The concept of digital twins has been around since the 1970s, but more recently we’ve started to see rapid adoption of the technology in a number of scenarios. With organizations in energy, transport, manufacturing, medicine and others creating digital twins to help understand how specific processes, instruments, even people will react to changing situations.
What are digital twins?
What are digital twins? Simply put a digital twin is a virtual replica of physical thing – whether that’s a process, product or even a person. The digital twin is created using real-time data from connected sensors on the physical asset, which can then evolve using machine learning and simulation. Digital twins allow us to see what’s going on with the physical asset in the real world right now, but also how it might react in the future too. While it sounds like the stuff of Sci-Fi – it really isn’t. It’s here and now.
IoT sensors are already collecting data on real-world machines for analysis in cloud environments. Building a digital twin for each machine is a cost-effective addition to an already connected model and will help businesses like manufacturers to improve machines and processes in the supply chain.
According to GlobalData there are three key technology trends currently driving the adoption of digital twins.
5G into 6G – as the networks are rolled out and embraced, digital twins will be able to take advantage of the high-speed wireless connectivity they offer without ‘temporal or spatial’ constraints. This will deliver data in real time back to the virtual twin
Digital relationships and partnerships – those organizations working in specific sectors are working together to deliver concepts for digital twins. This is already happening in the energy industry for example, where the partnerships ‘help oil and gas companies to improve asset performance
Personalized medicine – digital twins are widely used in testing to validate drug candidates. It won’t be long before we see a digital twin for patients created and then tested with 1,000s of drugs to see which provide the best result
How can I use digital twins?
There are numerous benefits and applications of digital twins. Business Intelligence (BI) leaders can use digital twins to optimize IoT applications and get the most out of their data by clarifying what IoT application will deliver the most impact to their physical counterpart.
Digital twins can also streamline the delivery of real-world IoT data. With digital twin systems designed to match the physical machines, data latency is minimized.
For those in engineering, manufacturing and testing, digital twins are a game changer. They let you apply and test changes to the digital asset before they’re applied to the physical version. This lets you see the impact and costs of other connected systems and the environment the machine works in. Predictive monitoring and maintenance of remote IoT assets or sensors is simplified as well. Using the virtual replica, you can review and predict likely faults or errors before they occur.
When combining the twin data with business rules, optimization algorithms or other prescriptive analytics technologies, digital twins can also support human decision making or even automate decision making.
This is especially useful in IoT environments, where data from machines can be fed directly into the integrated digital twin and business intelligence solutions. IoT sensors allow BI solutions to pinpoint problem areas and opportunities to optimize, before developing a comprehensive solution on which manufacturers can act.
What’s their future application?
While currently digital twin solutions are being adopted to ensure the integrity of a physical asset, it’s likely in the future that we’ll witness integration of the twins themselves, as we need to map increasingly integrated application environments.
If that sounds complicated, it’s not. It’s through edge analytics solutions – specifically, IoT-edge computing frameworks – that companies will be able to realize their ideal digital twin integrations. Using IoT Edge Frameworks – like Eurotech Everyware Software Framework – can help to simplify the concept of digital twins of real assets.
Designers, engineers and producers of machines have long recognized the value of being able to simulate physical assets on software on a computer screen. NASA scientists first looked at the concept fifty years ago, but with the increasing adoption of IoT and edge computing the value of digital twins in understanding and identifying product (and human) reactions is easier than you think.
Contact us to find out more about how digital twins can help your business.
Industrial automation is undergoing a digital revolution as ‘Industry 4.0’ and Digital Twin technologies introduce new and innovative ways to execute manufacturing processes.
With IoT, artificial intelligence (AI), machine learning, and deep learning at the forefront, manufacturers will have spent close to $70 billion on these new technologies by 2020 to meet new demands.
Among these technologies, Digital Twin is changing how industrial manufacturers approach product design, operations, and post-sale services.
“Digital Twin solutions open new opportunities for industrial firms to speed up their Big Data- and AI-driven innovation transformation efforts,” Forrester reports.
