We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!
Digital twins have the potential to transform the way products are designed, built and operated to improve sustainability and profitability. But most digital twin projects to date have focused on a specific use case. Emerging digital twins standards promise to help connect the dots between individual digital twins to enable systems of systems.
Multiple standards are being developed by the International Standards Organization (ISO), the Industrial Digital Twins Association (IDTA) and the Digital Twins Consortium (DTC) — creating some challenges for unifying digital twins into systems. At the Digital Twin Summit, experts from each organization weighed in on where these standards are today and what is ahead.
Irene Petrick, senior director of industrial innovation at Intel, said people tend to think about digital twins as monolithic constructs.
“Instead, we really need to be thinking about digital twins as being anchored in a point of view,” Petrick said.
System of digital twin systems
What a digital twin needs and how it adds value are perceived differently when considered at the level of a machine, a factory production cell, engineering and design or a leadership level, Petrick explained. A broader adoption of standards could support the interoperability required to enable this system-of-systems approach.
Sameer Kher, senior director of product development at Ansys, who also sits on the steering committee for the DTC, pointed out that a manufacturing facility generally consists of various kinds of equipment. Each of these elements is, in turn, composed of engineering systems such as robot arms, motor drives and software. Standards are essential for all these things to work together in the real-world and virtual world.
The DTC approaches this problem by developing a collection of open-source software to implement best practices and support open-source collaboration.
Enabling different types of questions
As one example, Boeing has been a big proponent of digital twin standards and has been working with various groups to evolve standards from the physical world to the digital.
Kenneth Swope, senior manager of enterprise interoperability standards at Boeing, said that, “Historically, these [standards] all used to be with respect to things like processes and properties and characteristics of physical things. And they continue to be so, but in addition, they are now about the data as well.”
His team is looking at improving the utilization and exchange of data across groups involved in engineering, manufacturing, supply chain, service and support. Teams have been exchanging data for a while.
“But the really neat thing is about digital twins is the opportunity to ask different types of questions, to really solve age-old business problems of how can I make my product safer, how can I make my product with higher quality and how can I make my product more efficient,” Swope said.
Boeing worked on the ISO-23247 standard to show how digital twins could help optimize the design and assembly of fasteners in a jet wing. Using digital twins in concert with instrumentation in the manufacturing process helped identify opportunities to reduce the size and number of fasteners, shaving hundreds of pounds off the weight of a wing compared to a traditional approach.
Digital threads promise to connect the digital twin dots
Gordon (Guodong) Shao is a computer scientist in the lifecycle engineering group at the National Institute for Standards and Technology (NIST). He has been helping coordinate efforts on the ISO-23247 standard and developing best practices as part of a NIST manufacturing testbed. He also wrote up a deep dive on various use cases of the technology.
Shao observed that digital twins require the interplay of components for data collection, processing, communication, modeling, analytics, simulation and control. Some of these can be distributed, so there is a significant challenge to seamlessly integrate these various pieces because of that.
“We need to take a system-of-systems approach to characterize and manage these subsystems to ensure cross-disciplinary interoperability and maintain the credibility of digital twins,” Shao said.
ISO-23247 provides a generic digital twin development framework to help manufacturers choose building blocks for digital twin implementations. It can help them analyze digital twin project requirements and use common terminology when communicating with suppliers, partners and customers.
A vital feature of this new standard is support for digital threads that connect the dots between data from different parts of a product’s life cycle. Digital threads will help optimize processes at the enterprise level that are greater than the local optimization available with siloed digital twins.
A lifecycle approach
Digital twins today are mostly application-driven.
“But what we really need is the interoperable digital twin so we can realize the interoperability between these different digital twins,” said Christian Mosch, general manager at IDTA.
The IDTA Asset Administration Shell standard provides a framework for sharing data across the different lifecycle phases such as planning, development, construction, commissioning, operation and recycling at the end of life.
It provides a way of thinking about assets such as a robot arm and the administration of the different data and documents that describe it across various lifecycle phases. The shell provides a container for consistently storing different types of information and documentation. For example, the robot arm might include engineering data such as 3D geometry drawings, design properties and simulation results. It may also include documentation such as declarations of conformity and proof certifications.
The Asset Administration Shell also brings data from operations technology used to manage equipment on the shop floor into the IT realm to represent data across the lifecycle. For example, the robot arm generates a stream of operations data once it hits the shop floor that is gathered using standards like OPC UA. Teams may also create processes that use this robot using specifications written in automation ML.
Digital twin standards are still young. The ISO-23247 standard was only finalized last October. But panelists expect widespread adoption could play a crucial role in digital transformation for physical industries.
“If you think about standards, both from the systems of systems approach and the lifecycle approach and use the digital thread, you can save a lot of effort on the information exchange, reuse the information and avoid customized effort,” Shao said.