Siemens Envisions AI-Powered Digital Twins as Part of Industry Transformation

In his keynote at Siemens Realize LIVE in Las Vegas, Tony Hemmelgarn, President and CEO of Siemens Digital Industries, laid out the fabric of living, breathing digital twins—the digital thread. 

“What we mean by a digital thread is currently connected data, the information flow,” he said. “[You get] the biggest payback if you do the full digital thread. But many times, our customers start with just one or more components.”

This year, there’s a new thread to add to the digital twins discussion: how to weave AI (artificial intelligence) into its predictions and operations. 

AI in Teamcenter

Siemens recently struck up a partnership with Microsoft to bring AI-powered collaborative apps that accept incident reports in natural speech. “Studies suggest over 70% of quality issues go unreported. The reason is, there's no access to the tools. It's hard to use paper-based tools,” said Hemmelgarn. 

Justin DiNunzio, Marketing Manager, Siemens Digital Industries Software, clarified, “The natural language processing occurs in the Teamcenter app for Microsoft Teams, which leverages Azure OpenAI Services to translate and prepopulate the problem report.”

The partnership is also expected to bring more AI-powered functions to equipment diagnostics, and asset and process management. “Siemens and Microsoft are also collaborating to help software developers and automation engineers accelerate the code generation for Programmable Logic Controllers (PLC), the industrial computers that control most machines across the world’s factories,” according to the announcement of the partnership.

Hemmelgarn said, “[The generative AI] is automatically writing the PLC (Programmable Logic Controller) code that has to be adjusted ... So the AI is writing the code to make the change. The manufacturing engineers not only receive a notification on what's going on, but he gets direct access to the updated code. He can use it on the robot, bring it right to the edge device and upload it.”

Himanshu Iyer, Senior Industry Manager, Manufacturing/Product Development, at NVIDIA said, “Many of the processes associated with building a digital twin are graphics and compute intense. NVIDIA GPUs and platforms like Omniverse enable teams to connect 3D design and simulation applications for collaborative design workflows and allow users to build physically accurate virtual worlds for training, testing and operating AI agents such as robots and autonomous machines.”

AI in NX

AI will also become part of NX, changing how you might learn and use the software. The key is to let the AI do the heavy lifting, so you won’t have to remember every menu bar and command. “We have the machine learning capability, or the technology that predicts commands based on usage patterns. This allows the designers to become really proficient [in NX] without years of practice,” revealed Hemmelgarn during his keynote.

S. Ravi Shankar, Siemens’ Director of Solutions & Technical Marketing, provided details about a preview technology, not yet ready for public distribution. “You can interact with NX through a ChatGPT-like interface to ask the program to build you a simulation model for a certain purpose. In the background, AI works with Simcenter tools to build a model. Then you can say, ‘Run these types of scenarios.’ So you’re interacting with the software like an assistant. Your job becomes much more about interpreting the simulation results and making some changes and tweaks. This can speed up the process of performing simulation,” he explained. 

AI on the Shop Floor

One area AI has proven to be quite powerful is in image processing. Siemens appears to be exploring ways to leverage this for search and retrieval operations. Hemmelgarn said, “We see examples of frontline workers going out to an assembly to perform some kind of service operation on a component but are unable to find the part number.” One solution, he speculated, would be AI-aided image-based identification. “All these images can come together and, from those, we can automatically determine what that part is back in the database,” he explained.

Siemens and NVIDIA are collaborating to bring together Siemens Xcelerator’s vast industrial ecosystem and NVIDIA Omniverse’s AI-enabled, physically accurate, real time virtual world engine. This enables full-design-fidelity, live digital twins that connect software-defined AI systems from edge to cloud. The concept was demonstrated at the Hannover Messe trade show in April 2023. 

At the event, Siemens unveiled a digital model of new FREYR Battery factories created using Omniverse. Leveraging an integration between Siemens Xcelerator and NVIDIA Omniverse, the demonstration included 3D representations of the entire factory. Such a model can be used for production planning, virtual client meetings, and simulating potential changes in a facility.

“Highlighting the integration between Siemens Xcelerator and NVIDIA Omniverse, the demo features 3D representations of the infrastructure, plant, machinery, equipment, human ergonomics, safety information, robots, automated guided vehicles, and detailed product and production simulations,” NVIDIA wrote in a blog post.

“You know, if you look at almost every engineering project today of any significant complexity, we simulate the product before we go to production,” said NVIDIA CEO Jensen Huang in a recent blog posting. “And yet, for most plants and most factories, it’s nearly impossible to do that today … and so we needed to create a very large-scale simulation platform: Omniverse.”

For more on Siemens Realize LIVE, read DE 24/7’s report here.

More Dell Coverage

Artificial Intelligence for Design and Engineering Workflows

In this white paper, learn how artificial intelligence and machine learning can improve design and simulation.

More NVIDIA Coverage