As a disruptive new force within manufacturing, digital twins act as a means of accounting for the impacts of design choices.
Named one of Gartner’s top ten strategic technology trends in 2017, 2018, and 2019 while receiving further mention in 2020 as a key component of hyperautomation, digital twins are a bridge between designs and products.
This is accomplished through the careful digital modeling of physical objects or systems. The technology is capable of simulating a product’s performance under any amount of independent variables through an intricate combination of machine learning, artificial intelligence, and predictive analytics. Digital twins are redefining how products are made, making it so manufacturing only needs to occur once a design performs precisely to your requirements.
The wealth of operational data collected in order to support digital twins quantifies a number of previously unquantifiable product traits, qualifying designers and manufacturers alike to assess product performance from previously unthought of perspectives. In turn, this allows for the real time improvement of products and services as customers use them.
Digital twins have a number of defining characteristics that set them apart from other virtual models, such as those created with CAD (Computer Aided Design). Connectivity plays an important role, particularly in regard to the IIoT. IIoT, or Industrial Internet of Things, is a term which describes the interconnectivity of the many contemporary material objects and systems now enabled to send and receive data. IIoT makes this connection between the physical and the digital possible; without it, digital twins would not exist. The better a digital twin can replicate its physical counterpart, the more useful it will be. In creating an accurate digital representation of the real world, it is necessary to use smart technology to best enable IIoT sensors. Every digital twin creates a digital trace of a physical object using these sensors.
These traces guarantee that digital twins are exact replicas, containing all of the same information as their physical counterparts. They can also prove useful during any troubleshooting of the data delivery systems essential to supporting this technology.
Digital twins help to standardize the manufacturing process through modularity. Modularity refers to the customization of production components, allowing you to optimize every machine on the production line. As previously mentioned, the capacity for digital twin technology to simulate a product’s function assists manufacturers in identifying poorly performing components and altering their designs without ever needing to order the development of a prototype. With regard to systems, digital twins can even simulate the workflow of an assembly line, giving some idea of the time and cost associated with upscaling or otherwise making changes to manufacturing.
What is a digital twin and how are they important to manufacturing?
Digital twins are virtual replicas of physical objects or systems, made possible through the generation and collection of digital data that represents such objects or systems.
At a fundamental level, the technology is the result of gathering data and modeling it virtually to visualize exact replicas of designs. Therefore, they contribute to the value already offered by nuanced data collection and help to prepare manufacturers for an increasingly automated future. They also allow for the development cycle to be undertaken entirely digitally, removing a significant amount of any potential cost barrier for manufacturing new products.
Digital twins can be further defined by the stage at which they are used in the manufacturing process.
The benefits of digital twins vary depending where and how they are used. A Digital Twin Prototype (DTP) is any digital twin made as part of the initial design phase before any physical product is created. DTPs are particularly invaluable at this phase because of their ability to test designs digitally before any costly physical production occurs. After production begins, a Digital Twin Instance (DTI) is made to run tests on different usage scenarios to identify any potential issues with the existing product. Digital Twin Aggregates (DTA) gather data from the DTI to test operating parameters and identify opportunities for further product development. Regardless of where you choose to apply it in your manufacturing process, the primary purpose of digital twin technology is to save money.
Efficiently implementing digital twins into a workflow may seem daunting at first, but the various fine-tuned instruments, tools, and devices employed in many manufacturing processes prove to be ideal for the creation of digital twins. With their ability to obtain precise measurements, it becomes possible for the building of a detailed digital twin to be accomplished relatively quickly and easily.
To start, a variety of data about an object’s status and function is gathered by integrating smart sensors into that object. These sensors are all connected to a cloud computing system, which receives and processes all data from all of these sensors. Next, newly processed data is compared to other data from your business, as well as any other contextually relevant data that may be required for analysis. The results of this analysis are then applied to real world objects and systems. The opportunities identified by digital twins make them an invaluable contributor to the never-ending refinement and optimization that occurs with any manufacturing process.
However, keep in mind that there does not yet exist a standardized platform for building digital twins. Current software is not only industry specific, but often organization specific as well.
Digital twins can be as simple or as complex as the objects they simulate. The precision of any given simulation is dependent upon the quantity of data available for both the initial building phase and any future updates. Any attributes which could be measured or observed from the real asset should also be accessible from its digital twin. Twins are constructed to receive data from sensors on their real world counterparts. Computer programs utilize real time, real world data to best visualize predictions about how potential design changes might affect the trajectory of various product attributes.
For example, a digital twin could test how new exterior surface materials might perform relative to fluctuating degrees of wear and tear over the lifespan of your product. Furthermore, they are also indicative of current performance and can be used to identify any potential issues with existing products. The unique real time perspective they provide can reduce maintenance issues, anticipate downtime, and lead to overall improved customer support service. The insightful possibilities offered by these predictive models are applicable across all stages of a product’s lifecycle.
An appeal of digital twin technologies is their ability to improve decision making.
