Digital Twin

What Is a Digital Twin?

A digital twin is a digital representation of a device, process, or environment that exists in the real world. With this technology, companies create virtual copies of assets, run accurate simulations, and apply tested success in the real world.

Colin J Parris Ph.D., GE Digital's Chief Technology Officer, describes a digital twin as a “living model which drives a business outcome.”

Although a simple explanation of complex technology, this explanation works beautifully. The digital twin is a living model that behaves like the real world because it is modeled after real-world actions and components. Businesses use digital twin technology to examine their business and drive their desired outcomes.

A Brief History of Digital Twin Technology

Digital twin technology may seem like a new capability of Industry 4.0, it actually dates back as far as the 1970s when NASA pioneered the “twin” concept. Although, the technology wasn’t digital in its first implementation.

In 1970, the famous Apollo 13 mission lost its oxygen tanks in transit to the moon. A fight for survival ensued in both the spacecraft and on the ground. With limited time, NASA engineers on the ground needed to find a way to troubleshoot and repair the spaceship remotely.

A team of engineers on the ground used a physical twin of the spacecraft to re-create the problems experienced by the Apollo 13 crew. With this information, NASA was able to find the appropriate solutions and rescue the astronauts on board.

How IoT Makes Digital Twin Technology Possible

Modern digital twin technology is, for the most part, dependent on IoT sensors and devices that gather data from the real world and apply it to the digital world. In order to achieve this level of modeling, there are a few essential factors that need to be present.

Constant and Consistent Analytics

First, digital twin technology needs to have strong and consistent analytics processes. Data needs to be captured in real time so that it can be analyzed at every step. Only then, can your digital twin be responsive and predictive to real-world problems and solutions.

No matter the size of the device, process, or environment, the amount of data required to create a complete digital copy is enormous. The data needs to be organized and structured in a way that allows people and systems to pull valuable information at any point in the design, build, and operation.

Layered & Open Data

With the digital twin gathering data from numerous sources and IoT devices, the data needs to collaborate harmoniously. Or in other words, the information needs to form a federated model that consists of layers and layers of data, emphasizing the importance of our first essential factor above.

A layered model also enables multiple machines and devices to connect and learn from each other’s environments and digital twins, creating an open environment for quick data flow and learning.

Contextual Application

While capturing the engineering aspects of a machine, process, or environment, digital twin technology also enables companies to apply different models based on the context of the industry.

For example, turbines used for air or water will behave differently, output distinct data, and have contrasting goals. The same can be said for cutting tools that cut different metals and pumps which enhance the flow of various liquids.

Based on the purpose of the industry, the digital twin needs to reflect the real world’s unique environment and outcomes. This level of differentiation gives companies incredible insight and control throughout a process or product’s complete lifecycle.

The 3 Phases of Digital Twin Technology

Digital twin technology has 3 phases. Each phase uses a complex modeled digital twin to achieve the highest levels of configuration, manufacturing, and functionality.

Phase 1: Design - Plan for Success

During the design phase, you have an open canvas to create and test your ideas. There is no need to go through costly prototypes.

Digital Twin technology enables physical elements such as BOM (Bill of Materials) and physical assets to meet digital assets like software and company knowledge. By bringing the physical and digital components together early on in the lifecycle of a product, manufacturers can plan for success and efficiently understand the outcomes of their designs.

With a digital twin, manufacturers can shift from a prototype-oriented design to an operationally oriented one. This orientation ensures that products are built with clear goals and reliable objectives that have already been digitally achieved.

Phase 2: Build - Better Manufacturing

The build phase is all about better manufacturing. So far within the design phase, we’ve been looking at the digital twin of only one product. But in the build phase, ideally, the machines and processes used to manufacture the product also tap into a digital twin. By using IoT devices and tools within manufacturing, companies gain an enhanced understanding of how every action affects the build phase.

Digital twin technology gives you various insights during the build phase:

1. How your manufacturing processes affect the product’s outcomes.
2. The specific stresses and tolerances of the design and manufacturing.
3. What actions and measures are needed to drive the correct outcomes.

Phase 3: Operation - Real-Time Information

The operation phase is where the product is being used in the real world while still adhering to optimal performance goals through real-time monitoring. With IoT sensors, the digital twin extrapolates and measures factors that will change the product's functions and outcomes.

For instance, let’s say the product in operation is a turbine engine for a commercial jet. Depending on humidity levels, temperature, and wind speeds, the turbine engine will need to function differently to achieve optimal performance.

Digital twin technology creates an environment of constant information feedback that is shared between the physical and digital worlds. And people and systems can take advantage of this technology to peer into information and knowledge that they would have never had before.

What sets the operation phase apart is time. This phase is generally the longest part of the product's lifecycle, meaning that there are incredible opportunities to learn, grow, and continuously improve. The captured in-process data is then used to improve the operation phase as well as improve the design and build phases of subsequent projects.