Siemens Spotlights Process Preparation X & Digital Twin Composer at NVIDIA GTC Berlin
Explore the recap of Siemens at NVIDIA GTC in Berlin from October 20-22, 2026. NVIDIA GTC in Berlin is NVIDIA’s premier European AI conference, bringing together developers, researchers, industry leaders, and technology innovators to explore the latest advancements in artificial intelligence, accelerated computing, robotics, simulation, and digital twins.
Siemens was centerstage showcasing how AI-powered digital twins can help manufacturers manage growing product complexity, accelerate innovation, and improve production readiness. The event serves as a platform for showcasing breakthrough technologies, customer success stories, and collaborations that are shaping the future of industry.
What were the key takeaways at the event? First, Siemens announced its latest collaboration with NVIDIA, combining Teamcenter software and Process Preparation X and NVIDIA Cosmos to advance PCB inspections from design to the factory floor. Electronics manufacturers will be able to create and optimize AI-powered workflows for automated optical inspection (AOI) processes and work with a virtual “golden board” using Digital Twins to:
- Boost engineering efficiency and quality while reducing manual effort
- Accelerate production planning, inspection, prep and optimization
- Decrease dependence on physical samples by generating and analyzing potential defects before physical production begins
- Establish more agile, adaptive factories
At the forefront, Siemens Electronics Factory in Erlangen, Germany will serve as “Customer Zero”, a use case to find the path to achieving big goals: up to 85 percent less manual engineering effort in AOI program generation, 40 percent fewer false calls and more than 5 percent first-pass yield improvement.
In addition to accelerating PCB inspection, Siemens also highlighted its vision for the Industrial Metaverse through Digital Twin Composer, illustrating how AI-powered digital twins can bridge the gap between physical and virtual operations, helping manufacturers make smarter decisions, reduce risk, and bring products to market faster.
Interested in learning more? Explore a recap below of what Siemens demonstrated at NVIDIA GTC.
What if PCB inspection could start before the first board exists?
PCB inspection traditionally waits for the first physical board. See how Siemens and NVIDIA are using the digital twin and industrial AI to prepare AOI earlier and support inspection decisions in production. In electronics manufacturing, every day gained between design and production matters. A critical part of the process is ensuring that every printed circuit board (PCB) is assembled correctly and free from defects before products move downstream.
Automated Optical Inspection (AOI) plays an important role in this process. Using high-resolution imaging, AOI systems help verify component presence, orientation and assembly quality across large production volumes. However, preparing and optimizing AOI programs traditionally depends on a physical reference PCB, often called the “golden board.” That means an important part of production preparation must wait until the first physical board becomes available.
But what if inspection preparation and optimization could begin with the digital twin, before physical production starts?That’s where Siemens’ Electronics Factory in Erlangen, Germany enters the demonstration and is the success story where Siemens and NVIDIA are demonstrating this new approach with Process Preparation X, shifting more of AOI preparation and optimization into a digital and AI-supported workflow.
Moving inspection preparation into the digital world
The idea starts with information manufacturers already have. Teamcenter provides PCB design, bill of materials and component information, while Process Preparation X validates, normalizes and assembles this information into a comprehensive digital twin of the printed circuit board assembly, or PCBA. NVIDIA Cosmos then uses this digital context to create a photorealistic virtual reference board, while NVIDIA Cosmos Data Factory generates synthetic circuit-board defect images to support inspection preparation and calibration.
One result is a photorealistic virtual reference board. Instead of waiting for the real reference board before important AOI preparation can begin, manufacturers can start preparing the inspection process using the digital twin before the first PCB reaches the production line. This is a meaningful shift, but the workflow goes beyond generating a virtual reference board. It can also generate synthetic, photorealistic samples representing different production conditions, supporting earlier AOI program generation and threshold calibration without relying exclusively on physical boards or manually introduced defects.
The objective is therefore broader than replacing a real golden board with a digital one. It is about moving more of the inspection-preparation process upstream, where engineering and manufacturing data already exist digitally.
From virtual preparation to AI-supported inspection
The workflow does not stop when production starts. When an AOI system identifies a potential defect, an AI-supported secondary inspection capability can help analyze the detected condition and determine whether it represents a real defect or a false call.
That creates a connected flow spanning the digital and real stages of manufacturing:
For manufacturers, the intended value spans the workflow: earlier inspection preparation, reduced manual engineering effort, less dependency on physical samples, better-calibrated inspection processes, reduced manual verification, faster production readiness and improved manufacturing quality.
At the NVIDIA GTC event, it was shown how Siemens Electronics Factory Erlangen, where the technology is being demonstrated as part of the broader Siemens and NVIDIA work on adaptive, autonomous and AI-powered production, the project is targeting up to 85 percent less manual engineering effort in AOI program generation, more than 40 percent fewer false calls and more than 5 percent first-pass yield improvement. These are project targets rather than established customer results. The significance goes beyond any individual metric or capability. It demonstrates how the role of the digital twin can evolve within manufacturing.
When the digital twin becomes part of the manufacturing process
The comprehensive digital twin is often used to represent, understand and virtually validate a product or production environment. Combined with industrial AI, it can also become an active foundation for manufacturing and quality processes. PCB inspection makes this tangible. The PCBA digital twin provides the product and manufacturing context needed to support virtual AOI preparation and calibration. Once production begins, AI-supported secondary inspection can continue the workflow by helping distinguish real defects from false calls.
In this way, the digital twin connects engineering intent with real-world inspection across the digital and real stages of production, evolving from a virtual representation of the product into an active foundation for AI-powered manufacturing and quality processes.
Rather than waiting for the physical product to begin inspection preparation and optimization, manufacturers can start with the digital twin and continue using AI to support inspection decisions once production begins.
Siemens Digital Twin Composer featured at NVIDIA GTC Berlin
In addition to Siemens and NVIDIA announcing a new collaboration to accelerate production planning, inspection preparation, and optimization, Siemens also showed the second generation of Digital Twin Composer, and how it’s advanced into the decision cockpit of the Industrial Metaverse, bringing together engineering data, operational intelligence, AI and simulation in a single immersive environment.
Powered by NVIDIA Omniverse technologies, Digital Twin Composer 2612 is now connected to real operational data like IoT – turning the digital twin into an always-on, living representation of the factory, enabling manufacturers to analyze disruptions, validate alternatives with simulation and optimize operations before actions are executed in the real world. This gave attendees a firsthand look at how AI, simulation, and digital twins are converging to transform manufacturing.
The demonstration will also center on the Siemens Electronics Factory Erlangen, showcasing how a real-world production facility can be represented as a continuously synchronized digital twin powered by NVIDIA Omniverse.
Siemens shares the common manufacturing challenge: planning changes to a production hall or line requires digital and physical assets to remain aligned. The goal is to demonstrate how Digital Twin Composer addresses this challenge by automatically bringing real shop-floor changes back into the digital twin. By combining engineering data, operational intelligence, IoT data, AI, and simulation in a single environment, Digital Twin Composer creates an always-on digital twin that remains connected to the real factory.
Attendees saw examples from the Erlangen Electronics Factory, including physical AI simulation, live operational technology data displayed directly within the digital twin, SMT production monitoring, end-of-line testing, and AGV heatmap analysis. Together, these solutions show how manufacturers can analyze disruptions, evaluate alternatives, and validate decisions in the digital world before taking action in the physical factory.
The key message is: the digital twin no longer just shows how the factory was designed. It helps determine how the factory should respond, enabling smarter planning with a living, continuously synchronized representation of operations and the potential to reduce planning effort by up to 50%.
Stay tuned for more information following the event.
