Digital twinning is one of aviation’s success stories, with the industry reaping the benefits of having been at the forefront of this technology. But what is it at its technological core, and what are the new elements emerging as digital twinning evolves? We sat down with Julie Shainock, Microsoft’s global leader for the travel and transportation industry, to understand more about the growing benefits of digital twinning — and where the opportunities are to be found over the coming years — for the latest in our ab initio primer guides.
“The concept of digital twins has been around for quite some time when you think about the significant use of advanced modelling and simulation for design, testing, and training,” Shainock says. “Overall, the industry sees great benefit in this technology and also greater future potential for operationalising it.”
Digital twinning today spans a variety of segments within aviation, including:
- predictive maintenance
- performance optimisation
- training and simulation
- design and development
- supply chain management
- asset lifecycle management
Some of the earliest work on digital twinning was led by enginemakers, and indeed early limitations to digital twinning technology and methodologies meant that the discrete environment of an engine — complex enough to realise substantial benefits, but separate enough to exercise control, and manufactured apart from the rest of the aircraft — was in many ways optimal for the development of digital twinning.
Today, commercial aircraft and their subsystems may well have multiple digital twins at different levels: two, three or four for the engines, one for the inflight connectivity network, another for the entertainment system, one for the airframe, one for the avionics, and so on.
On the carrier side, Shainock explains, “airlines have built some of their own proprietary systems and will continue to do so, but also will ensure that they always are looking to best of breed product that is open, flexible and futureproof — so they work with the likes of big systems integrators and smaller independent software vendors to make that happen.”
New technology stack integration offers solution for challenges, both new and legacy
Aviation’s legacy systems challenge is not unique to the industry, but given its early adoption of many computerised systems, together with the subsequent technology stacks at airlines, airframers and manufacturers, all means that there are intricacies not present in other sectors.
“We see several challenges within the industry including legacy systems, acceptance of new tools and technology, data sharing, and connectivity. Legacy systems do hamper the ability to capture and utilise data but with the dawn of generative AI there is an opportunity to reach into all of these disparate silos to grab information and collectively reason over it,” Shainock says.
Integrating new technology on top of — or alongside — an existing stack can be complicated, but the benefits can be substantial.
As an example, Shainock explains, Microsoft’s technology stack offering includes “analytics, cloud computing, connectivity, data collection/ingestion, edge computing, IoT devices, modelling and simulation, security and data governance, visualisation tools such as 2D/3D rendering, and augmented reality” — including its Hololens headset for augmented and mixed reality.
Hololens and other ways of connecting people to data and technology are good examples of how the industry needs to join up the physical and digital. It’s not just about the systems: it’s about the people too.
“We also see resistance to accepting new technology and tools as a major factor,” Shainock says. “Within the industry this apprehension is at a human level. Without implementing new technology like digital twins and AI, companies may put themselves at a disadvantage to competitors. Lastly, in operations beyond data sharing, a big hurdle is connectivity — and sharing key data sets in real time. By sharing real time data, employees are better able to make successful decisions.”
Artificial intelligence and machine learning, if integrated successfully with other technologies, will bring many opportunities
Artificial intelligence is one of the new technology pillars of this decade and the next, with its evolving forms like generative AI promising incredible new benefits through automation and pattern recognition — and the prospects for integrating it with digital twinning are immense.
“Generative AI has the ability to provide real time feedback through its sentiment analysis to drive more operational efficiency and a better customer experience,” Shainock predicts. “This technology also has the ability to transform many areas of an airline’s business operations including marketing, sales, call centre, operations, et cetera. The ability for each individual to have AI assist them in their every day job will be a tremendous asset to all.”
Early forays into AI within aviation — notably for passenger experience — have not been universally successful. The concept of ‘ChatGPT, but for airlines’ is not yet robust enough for primetime. Technologies like this will, however, continue to evolve and improve, with a critical factor in their success being how they are integrated into other technologies.
Within the world of digital twins, Shainock says, next-generation digital twins will leverage AI technologies, such as machine learning and deep learning, to enhance their analytical and predictive capabilities. AI algorithms can process large amounts of data and extract valuable insights, enabling more accurate simulations, predictive maintenance, and optimised performance.”
The Internet of Things (IoT) is a critical technology family for digital twinning, with platforms, sensors, devices, physical assets and environments providing more data — and more rich data — to both design digital twins and to keep them updated.
Virtual reality (VR), including mixed reality (MR), augmented reality and assisted reality (both referred to as AR), are also key technologies to integrate. “Users will be able to visualise and interact with virtual representations of assets, systems, and processes. This can improve training, remote assistance, and troubleshooting capabilities,” Shainock says.
Cyber-physical systems will also speed up digital twinning iteration, Shainock explains. “Next-generation digital twins will bridge the gap between the virtual and physical worlds by integrating with cyber-physical systems. This integration enables real-time feedback loops, where data from physical assets is used to update the digital twin models, and insights from the digital twins can inform physical operations.
Collaboration vertically and horizontally within the supply chain is likely to be a further hallmark of next-generation digital twins, allowing expertise at multiple levels in multiple organisations to contribute — and to benefit from others’ contributions.
Lastly, predicts Shainock, “next-generation digital twins will incorporate advanced simulation and optimisation techniques. These techniques can simulate complex scenarios, optimise performance parameters, and facilitate scenario-based decision-making. They help identify optimal strategies, improve efficiency, and reduce risks.”
Author: John Walton
Published: 14 September 2023
;
