Digital twin
A digital twin is a virtual model of a physical object, system, or process that is kept synchronized with its real-world counterpart through data collected from sensors. Unlike a static three-dimensional model or a one-off simulation, a digital twin updates continuously as the physical object changes, so the two stay aligned over the object's lifecycle and the virtual copy can be used to monitor, analyze, and predict the behavior of the real thing.[1][2]
In virtual reality (VR) and augmented reality (AR), the digital twin supplies the data-driven, geometrically accurate model that users view and interact with. A technician can see a digital twin overlaid on a physical machine through an AR headset to read live sensor values and guidance that would otherwise be invisible, and an engineer can step inside a VR copy of a factory or product to test designs and rehearse procedures without touching the real equipment.[3][4] Digital twins are central to industrial uses of immersive technology, including Enterprise AR and Enterprise VR applications for design, simulation, training, and remote monitoring.[3][5]
Definition
A digital twin combines three parts: the physical object, a virtual representation of it, and a data connection that links the two. Sensors built into or placed around the physical object collect data continuously (for example temperature, pressure, position, vibration, wear, and energy use), and the virtual model uses those data to change in step with the real object in real time.[2][1] The synchronization is what separates a digital twin from an ordinary simulation. A simulation typically studies one scenario in isolation, while a digital twin runs against a live feed from its physical counterpart and can therefore reflect the actual current state of that specific object rather than a generic model of its type.[2][6]
Many modern digital twins are bidirectional: they mirror the physical system and can also influence it, for instance by sending control commands back to the equipment. By pairing real-time data with mathematical models and, increasingly, machine learning, a digital twin can simulate performance, test designs, and explore "what-if" scenarios.[6][2]
A common taxonomy, used by cloud and software vendors, distinguishes digital twins by scope:[1]
| Level | Scope |
|---|---|
| Component (or part) twin | A digital representation of a single piece of a larger system. |
| Asset twin | Two or more components modeled together as a working unit. |
| System (or unit) twin | How several assets work together as part of a broader system. |
| Process twin | A whole environment, showing how components, assets, and units operate together. |
Origin
NASA is often cited as the precursor of the idea. In the 1960s the agency built physical replicas of spacecraft so it could study how they might behave under different conditions before and during missions.[6] The modern concept and model of the digital twin was introduced in 2002 by Michael Grieves, then at the University of Michigan, in a presentation on Product Lifecycle Management to the Society of Manufacturing Engineers in Troy, Michigan.[7] Grieves first called the model by other names, including the "Mirrored Spaces Model" and the "Information Mirroring Model"; the term "digital twin" itself was coined by John Vickers of NASA and appeared in a 2010 NASA technology roadmap.[7][2]
The approach was made practical by the spread of cheap sensors, the Internet of Things, faster networks, and more affordable computing and storage, which together allow a virtual model to be fed with continuous data from the field rather than periodic manual measurements.[2][1]
Relationship to VR and AR
Digital twins and immersive technology are complementary. The digital twin provides an accurate, data-rich model of a real object or facility; VR and AR provide the means to inspect and interact with that model in three dimensions rather than through dashboards and two-dimensional drawings.[4][3]
In an AR workflow, the twin is registered to the corresponding physical object so that information appears in the correct place in the user's view. A maintenance worker wearing a head-mounted display such as the Microsoft HoloLens 2 can see holographic step-by-step instructions and live data anchored to the actual machine, and can share that view with a remote expert who guides the work hands-free.[3][4] In a VR workflow, the user enters a fully virtual copy of the asset or environment, which is useful when the real object does not yet exist (a product still being designed) or cannot be entered safely (a running production line). Engineers can walk through a virtual plant, simulate faults, and assess layouts before anything is built.[3][5]
The combination underpins what industry sources call the Industrial Metaverse, in which digital twins, industrial IoT data, and immersive visualization let teams operate inside persistent, data-rich virtual versions of factories, energy grids, and logistics sites.[8]
Platforms
