If you haven’t heard the term “digital twin,” you’ve been living under a digital rock. The technology has also been referred to as virtual prototyping, hybrid twin technology, virtual twin, and digital asset management. With an estimated 25 billion global sensors being brought online by 2021, MIT Sloan Management Review calls digital twin technology "an innovation game changer." 

So, what exactly is a digital twin, in the context of manufacturing automation?

A digital twin is…

… a digital replica that uses real-time data, collected from connected sensors, to mirror a unique physical device (referred to as a digital twin Prototype), or a process (referred to as a digital twin Aggregate).

digital twins enable devices, processes, systems to be understood, analyzed, manipulated, or optimized across their lifecycles. Engineers can build simulations of how the asset will be affected by different inputs, using an exact replica of the asset. The aggregation of multiple digital twins provides a composite view across an enterprise, driving faster, more accurate decisions. A digital twin Aggregate is dynamic and uses artificial intelligence to continuously correct its process simulation based on real-time feedback from remote sensors.

Why digital twins Matter

digital twins are gaining popularity because they decrease the complexity of monitoring and optimizing IoT environments. They provide safe, efficient, inexpensive ways to monitor real-time data, assess health and performance of an asset, and predict different outcomes based on variable data inputs. 

The primary benefits of digital twins include reduced costs, future cost avoidance, increased productivity, and improved overall asset value.

End users are realizing these benefits by using digital twin technology in the following ways:

• Visualizing assets while they’re in use, by real users, in real time
• Monitoring and troubleshooting remote devices, processes, and systems
• Leveraging predictive analytics and variable data to run scenario testing on devices and systems
• Streamlining the deployment of assets in an IoT environment, helping design engineers make functionality and operations decisions
• Using prototype twins to qualify and certify equipment remotely, test integrations, verify tolerances, and conduct inspections—all from a desktop anywhere in the world

Under the Hood of digital twin Technology

Data scientists identify the physics and functionality of a physical object or system and use that data to develop a mathematical model that simulates the asset in a digital form. The model is connected to hundreds of sensors gathering data from its real-world counterpart, offering powerful insights into real-time performance and enabling prototyping and testing by using variable data inputs. The digital twin is continuously updated to mirror the current state of the asset, and when combined with business rules, optimization algorithms, or prescriptive analytics, it can support human or automated decision making.

digital twins act as a proxy for their real-world counterparts, and any application that needs data from the asset can leverage the digital twin instead. By encapsulating the data, changes can be made to the twin without affecting any other applications.

Many different manufacturing sectors are leveraging this technology in common ways:

• Continuous monitoring of product data from IoT devices—rather than random inspection, digital twins monitor and model every part of the production process to identify quality issues and analyze the composition of the product being created
• Virtual start up—historical data from similar systems can be mapped into a digital twin prototype of a new system before its constructed; this allows you to model various scenarios and environmental constraints to determine the best way to connect and deploy assets
• Product development and customization—manufacturers can leverage data from production and enterprise layers to include real-world feedback into digital twin models; users can also simulate changes to a product’s design and the resulting impacts on the production process without expensive re-tooling; this enables more affordable and efficient creation of smaller quantities of variations of products
• Supply chain optimization—digital twin replicas of manufacturing processes provide an opportunity to understand how materials are being used, enabling efficient and perhaps even automated inventory management

What Does the Future Hold for digital twin Technology?

Information Age cites research suggesting that digital twin isn’t going anywhere anytime soon:

• 62 percent of organizations that used IoT were also using or planning to use digital twins
• By 2022, more than two-thirds of companies with IoT projects will be incorporating at least one digital twin into production processes

digital twins are a powerful enabler for the promise of IoT—they can help manufacturers and facilities detect problems, improve productivity, and optimize processes. They can even help simulate and predict performance using variable data. This, in turn, drives scenario-based innovation.

Most of the time, digital twins will be built by OEMs and delivered alongside their physical components, systems, and assets. Starting now, if they haven’t already, end users should consider the availability and robustness of digital twins within their purchase decision criteria. Asset owners will also need to create comprehensive management strategies that address dozens of key questions, including:

• What are the organization’s goals in deploying digital twin technology? How will you catalog, organize, and optimize the use of digital twins in various parts of your workflow?
• Data formatting and data quality—how will the digital design data feed into an integrated digital twin? Is that data formatted correctly and secured?
• If data is contaminated or needs to be reformatted, whom in the organization takes the lead on that project and how will costs be covered? What kinds of approval processes need to be in place with contractors and internal teams?