Digital Twin: Vision and Boundaries

[Digital Development for Human Beings and Things - NTT R&D]

A digital twin revolution is coming. A digital twin is a digital representation of a physical object, process or service. Digital twins can be digital replicas of objects in the physical world, such as jet engines or wind farms, or even larger objects such as buildings or even entire cities. 

The concept of digital twins is not new. In 2002, at the University of Michigan, computer engineer Michael Grieves mentioned the possibility of creating digital representations of physical systems with their own entities in a conversation with NASA technical director John Vickers. Less than two decades later, in 2018, consulting firm Gartner named digital twins one of the top 10 technology trends of the year.

In a speech to industry in the early 2000s, Michael Grieves proposed that the construction of digital information about a physical system could be created individually as an entity. The virtual system contains all the information about the physical system and is associated with the physical system throughout the life cycle of the system. 

NASA introduced the concept of digital twins to this idea, which are ultra-high-fidelity simulations of space vehicles that allow engineers on Earth to mirror the exact and actual conditions of real vehicles during missions.

- Principles

The principles behind the digital twin vision originate from the field of product lifecycle management. From this perspective, there is a strong need to integrate all product-related information into a comprehensive product management system that can be accessed by any user at any stage of the product life cycle, such as full life cycle performance information, design optimization and Manufacturing system improvements. It is recommended to use the digital counterpart of each physical product as a central means of managing product data during the product lifecycle.  

A digital twin is a concept, not a single product or technology. Multiple technologies – 3D simulation, IoT, 4G/5G, big data, blockchain, edge computing, cloud computing and artificial intelligence – work together to make this concept a reality. The core principle is, for a physical entity or asset, the digital equivalent in the virtual world.

Essentially, a digital twin is the virtual digital equivalent of a physical object. They connect remotely to real objects in real-time and provide rich representations of those objects and their context. This digital twin goes beyond static product designs, such as CAD models, to include dynamic behavior. 

This dynamic nature of the digital twin may include a representation of the current behavior of real-life objects, as well as simulations or predictions of future behavior and recall of historical behavior.

Looking back over the past 30 years or so, we can see that every 10 years or so, the focus of digital has alternated between people and things.

• 1985: Focus on the digitization of "Humans" (Humans - first generation). With the introduction and popularity of email, the digitization of "human" communications.
• 1995: Focus on the digitization of "things" (things - first generation). Digitizing "things" such as timetables and maps through the spread of the Internet.
• 2005: Human-Centric Digitization (Humans - Second Generation). With the introduction and popularity of social media, the digitization of "human" connections and networks.
• 2015: Focus on digitization of "things" (things - second generation). With the introduction of the Internet of Things and artificial intelligence, the digitization of various "things" such as parking lots.

So, what will 2025 bring? Following the same pattern, next we should see progress in the digitization of "humans".  

- Digital Twin Adoption Grows

A digital twin is often defined as a software representation of a physical asset, system or process designed to provide business value through real-time analytics to detect, prevent, predict and optimize. 

The concept of digital twins has been around since the early 2000s. The company will program the replica on their computer or server and enter the necessary data to update the design or environmental specifications. This information is rarely received in real time. This method is a very clumsy and slow way of rendering these virtual copies. 

With the advent of IoT technologies such as smart sensors and cloud systems, digital twins have become easier and cheaper to create, maintain and use. 

We have now reached the stage of intelligent enterprise asset management supported by an influx of connected information from digital manufacturing equipment, hyperautomation, and other sources. 

Digital twins are becoming more common in and out of factories, as these exact virtual replicas of systems allow for larger experiments with various scenarios to determine outcomes. It's a surefire way to test new ideas and different conditions. Positive results can then be communicated to the physical twin for implementation. 

The idea of ​​a digital twin as a real-time digital representation can be applied to more complex physical structures, often combining connected assets and products with specific business outcomes. Visualization allows companies to test for potential machine failures, optimize processes, plan for future capacity, and more.

- Popular Use Cases for Digital Twin

Today, the application of digital twins is widely used in many industries. Applications of digital twins provide accurate virtual representations of objects and simulations of operational processes, enabling businesses of all kinds to rapidly identify areas of innovation and improve business processes and performance. 

Popular use cases for digital twins include: manufacturing, healthcare, utilities, disaster management, insurance, smart cities, and more. 

Examples of digital twins include product digital twins, component digital twins, and performance digital twins, all of which are used to visualize and process objects, whether single or more complex. 

Industry 4.0 is sometimes referred to as the "Fourth Industrial Revolution". The current trend is that manufacturing is increasingly automated and relies on capturing and exchanging data through the Internet of Things (IoT) and cloud computing, often referred to as "big data". Industry 4.0 creates so-called "smart factories". A digital twin as a computer representation of the product fits into this model.