Several vendors offer software for building and visualizing industrial digital twins with immersive output. NVIDIA Omniverse is a platform for creating physically based three-dimensional digital twins and supports VR and AR viewing of those environments for training, maintenance, and remote inspection.[9] In June 2022 Siemens and NVIDIA agreed to connect tools from the Siemens Xcelerator portfolio to Omniverse, beginning with process simulation from Siemens Tecnomatix, so that engineers in different locations could immerse themselves in the digital twin of a plant operating in real time, "down to the physical behavior of the robots."[5] A Siemens executive summarized the aim by saying that the twin "is physics-based; it doesn't just look like the real thing, it behaves like the real thing."[5]
Other companies pair their AR products with twin data. PTC, for example, positions AR (through its Vuforia products) as a "lens" onto the digital twin, presenting a technician with information they could not otherwise see directly on the equipment.[4] Microsoft has promoted the Microsoft HoloLens 2 for guided assembly, holographic work instructions, and remote assistance in manufacturing, often connected to back-end twin and field-service data.[10]
Applications
Reported uses of digital twins, many of which involve VR or AR for visualization and interaction, span several sectors.[1][2]
| Sector | Example uses |
|---|---|
| Manufacturing | Monitoring equipment, validating plant and workstation layouts, predictive maintenance, quality assurance, and operator training. |
| Energy | Optimizing wind farms, solar projects, and offshore installations. |
| Construction and infrastructure | Project planning, progress monitoring, and assessment of buildings and structures. |
| Automotive and aerospace | Vehicle and aircraft models for design optimization and predictive maintenance. |
| Smart cities | Modeling urban infrastructure and predicting the effects of extreme weather and traffic. |
| Healthcare | Virtual models of hospitals, labs, and human organs to plan treatment and test scenarios. |
Training and maintenance are among the largest immersive use cases. Combining a digital twin with mixed reality lets new staff learn complex tasks on a virtual copy of equipment and lets technicians follow visual instructions overlaid on real machines, which reduces reliance on long printed manuals.[3][10] In manufacturing, twins that engineers can examine in immersive form are used to find problems before physical changes are made, which can cut commissioning errors and rework.[9][3]
Academic work has demonstrated specific pairings of digital twins with VR. A 2019 study on industrial workstations described a co-simulation environment that links a digital twin with virtual reality so that workstation designs and worker ergonomics can be assessed before the workstation is built.[11]
Current status
As of 2026 digital twins are an established part of industrial software and of enterprise XR strategy, although adoption varies by sector. Market analysts describe rapid growth driven by the joint build-out of digital-twin platforms, XR hardware, and industrial IoT, with manufacturing, energy, transportation, and healthcare among the leading adopters.[8][1] The U.S. GAO has noted that wider use still faces obstacles, including immature standards, high development costs, and unresolved questions about data ownership and privacy, particularly where twins represent people rather than machines.[2]
See also
References
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 "What is Digital Twin Technology?". https://aws.amazon.com/what-is/digital-twin/.
- ↑ 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 "Science & Tech Spotlight: Digital Twins, Virtual Models of People and Objects". 2023-02-14. https://www.gao.gov/products/gao-23-106453.
- ↑ 3.0 3.1 3.2 3.3 3.4 3.5 3.6 "Digital twinning use cases strengthen with AR, VR". https://www.techtarget.com/searchcio/tip/Digital-twinning-use-cases-strengthen-with-AR-VR.
- ↑ 4.0 4.1 4.2 4.3 "Augmented Reality: A Lens to the Digital Twin". https://www.ptc.com/en/blogs/corporate/combining-augmented-reality-ar-digital-twin.
- ↑ 5.0 5.1 5.2 5.3 "Siemens and Nvidia aim to bring more virtual reality to digital twins". 2022-06-30. https://www.techdesignforums.com/blog/2022/06/30/siemens-nvidia-xcelerator-agreement/.
- ↑ 6.0 6.1 6.2 "Digital twins: Virtual models with real-world impacts". https://www.nsf.gov/science-matters/digital-twins-virtual-models-real-world-impacts.
- ↑ 7.0 7.1 "What is digital-twin technology?". https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-digital-twin-technology.
- ↑ 8.0 8.1 "Industrial Metaverse Market to Reach USD 250.67 Billion by 2031 as Digital Twins, XR Training, and Private 5G Accelerate Adoption". 2025. https://industrytoday.co.uk/it/industrial-metaverse-market-to-reach-usd-25067-billion-by-2031-as-digital-twins-xr-training-and-private-5g-accelerate-adoption-1.
- ↑ 9.0 9.1 "NVIDIA Omniverse, Digital Twin Applications in Industry". https://www.rs-online.com/designspark/nvidia-omniverse-for-digital-twin-applications-in-industry.
- ↑ 10.0 10.1 "Microsoft HoloLens 2 and mixed reality bridge physical reality and digital experiences". 2022-02-23. https://www.microsoft.com/en-us/dynamics-365/blog/business-leader/2022/02/23/microsoft-hololens-2-and-mixed-reality-bridge-physical-reality-and-digital-experiences/.
- ↑ (2019). "Digital twin and virtual reality: a co-simulation environment for design and assessment of industrial workstations".{Template:Journal. https://www.tandfonline.com/doi/full/10.1080/21693277.2019.1660283. Retrieved 2026-06-15.