Siemens PLM has a strategy to reinvent manufacturing by threading digital information from concept, through design to monitoring performance in the field. It is easy to get confused as to what all this means, but some concrete examples and definitions help explain it all.
INDIANAPOLIS — There is a lot of talk about digitalization in manufacturing. Siemens PLM is not alone in discussing digital twins and threads, cloud computing, big data, Industrial Internet of Things, and agile manufacturing.
One could be forgiven if it seems a bit nebulous – the sheer scope of the task means it is hard to get a handle on it.
Every company offers its own vision. Boundaries are shifting upstream and down as PLM companies buy or build new capabilities. The goal seems to be to offer, as much as possible, one stop shopping to any potential customer. Siemens PLM is certainly taking that approach, both on a big scale and small. Both were on display at the 2017 Siemens PLM Connections conference, held in Indianapolis in May 2017.
In particular, one discussion and demonstration made all the buzz seem, well, less buzz and more real.
Big Picture, Digital Threads and Twins
“There are a number of enabling technologies that are changing the way you think about the digital twin,” said Siemens PLM CEO Tony Hemmelgarn, speaking in a keynote speech at the conference. The digital twin concept is a computer representation of any physical asset that is constantly updated with data from a variety of sources. Its purpose is aimed at maintenance, repairs, updates, and future designs.
While the concept may have started with a product (an aircraft engine in fact), Hemmelgarn stressed that digital twins could also reflect both production processes and performance models – of most anything.
He put the key enabling technologies into three categories, which made a lot of sense to me and helped me understand how all of this could work. The essence of his message was simple, tools for Ideation, Realization, and Utlization.
Ideation. These are existing and emerging tools to bring products “to life,” in Hemmelgarn’s words, through the R&D and design phase. These include generative design technology, a buzz phrase that comprises both CAE simulation and optimization techniques that let computers do some of the design work. Optimization includes topology and shape optimization as well as multidomain optimization, or MDO. MDO will be critical in that any product today and in the future is both smart, meaning lots of software, and composed of mechanical, electrical, and mechatronics systems. The recent purchase of Mentor Graphics, the electrical design company, shows Siemens is thinking of this convergence of disparate disciplines.
Realization. This is the bundle of tools used to actually make the product. An exciting and emerging tool is additive manufacturing, but there is also a lot going on in advanced robotics and even machine learning. Machine learning techniques could help manufacturers optimize planning and decision making, such as supply chain management and even electricity usage, according to the company. Combining additive manufacturing with robotics is the key to larger parts, and there is no shortage of good ideas in this area as well.
Utilization. This is where the true value of big data analytics, cloud technology, and knowledge automation converge. Collecting data through the cloud and then automating how knowledge is extracted from noise by applying the emerging big data techniques was another of his themes.
What might make Siemens PLM unique is that its parent, Siemens AG, is one of the largest industrial conglomerates in the world. It supplies gas turbine engines, manufacturing equipment and automation, as well as manufacturing operations management software. In his view, it can test out practical techniques in demanding situations, providing both a laboratory and credibility.
It also seems to let them think in truly end-to-end fashion. While acknowledging that there are a number of suppliers for both Ideation and Utilization, “many miss the middle part,” said Hemmelgarn. “Someone has to make it and build it. We make things and use our own software doing it and we learn a lot from doing it.” The Siemens Digital Enterprise Suite is the result. It covers nine different segment areas of what, in his opinion, a manufacturing entity needs to employ.
Future Directions of MFG
One of the key insights that CEO Hemmelgarn stressed was that digital twins come in three ways, in some ways matching the three key words he used to organize his talk. One is the digital product twin, matching to Ideation tools. Another is the digital production twin, the production process used to make the product and matches Realization. The third is the digital performance twin, matching the set of big data and analytics under utitlization.
The Siemens team also provided a tangible example of these tools used to help One Aviation, a maker of small general aviation airplanes, including a twin engine very light jet, or VLJ, called the Eclipse EA500 and variants. The CEO and founder of One Aviation, Alan Klapmeir joined Jim Rusk, the chief technology officer and senior vice president for Siemens PLM. Klapmeir stressed the importance of all three digital twins at the aircraft company, and how maintaining the digital twins allowed them to plan better for future upgrades.
The pair then went through all of the steps needed to change the design of the basic Eclipse into a larger version called the Eclipse Canada. A key point about most any transport vehicle, car or airplane, is how smart it has to be and how much electronics and wiring needs to be included. In fact, Hemmelgarn eluded to the fact that their purchase of Mentor Graphics was spurred by customers demanding that Mentor Graphics products, especially wire harness design, be compatible with both Teamcenter and their CAD NX package.
A number of software tools were used to build and maintain digital twins for product, process, and performance. The demonstration included simulation of air flow around landing gears to programming a robot that would build the plane, with other robots that would sonically inspect the carbon fiber panels. There was even a demonstration of a Stratasys additive manufacturing device attached to a robot that built up a prototype nose cone for evaluation on the Eclipse Canada.
Not that anyone is constrained to using only Siemens tools. Hemmelgarn stressed that any single Siemens product may not be the preferred solution, and some times may not be. The key take away idea is that here is at least one version of how to make sense of this emerging paradigm for manufacturing.