From a customer perspective, they assist in streamlining the communication of customer needs while enhancing already available products and services. On the business end, additional data analytics software and personnel can be essential tools to best understand simulation results and further optimize efficiency.
Predictive analytics (those which use historical data to make projections about the future), for example, can be used to refine your assumptions and drive forward the decision making process.
“For every physical asset in the world, we have a virtual copy running in the cloud that gets richer with every second of operational data,” says Ganesh Bell, chief digital officer and general manager of Software & Analytics at GE Power & Water.
GE uses careful cultivated digital representations of their contract sites to inform system configurations, such as which direction a wind turbine might face, before any construction occurs. The data associated with digital twins, therefore, remains of an exponentially increasing value to more and more companies.
Digital twins can also be used to enrich organizations through servitization, the process of adding value through the offering of services. For example, if you are a manufacturer of airplane engines, digital twins could enable your company to provide the service of ongoing maintenance (if they do not already) or improve engine efficiency through incremental upgrades over the course of its lifecycle. Digital twins have already made themselves a mainstay in factories and production plants around the world. By simulating their manufacturing processes, manufacturers optimize assembly to minimize downtime while maintaining a cost effective operation.
Examples of Digital Twins
Digital twins have a far reach of applications for a wide array of industries. An early form of digital twins were first utilized by NASA over the course of the Apollo program, giving the association a significant advantage over their competitors in aerospace. Creating exact replicas of spacecraft systems turned out to be the only feasible way to problem solve in space from the ground in a pre digital era. Artificial intelligence and machine learning are used to analyze the model of operations represented by the digital twin regardless of where the real equivalent is actually located. “The ultimate vision for the digital twin is to create, test and build our equipment in a virtual environment,” says John Vickers, a leading manufacturing expert at NASA and manager of the National Center for Advanced Manufacturing. As this digital convergence grows closer, it is all the more fundamental for manufacturers to recognize the value offered by digital twins. The following are a few selected use cases, sourced from early adopters in established industries.
Automotive manufacturers have already embraced digital twin technology in preparation for the ever approaching launch of fully automated vehicles. Digital twins of automated vehicles significantly cut down on development time through their ability to undergo the large number of tests mandated by vehicle safety regulators entirely digitally. This cuts down on the cost of testing by not having to set aside units for or carry out any road tests. Another application within automotive manufacturing would be digital twins’ ability to forecast component durability.
Hypothetically, let’s say you’re a business data analyst working for Toyota as they prepare to roll out a new line of vehicles. Simulations have shown that the drivetrain in this year’s new line of sedans will wear out after some 125,000 miles. Thus, the testing enabled by digital twins has demonstrated it might be necessary to include a 125,000 mile warranty with every new car sold. This is but one example of how digital twin technologies can add cost effective functionality to the manufacturing process.
The powerful platforms that enable digital twin technology are able to account for an astronomical amount of variables, meaning that this technology has applications even in urban planning. A digital twin is used by the city of Singapore as part of their planning process. Aptly called “Virtual Singapore”, this 73 million dollar investment assists the government and local businesses in building a city best prepared to meet whatever challenges the future may hold. Water, food, and people are just a couple of the resources this digital twin takes into account in the context of managing a city. City planners are able to utilize this data to improve the lives of their citizens and ensure that things are operating efficiently.
Healthcare is one field where the potential of digital twin technologies are just beginning to emerge. Patient care has been revolutionized by bandages equipped with IoT sensors. These sensors monitor patient vitals, producing digital models that can be used to improve care. Digital twins are helping to personalize care, enabling an unprecedented level of insight into how genetic level differences between individuals impact their response to any received treatment. Additionally, digital twins of organs and bones allow healthcare professionals to practice complex surgical procedures without the added pressure of a real patient.
From manufacturing to healthcare, the potential for digital twin technology to innovate and improve our world is limitless. They will also assist in eliminating some of the gatekeeping elements and costs associated with the manufacture of a product.
As more and more companies develop digital twins to maintain a competitive advantage, inevitable improvements will occur to ease the process of development and deployment for both established organizations and burgeoning start ups.
Digital Twins are the Future
Within the next few years, digital twins will come to directly represent billions of everyday things from all around our world, including products, cities, and even human beings.
A likely effect of this will be a growing convergence between the work of manufacturers and data scientists. Developing these kinds of collaborative relationships will become paramount as such an inconceivable amount of data continues to be collected; the more data that you have, the more valuable that someone who can interpret it within the context of your organization’s goals becomes.
As with any other new technology, it is important to consider how the use of digital twins will fundamentally change the way your organization does things. There is a slight risk associated with being a relatively early adopter of digital twin technology that should be considered. Implementation cost, compatibility with existing systems, and data security are a few key factors in play. Despite their innumerable applications, the implementation of these systems could cause an unnecessary rise in complexity relative to the problems you intend them to solve. Keep this in mind as you consider deploying digital twins in order to either gain or maintain a competitive advantage.