The High Energy Photonics Center: Siemens Healthineers optimizes production with Azure

The challenge: Complex manufacturing processes need modern solutions

If you think of the X-ray machine as a lamp that illuminates patients to enable precise diagnoses, then the X-ray tube is essentially the light bulb of that lamp. This bulb must function properly, last as long as possible, and indicate when it needs to be replaced before it breaks down—that’s the only way to guarantee the provision of medical care. Siemens Healthineers produces just such X-ray tubes, as well as the generators and control elements required for these imaging systems. Its manufacturing facility is divided into different areas dedicated to the various steps of the highly complex production process. Logistics areas ensure the production lines are supplied with material, and the tubes’ vacuum components are manufactured in a clean room under controlled conditions using robotic solutions. On the shop floor, employees install tubes in the X-ray tube assembly (XTA) on assembly lines. Siemens Healthineers then conducts rigorous trials in a test room before installing the XTA in computed tomography and angiography devices and delivering them to medical facilities all over the world. Demand for the end results of this complex process—the X-ray tube—is growing steadily and rapidly. "A reliable and networked manufacturing environment is fundamental to meet the high demands of the medical sector," says Peter Schardt, CTO of Siemens Healthineers.

Creating this kind of environment was a major task: to better coordinate production lines and the individual process steps, a factory needs a lot of space—but existing sites were reaching their physical limits due to increased production volume and couldn’t be expanded any further. Although the company had already digitalized a large part of its process, it wasn’t yet able to stream real-time information or conduct in-depth data analysis of, say, process anomalies along the production line.

“That’s why we decided to set up our first natively digital factory in close cooperation with Microsoft: the High Energy Photonics Center, or simply HEP, in Forchheim,” says Horst Schmidt, Head of the Power & Vacuum business at Siemens Healthineers. “To digitalize all areas of the factory, we needed a wide variety of digital solutions that interlock seamlessly.” Siemens Healthineers applied several Microsoft technologies to realize the smart factory.

The solution: Automation and digital connectivity with Microsoft in all core areas

Today, 800 people work at the 69,000 m² HEP Center in Forchheim on state-of-the-art diagnostic imaging solutions and on developing and manufacturing core parts for the company’s computed tomography, X-ray, and angiography systems. Three production floors, six office floors, and a utility building are all 100% digitally connected based on cutting-edge cloud technology.

Connected machines: Cutting-edge cloud technology streamlines production lines

Using Azure and the integration of IoT technologies in the HEP Center, Siemens Healthineers has connected not only individual machines but the entire production environment. A digital cloud infrastructure increases process efficiency and helps the company make informed production decisions.

Digital twins, i.e., virtual models of the production facilities, are a key element. They serve as the basis for simulations to optimize production processes and predict potential problems. “This means we’ve introduced predictive maintenance not only for our products in the field but also in our factory,” explains Jens Fürst, Principal Key Expert Digitalization & Automation at Siemens Healthineers. AI helps the HEP analyze data from the systems: using Azure Machine Learning and Azure AI Studio, Siemens Healthineers leverages real-time information that’s collected, connected, and aggregated by machines in the HEP. “With the help of machine learning, we’ve developed AI applications for the HEP that allow us to analyze huge amounts of data and detect anomalies within our production lines,” says Markus Kaupper, Head of Digitalization for the HEP Center at Siemens Healthineers. This enables holistic, data-based management of production. For example, employees can service machines preventively before they break down, and they can also detect potential unknown anomalies during production in real time so that they can take action before a failure occurs.

“Thanks to our AI applications from Azure AI Studios, we’re reducing disruptions and are thus bringing down our production costs,” Kaupper says.

Connected people: Teams and DevOps for agile collaboration and innovation

Just as important as connected machines in the HEP is a good system of interdisciplinary networks among people. Microsoft Teams serves as a central platform for collaboration between the shop floor and offices, thus helping the company to make decisions quickly and to continuously improve its processes. “Production orders received in the office can be transferred to production rapidly and seamlessly,” Kaupper says. Another key innovation was the connection between Azure DevOps and Microsoft Teams: if employees in the factory detect a fault, they can immediately inform their colleagues in the office using a ticket system.

However, implementing these solutions is only the beginning. Ongoing innovation and adaptation to new challenges is part of the Siemens Healthineers culture. Digital skills are constantly being expanded, while the focus is on a culture of learning and continuous improvement.

“Azure DevOps gives us the opportunity to react flexibly and quickly to rapidly changing digital requirements,” Fürst adds. “In this way, we’re creating a culture of experimentation. This means that we can develop customized solutions for specific customer requirements and build them as prototypes fast.”

Connected expertise: Integrating language models with Azure OpenAI Service

A central tenet of Siemens Healthineers’ digital strategy is the use of language models to facilitate access to knowledge that’s already stored in data and documents. Thanks to the implementation of large language models with the help of Azure OpenAI Service, employees can communicate with documents and data. This makes complex information easy to understand and simplifies access to specialized knowledge about internal factory data.

“Azure OpenAI Services creates dialogue-oriented access to knowledge that helps employees make informed decisions faster,” Kaupper explains.

The production lines are fully digitally connected, which reduces disruptions and cuts production costs. The high level of networking among people within the HEP reduces gaps in communication and makes collaboration between the shop floor and the office more efficient. And by networking knowledge, answers to key questions are always available to all employees when they need them.

The HEP is already the central element of a highly efficient smart factory network of the future. “We’re experiencing the digital transformation as a journey of discovery, on which Microsoft is supporting us with its extensive technology portfolio and pioneering concepts,” Fürst says. “And this journey is far from over.”