A combination of vendor marketing and education—along with real-world benefits realized in Digital Twin use cases—are driving Digital Twin adoption among industrial manufacturers as they become an integral part of companies’ IoT and digital strategy investments.
What is a ‘Digital Twin’?
“Digital twins — digital representations of people, processes and things — will become more connected, providing a rich and real-time view of company processes and assets. Enterprise architecture and technology innovation leaders must plan for a diverse environment with integration challenges.”
Gartner, Top 10 Strategic Technology Trends for 2019: Digital Twins, 2019
A ‘Digital Twin’ is a digital model of a real-life object, most often a physical machine. Gartner defines a digital twin as “a software design pattern that represents a physical object with the objective of understanding the asset’s state, responding to changes, improving business operations and adding value.”
For industrial manufacturers, a digital twin allows them to create digital models of real machines, then combine both digital and physical data to deliver a better understanding of the performance and potential value of those physical machines.
This applies to all aspects of the machine’s functionality—production, maintenance, user experience, and more.
Each Digital Twin is used as a proxy for a real machine, allowing manufacturers to test concepts and hypotheses before applying them to physical machines themselves.
This digital technique—called ‘encapsulation’—allows manufacturers to adapt the digital ‘machine’ before applying changes to real hardware, which is costly and may impact other connected systems.
Digital Twins therefore have a wide variety of benefits and applications. Manufacturers can simulate plans, processes, and systems with zero risk, helping them visualize any variety of new strategies—and their results—before investing real resources in their development.
Conceptually, Digital Twins are simple. There is one Digital Twin ‘per thing,’ and that Digital Twin is updated alongside the ‘thing.’
This allows for the best possible results during experimentation. But despite its simplicity, Digital Twins have a wide range of applications, and are suitable for just about any machine model.
How to use Digital Twin for industrial automation?
Digital Twins have proven especially useful when integrated with IoT systems in industrial manufacturing.
“We see digital twin adoption in all kinds of organizations,” says Benoit Lheureux, research vice president at Gartner. “However, manufacturers of IoT-connected products are the most progressive, as the opportunity to differentiate their product and establish new service and revenue streams is a clear business driver.”
This is especially critical as IoT becomes more commonplace in industrial manufacturing environments. IoT sensors are already collecting data on real-world machines for analysis in cloud environments—building a congruous Digital Twin with each machine is a cost-effective addition to this already connected model, and can only help manufacturers improve machines and processes in the supply chain.
“The Digital Twin is not a new idea: Designers and makers of machines have long recognized the value of being able to simulate those machines on a computer screen. But growing enthusiasm for connecting machines to the Internet of Things (IoT), and for delivering services instead of only selling products, means that the Digital Twin is becoming relevant to a broader audience.”
Forrester, “Digital Twins Combine Enterprise Data And IoT To Drive New Business Value,” 2019
In this way, Digital Twins significantly reduce the cost and work required to maintain and enhance physical things.
AI, machine learning, and deep learning allow for increased automation in these processes as well. Manufacturers can introduce new concepts in digital environments, then count on digital tools to both optimize them and execute them among their physical counterparts.
Digital Twin and industrial automation use cases
Industrial automation streamlines processes in multiple manufacturing environments and use cases, and its integration with Digital Twins is no exception. Humans can manipulate digital environments, suggest or make changes, and introduce additional resources to optimize Digital Twin (and subsequently machine) performance.
Industrial automation allows for these processes to unfold automatically in digital environments before applying new revelations to physical ones. This relieves humans of these responsibilities but safeguarding equipment until automated solutions are confirmed in digital environments.
As Gartner describes, “When combining the twin data with business rules, optimization algorithms or other prescriptive analytics technologies, digital twins can support human decisions or even automate decision making.”
This is especially useful in IoT environments, where data from machines can be fed directly into integrated Digital Twin and business intelligence (BI) solutions. IoT sensors allow BI solutions to pinpoint problem areas and opportunities to optimize, before developing a comprehensive solution on which manufacturers can act.
