Increasingly, physical products are born with digital twins, and the shared DNA that unites the two is a blend of product idea and real-world physics. The digital twin is conceived along with the product idea, serves as a virtual template for production, grows and develops in the product creation phase, and remains ‘joined at the hip’ with the physical product throughout the latter’s life cycle. And it is very much a child of Industry 4.0, Automation, Big Data and the Internet of Things.
Digital twins are possible for all kinds of physical products – from microchips to luxury cruise liners – and their potential benefits for industry are immense. Developers can avoid the expense of physical prototypes and endless testing iterations by using digital twinning to run through myriad scenarios in next to no time, develop and sift through multiple solution strategies, and explore and implement improvement options. In this way they can, for instance, develop services and have them at the ready before customers even realize they need them.
Aging at the push of a button
Take predictive maintenance, for example. This is where sensors continuously collect machine condition data and send it to the cloud where it is aggregated by the machine manufacturer and used to calculate component wear rates, loadings and life spans. The machine operator is thus able to determine the optimal time for maintenance, thereby avoiding the cost both of major repairs and premature or unnecessary maintenance.
Digital twins take this method to a whole new level because they enable a more profound understanding of the collected data and hence far more accurate predictions. For one thing, time can be manipulated in the digital world, enabling operators to preview what sort of condition the machine will be in after several thousand hours of operation. Different climate conditions can also be modeled in much the same way.
It is therefore possible not only to accurately predict when critical wear states will arise, but also to determine the fastest way of remedying them in various scenarios. SAP’s Asset Intelligence Network is a case in point. It is a cloud-based hub that connects processes and systems at the automation level. It also enables in-service assets to be controlled and fed configuration and other data via the network.
Digital twinning provides an integrated approach that maximizes asset availability and streamlines spare parts procurement. What’s more, parts can be repaired much more quickly because key information on their structure and geometry is stored in the digital twin and hence is readily available.
Of course, digital twins are also used earlier in the product lifecycle than the maintenance phase. Much earlier. The German packaging machinery manufacturer Optima , for example, uses simulations to digitally map, test and validate its products. The company is able to model and optimize changeovers and product flows across the entire lifecycle of its machines before they are even built. Once the physical twin is built and installed, it sends throughput data, downtime analyses and energy data via the cloud to its digital twin, thereby enabling the machine operator to optimize planning.
Digital is faster and more cost-effective
The potential savings from digital twins are enormous, especially when it comes to prototyping . With conventional product development, physical prototypes tend not to be built until very late in the process, by which time the design is very close to being production-ready and is thus less prone to changes. The main reason for this is cost: modifying the physical prototype every time there is a change in the design would be prohibitively expensive. Hence creating a digital prototype that can be used to run simulations and can be modified at any time at minimal cost is by far the better option.
These digital models are far more sophisticated than mere 3D visualizations. They incorporate big data and the Internet of Things, so they contain vast stores of product information and respond to physical factors in the same way as their physical counterparts.
Not that the applications of digital twinning are confined to production. General Electric (GE), for example, is fitting cruise liners with sensors that are designed to perform a range of functions, such as measuring component wear rates, collecting meteorological data and optimizing route planning. The system is designed to save time and fuel and in some cases even help the ship avoid rough seas. And if a component wears out quicker than anticipated, the system can alert the captain so that he or she can adjust the route so as to reduce the load on the part concerned.
The GE system is based on a digital twin that is anchored in the safe harbor of a data center, where the only headwinds are digital. GE already uses similar software to administer data from wind turbines, oil rigs and even aircraft. For example its aircraft system will collect anything between one and two terabytes of data per engine in the course of a flight between Frankfurt to London. The data is transmitted via satellite to a data center, where it generates a real-time digital twin of each engine. If a potential fault is detected, the maintenance personnel can have the necessary replacement part ready and waiting at the destination airport when the real aircraft lands.
Digital twins at HANNOVER MESSE
Digital twins and simulation feature strongly at HANNOVER MESSE, where they are explored in various forms at numerous showcases. One of these is the Digital Factory show, which, among much else, explores the question as to how manufacturers can get the most out of the latest digital technologies.