Image: Marisa Garcia

APEX Insight: The principle of using “digital twins” for modeling during program research and development is hardly new, but advances in computing mean we can use them to develop better aircraft and deliver better passenger experience at lower cost.

At the launch of ZAL TechCenter’s Innovation Days in Hamburg, Dr. Susan Ying, chief operations officer at international drones manufacturer XenorD International, discussed the importance of advancing digital product modeling as a strategy to reduce the costs and lead-times of new aeronautics products.

Ying is a fellow of both the American Institute of Aeronautics and Astronautics and the Royal Aeronautical Society, as well as president of the International Council of Aeronautical Science, and executive member of the Aerospace Council of the SAE International. She has held roles as chief integration officer at COMAC and director of Research and Technology at Boeing, and she has also worked in the DOE Research Labs and NASA Ames Research Center.

“The ‘digital twin’ [concept] actually came from NASA.” – Susan Ying, XenorD

“The ‘digital twin’ [concept] actually came from NASA,” she explained. “With the Apollo program, NASA had one chance to succeed, because you sent the rocket and if you lose it you fail, but you can’t do the actual experiment [in advance] so you do a lot of modeling and a lot of simulation.”

Tools like artificial intelligence, machine learning, and the internet of things, used to support ‘digital twin’ development processes can advance certification and testing of new products, and save significantly on the costs of real-world mock-ups.

“Today we not only have the capability of modeling and simulation but, we have also opened up the whole world of internet of things, as well as data analytics,” Ying said. “Because of all of the computing power available, data analytics is very powerful now. Now not only can we have a better understanding of our designs, and optimize our designs, but we can predict what is going to happen a lot better. We can have predictive maintenance and predictive manufacturing.”

Ying suggested that digital twins could also follow the product through into service: “Transformational tools and processes [of this kind] allow one to merge the power of a computer with the expertise of the engineer, so you can predict the performance before the build,” she says. “[It allows you to] model so that you fully understand the life cycle.”

Using this approach, designers and engineers can work together with marketing, sales and customer support teams, exchanging product performance data, which might prompt a redesign. But the systems themselves can also self-report on connected aircraft gathering data needed for design improvement or predictive maintenance.

“Doing everything virtually is a first path to success,” Ying said. “We are starting to model from the beginning to the end of the lifecycle. This allows us to go from concept to pilot architecture, as well as through to the 3D physical models, as well as through the many product and service model processes; so that you really understand the whole life cycle before you even do the manufacturing.”