital twin technology to improve the transport network in Leeds, York, and Hull.
Slingshot’s platform aims to auto-generate 3D models of all three cities and surrounding areas, allowing planners to test different ways to boost the capacity of the existing network, reduce congestion and air pollution.
The Yorkshire Geospatial Twin Partnership, which launched last year, unites Slingshot with the three local authorities that are all brining their own data to the table, as well as that of locally based businesses.
Data from the UK Department of Transport is also being used and other stakeholders include engineering and design group Arup – which is lending its city modelling and transportation planning expertise – while BT is providing the mobility data.
Minister of State at the Cabinet Office Nicolas True has estimated that the project will help unlock £2bn of economic value in the transport sector by improving services, enabling the efficient delivery of new networks and new transport corridors.
Industry is beginning to optimise the digitisation that has already made its processes and operations more efficient. The growing trend towards automation and data exchange in technology and processes within manufacturing is often referred to as ‘Industry 4.0’ and digital twins are set to play a crucial role in this, according to Issacs.
One example in this sector, cited by the Digital Twin Consortium’s Natural Resources Working Group, involves the use of real time data to optimise and maintain belt conveyor systems in mining.
One of the consortium’s members deployed a long conveyor monitoring system to create a digital twin of a 50-mile-long underground conveyor belt that would take a human being two hours to traverse in sub-zero temperatures.
“Over a six-month period using a digital twin it identified over 180 hours of preventable borer downtime that equated to approximately $4 million dollars of lost revenue,” Issacs reported in his presentation.
If NASA was the pioneer of the original digital twin, another space agency – the European Space Agency – is aiming to take digital twinning to the next level.
Rather than saving a spacecraft, ESA’s long-term mission is to save the Earth – through a complex series of predictive monitoring systems.
The ESA is currently working on a full stack for a digital twin of the Earth to help visualise and forecast natural and human activity on the planet. It is hoped that the model will be able to monitor the health of the planet, perform simulations of Earth’s interconnected systems with human behaviour and help support European environmental policies.
The predictive Earth Digital twin involves the bringing together of several different digital twins that monitor a range of activities across the different elements of which the planet comprises: Forest, Hydrology, Antarctica, Food Systems.
The digital twin of Antarctica, for example, will harness satellite observations, numerical simulations, and AI to track the whereabouts of melt water on and under the ice sheet, and to explore how fringing ice shelves melt in various scenarios.
A food systems digital twin, meanwhile, will simulate agricultural activities and interactions within ecosystems daily.
Drawing on multiple interactive models that can be combined to bring together an overall view, the Predictive Earth Project, like the Apollo 13 model 52 years previously, fits Calder’s definition of a ‘true’ digital twin perfectly.
Subscribe to our Editor's weekly newsletter
Share
