Digital twins are realistic digital representations of something in the physical world. They are rapidly being developed and shaking up industries, including healthcare, manufacturing, and city design. Digital twin technology intersects with advances in artificial intelligence, next-generation mobile communications, big data, and more.

Here, we explore this emerging technology in detail. We explain what it is, how it works, its advantages, and the challenges for implementing it, with examples throughout.

What is digital twin technology helpful for?

The aim of a digital twin is to obtain key information that can be used to improve the physical asset. They enable data-driven decision making, the monitoring of complex systems, product simulation, and management of an object’s entire life cycle.

These deep insights into systems can save time by enabling rapid iterations and the optimisation of product design by removing the need to build and test individual prototypes.

Moreover, by reducing the requirement for physical materials and resources during product design, DTs can increase sustainability. Also, at the end of a product’s life, DTs can aid in determining which materials need to be recycled and how to dispose of them effectively.

So, what is digital twin?

Defining digital twin

In short, a digital twin (DT) is a virtual replica of a physical object, person, or process that can be used to increase understanding and perform tests. They are linked to real data sources in the environment and update in real time to reflect changes in the physical world.

Here are two examples that can help clarify what this means.

First, a wind turbine could be fitted with various sensors related to vital functionalities. These sensors would collect data about different aspects of the turbine’s performance, like its energy output and temperature and external factors like the weather. The processing system would receive this information and apply it to a digital copy. This would aid in monitoring the turbine in its environment and finding areas for improvement.

Second, a common example is Google Maps. Essentially, Google Maps is a DT of Earth’s surface. It is linked to real-time data on things like traffic and roadwork to help optimise your commute.

Digital twins differ from simulations in a few ways. A DT is actually a virtual environment, not just the represented object (like how factors like weather are considered in the wind turbine DT), whilst simulations tend to study one particular process. Also, simulations do not tend to have real-time data, which is essential for DTs.

Brief history of digital twin

The ideas behind digital twin technology can be traced back to the 1990s, in work by David Gelernter. Dr Michael Grieves was the first to apply it to manufacturing in 2002, although the term was only coined in 2010 by NASA’s John Vickers.

NASA used physical twins as far back the 1960s, during the Space Race. During this time, each voyaging spacecraft was exactly replicated in an earthbound version that was used for study and simulation purposes.

Creating a twin helps further the testing process and aids in the monitoring of an asset. But let’s delve even deeper into the specifics of DT.

Key characteristics of digital twins

Here, we discuss the three key characteristics that constitute digital twins.

Representation of a physical system

We’ve already touched on the first characteristic—representation of a physical system. The digital twin is an exact representation of the physical version. However, DTs can and often do come first since they can be used to test different components and materials without having to procure or use them physically.

As for what DTs represent, this ranges from small parts of an object to entire systems:

• Component/part twin: this is the basic unit of a DT, representing a functioning component.
• Asset twin: when two or more components work together, they form an asset.
• System or unit twin: different assets come together to form an entire functioning system.
• Process twin: different systems come together to form larger processes, like production facilities.

Connection along the entire lifecycle

The second characteristic—connection along the entire life cycle—refers to two key aspects of digital twin. First, that they can be connected to represent entire systems or processes:

• Product twin: representation of a product.
• Production plant twin: representation of an entire manufacturing facility.
• Network twin: representation of a procurement or supply chain.
• Infrastructure twin: representation of physical infrastructure like highways, buildings, etc.

A DT of a store could be created, then connected to a series of DTs for warehouses, supply chains, and call centres, until an entire organisation is being represented.

Second, DTs can be created for a product along its entire life cycle. A wind turbine DT could be created during the initial planning stage before any physical parts are created. Different components and materials could be tested and then the turbine could be produced. Next, the turbine could be monitored until it finally is retired. Finally, the DT could be used to aid in the disposal process, ensuring parts are recycled and disposed of sustainably.

Bidirectional data exchange

Perhaps the key advantage of digital twins is the exchange of data back and forth between the physical and digital versions. This differentiates a DT from simply a digital model:

• Digital models do not integrate an automatic information flow from the physical world to the virtual version. Changes must be made manually.
• Digital shadows feature an automatic information flow from the physical to the virtual world.
• DTs feature an automatic information flow back and forth between the physical and virtual worlds.

Data, typically collected by sensors, are essential to DTs. They ensure the DT can update in real-time and entirely reflect the physical version.

Challenges for adopting digital twin technology

There are currently various factors holding back digital twin technology from being embraced on a broad scale:

• High costs of implementation due to the increased number of sensors and computational resources needed.
• A lack of standards, frameworks, and regulations for implementation.
• Issues related to data, such as trust, privacy, and cybersecurity.
• Communication network-related obstacles.

These issues reflect how DT is an emerging technology and is very early in its development and implementation. Likely, its potential value will encourage researchers and industries to address these issues.

Examples of digital twin technology

Generally, digital twins are useful for physically large structures or products bound by strict engineering rules (bridges and cars), mechanically complex projects (engines and generators), and for streamlining manufacturing projects.

Here, we’ll outline a smaller, consumer-based example of DT technology and then a broader, more functional one.

Baseball stadium twin

In 2022, the Atlanta Braves, a US baseball team, introduced a DT of the club’s Truist Park. This is a photo-realistic version of the baseball stadium. Fans can attend games virtually in the metaverse, which is a shared 3D virtual space.

Features include creating avatars, exploring the park, enjoying exclusive content, and more.

This DT helps improve the fan experience, providing opportunities for people who are not local and to create unique engagements with the team.

Twin of an entire city

A smart city is an approach to managing cities by integrating data and digital technologies to achieve various aims. Smart cities are enabled by the constant stream of data being generated from traffic, public transport, power generation, water supply, and waste management, among many other things.

DTs of apartment blocks, neighbourhoods, infrastructure, and utilities are connected together to create a digital version of the city. This can help in diagnosing problems and finding solutions, linking to various aims such as improving quality of life, economic efficiency, sustainability, and public transport.

For example, the Shanghai Urban Operations and Management Center built a DT of the entirety of Shanghai, a city with 26 million inhabitants. The twin includes 100,000 elements, ranging from refuse disposal and collection facilities to e-bike charging infrastructure. Data are also collected from satellites and drones.

City DTs are vital for crisis preparation and management, as floods or virus outbreaks can be simulated to test the city’s preparedness and identify potential weaknesses in the infrastructure.

Digital twins in manufacturing

Manufacturers are always looking for ways to innovate, especially in response to new demands. Implementing digital twins in manufacturing is one way that manufacturers are addressing these growing demands for sustainability and complexity.

In our article, Implementing Digital Twins in Manufacturing, we outline leading research on integrating digital twins in manufacturing, showing why it’s useful and how it intersects with other technologies.

Digital twins in healthcare

Digital twin technology has the potential to revolutionise personalised medicine and improve the efficiency of public health organisations. However, implementing DTs into healthcare has raised several ethical concerns and technical challenges.

In our article Applying Digital Twins in Healthcare, we cover how DTs are being applied in personalised medicine and to tackle public health challenges. Then, we explore the concerns and barriers being raised about widespread implementation.

Research on emerging technology

Digital twin technology reflects a convergence of various advanced technologies. It is already being embraced across various industries and will likely continue to spread as it becomes more cost effective and better understood.

It is an exciting field in which research is constantly being produced. Hence, we are likely to see significant advances in DT technology through the decade.

At MDPI, we are interested in exploring DT’s capabilities and showcasing exciting uses of it. MDPI makes all its research immediately available worldwide, giving readers free and unlimited access to the full text of all published articles. Click here to visit our full list of journals.