Digital Twin: A Bridge Between Physical and Digital Worlds

June 25, 2019

A digital twin is a virtual representation of a real object or system across its lifecycle, using real-time data to enable understanding, learning and analysing. Data has long played a fundamental role in manufacturing, today the ever-growing influence of the Internet of Things (IoT) is forging a closer connection than ever between the industries’ digital and physical worlds.

Digital twin technology was first used in practice by NASA to mirror and diagnose issues in orbit, which eventually paved the way to fully digital simulations. Today, NASA utilises a digital twin to monitor the entire Space Center. That has allowed NASA to create a digital twin that backs up facility decisions. For example, Flood Impact Analysis provides a 3D model calculation determining how many sandbags are needed for each doorway to keep critical projects out of harm’s way.

In general, digital twins are divided into 3 main categories:

·  Product

·  Production

·  Performance

Digital Twins Categorisation

The combination and integration of the three digital twins as they evolve together is known as the digital thread.

Product Digital Twins: Efficient design of new products

For understanding and showing how your product is acting in the physical world, a digital twin can be used to validate its performance. The product digital twin provides a virtual-physical connection, that allows to analyse how a product performs under various conditions. Furthermore, it lets you make adjustments in the virtual world to ensure that the next physical product will perform exactly as planned.

Even if complex systems and materials are in use, product digital twins help to navigate the complexity to make the best possible decisions in the design phase. All of this eliminates the need for multiple prototypes, reduces total development time of the product, improves quality of the final manufactured product and also enables faster iterations in response to user feedback.

Today, NASA uses digital twins to design next-generation vehicles and aircraft. That helps them to design a vehicle, which is as safe as possible, efficient and reliable. Furthermore, in case of critical situations, digital twin can be used to simulate problems and then find solutions, which can be then applied to a physical vehicle.

Production Digital Twins: Manufacturing and production planning

Simulating a manufacturing process before starting production helps to validate the manufacturing process while time and money on operating the plant. Simulating the whole process using the digital twin and analysing various scenarios allows companies to create a production methodology that is efficient under various conditions.

To be even more thorough, creating digital twins of all the manufacturing equipment enables to optimise the production even further. Connecting and using all of the data from the product and production digital twins, companies can prevent costly equipment downtime and even predict with great accuracy when preventative maintenance will be necessary.

Manufacturing operations that want to be faster, more efficient and more reliable require a constant stream of accurate information. For example, Chevron Corporation is using digital twin technology to predict maintenance problems in its oil fields and refineries. In addition, the high-value equipment will have additional connected sensors to prevent the breakdowns and thereby save millions of dollars each year.

Performance DigitalTwins: Capturing, analysing and acting on operational data

A massive amount of data is generated by smart products and plants regarding their utilisation and effectiveness. The performance digital twin captures all of the operational data and analyses it to provide actionable insight for informed decision making.

By leveraging performance digital twins, companies can create new business opportunities, gain insight to improve virtual models, improve product and production system efficiency and capture, aggregate and analyse operational data. Today, Formula1 teams use digital twins to simulate the cars behaviour under various conditions. That allows teams to gain insight into different elements of the car’s performance and improvement possibilities. In essence, digital twins are used to make the Formula 1 cars race better.

By using continuous streams of real-time data, it’s now possible to generate a digital twin of virtually any product or process. Manufacturers will then be able to predict outcomes more accurately, detect physical issues sooner and build better quality products. For example, while the output of the manufacturing process is a physical object, manufacturing itself begins with data from the design phase. That data is fed to the machines that execute designs – this would be the point of transition between the digital and physical worlds. In addition, more data is collected during manufacturing and also eventual use of the final product. All of the captured data can be extremely valuable for future designs and modifications, while creating an evolving cycle of innovation and improvement.

All of the product lifecycle data pieces stitched together represent a continuous flow of digital thread of data, that ties physical and digital worlds. Aggregated and integrated in real time, creating a virtual replica of a product or process that can reveal significant new insights and enable digital twin by providing the data it needs to function.

Digital twin offers a huge business value –companies are able to stop relying on analysing the past for better efficiency and use real-time data to predict the future. As more companies learn to leverage collected data, digital twins will help to optimise and enhance their processes and products, while driving the innovation of the business.

For example, General Electric has developed a software platform for the Industrial Internet named Predix, which allows wind farms to boost their energy production by 20 percent. Each wind farm is designed as a computer model based on the location parameters, engineers can configure each turbine individually to maximise the efficiency for the real-world wind farm. When the wind farm ready and it starts to produce power, more data is fed into the digital twin, which in return provides feedback and suggestions to further maximise the efficiency. GE uses digital twins to build the right wind farm at the right place and then using performance data to further optimise the farm’s performance.

All indications seem to predict that we are on the cusp of a digital twin technology explosion. Digital twins will disrupt many industries as they will radically change businesses. As the technology evolves, more and more companies will want to deploy their very own digital twins to gain a competitive edge. It is important to stay ahead of the competition– be the disruptor of the industry rather than being disrupted.

Martin Laid is the Chief Engineer at Neurisium. He is responsible for evaluating factory setups, developing Smart Factory solutions and helping to bridge the knowledge gap between manufacturers and Machine Learning solution creation. He has a keen interest in Machine Learning and a varied background of mechanical and civil engineering with an international experience of having worked on some world leading engineering solutions.

Feel free to also check out our other posts:

Leveraging Past Data on the Production Line

Digitalisation: Making Your Data Tell the Truth

Towards Industry 4.0: Digital Transformation in Manufacturing