In one scenario, a Digital Twin for industrial robots was implemented. The Digital Twin allowed manufacturers to program the robots offline for multiple scenarios, where virtual commissioning of the system could be tested.
This dramatically speeds up testing processes and minimizes downtime for the physical machines themselves, “resulting in a higher availability as well as a higher reliability with less failure risk.” The introduction of automation into this scenario rapidly speeds up the virtual commissioning process, delivering results faster and with greater efficacy.
As machines provide real-time feedback via IoT into a Digital Twin, manufacturers can use the Digital Twin to automated predictive aspects of machine performance and maintenance.
In another scenario, the Digital Twin of a locomotive can analyze real-time and historical data and provide recommendations about on-board system status for predictive maintenance and remote monitoring, with minimal human oversight.
The shared evolution of industrial automation, IoT and Digital Twin technologies
“The rise of Digital Twins coincides with the rise of the IoT. When buying machines and other assets, support for Digital Twins and continuous development of twin capabilities should be a selection factor.”
IoT empowers industrial manufacturers to learn more about their equipment and better understand processes in real time. However, IoT adds a new layer of complexity to manufacturing environments, where the optimal application and investment strategy for IoT sensors isn’t always clear upfront.
Industrial BI leaders can use Digital Twins to optimize IoT applications and get the most out of their data. Unlike incongruous systems, Digital Twins help manufacturers improve their IoT strategy, clarifying what are the highest-impact IoT applications to render desirable results in Digital Twin environments that reflect their physical counterparts.
Digital Twins streamline the delivery of real-world IoT data as well, where Digital Twin systems are designed to be congruous with those of physical machines. This is a desirable alternative to discerning IoT data in other analytical environments, where the data must be processed to suit the system.
Manufacturers and other companies are realizing the value of marrying IoT and Digital Twin technologies. Already, 13% of organizations implementing IoT projects use at least one Digital Twin; 62% are either establishing or plan to establish a Digital Twin alongside IoT processes, Gartner reports.
Optimizing Digital Twin with manufacturing intelligence
“Rapidly expanding Internet of Things and analytics programs powered by machine learning and artificial intelligence — as well as improving models and simulations — mean that Digital Twins are coming to fill a number of roles. And they’re going to make life better.”
Forbes, How Orlando’s Work with Digital Twins May Change How We Engineer Everything,” 2019
Industrial manufacturers have already identified dozens of successful use cases for automated and BI tools like machine learning, AI, and deep learning systems.
Digital Twins lend themselves to these environments as they connect seamlessly in existing machine applications. They simply serve as an alternative to physical machines in the application of these technologies to optimize results.
Scholars define this application as an ‘Intelligent Digital Twin’. It “consists of all the characteristics of a Digital Twin as well as artificial intelligence to realize an autonomous system”:
The Intelligent Digital Twin can therefore implement machine learning algorithms on available models and data of the Digital Twin to optimize operation as well as continuously test what-if-scenarios, paving the way for predictive maintenance and an overall more flexible and efficient production through plug and produce scenarios.
Automation Technology, “An architecture of an Intelligent Digital Twin in a Cyber-Physical Production System,” 2019
The introduction of automation allows manufacturers to increasingly remove human beings from processes where they are no longer required, instead applying human resources only to mission critical areas in much larger processes.
In this way, Digital Twins can realize their true potential—simulating vast industrial environments with intense precision, and without increasing workloads for human workers.
Conclusion: Digital Twins and our automated industrial future
Gartner predicts that by 2022, over two-thirds of companies who have already implemented IoT will have deployed at least one Digital Twin.
Today, Digital Twin solutions are most often delivered in coordination or even by the manufacturers of physical equipment in which companies invest. But while this model makes sense today, companies will increasingly look to integrate Digital Twins with one another to accommodate each ‘thing’ in increasingly integrated manufacturing environments.
It’s through edge analytics solutions—specifically, IoT-edge computing frameworks—that industrial manufacturers will realize their ideal Digital Twin integrations.