Siemens has launched Simcenter PhysicsAI as an add-on for Simcenter STAR-CCM+ to enable engineers to generate predictions up to 1,000 times faster via AI-enhanced reduced-order modelling, transforming design exploration and decarbonisation efforts.
Siemens has broadened its Simcenter engineering simulation and test portfolio with Simcenter PhysicsAI, a new add-on for Simcenter STAR-CCM+ aimed at helping engineers test far more design options in far less time.
The software uses geometric deep learning to turn CFD results and other simulation data into AI-based reduced-order models, allowing teams to generate quick predictions for new geometries without having to rerun full solver workflows each time. Siemens says the approach can cut exploration time from days to minutes and deliver predictions up to 1,000 times faster than conventional simulation runs.
For industrial decarbonisation teams, the attraction is obvious: faster model-led design cycles can support earlier optimisation of energy systems, thermal equipment, fluid handling, heat exchangers and other carbon-intensive assets where performance gains often depend on many iterative design choices. By making it easier to assess thousands of variants, Siemens is positioning the tool as a way to reduce the cost and delay that can otherwise slow engineering decisions.
Sam Mahalingam, executive vice president for Simulation, HPC and AI at Siemens Digital Industries Software, said the aim was not to replace physics-based modelling but to make it more scalable. He described the software as a way to help customers learn from existing work, accelerate exploration and use AI to identify the most promising solutions more quickly.
According to Siemens, Simcenter PhysicsAI can train predictive models on both historical and newly generated simulation data, including earlier design-of-experiments studies. The company says this should reduce the need for repeated CFD runs, while still allowing engineers to validate AI predictions against high-fidelity simulation results inside the same environment.
The product also adds GPU acceleration, with Siemens saying inference can run up to 100 times faster on a GPU than on a CPU. That said, academic work on GPU-enabled CFD has also highlighted that speed gains do not automatically translate into lower power use or better economics in every HPC configuration, particularly when more than one GPU is required.
Simcenter PhysicsAI is now available as an add-on to Simcenter STAR-CCM+. Siemens says it will continue developing the tool within its wider Simcenter and Altair ecosystem, alongside broader engineering data science and AI capabilities designed to embed surrogate modelling and reduced-order modelling more deeply into CAD, CAE and CFD workflows.
- https://www.aerospacemanufacturinganddesign.com/product/rapidly-explore-thousands-design-variants-using-ai-powered-models/ – Please view link – unable to able to access data
- https://news.siemens.com/en-us/siemens-simcenter-physicsai/ – Siemens has introduced Simcenter PhysicsAI, an add-on to Simcenter STAR-CCM+, which leverages geometric deep learning to accelerate AI-powered design exploration. This software enables engineers to rapidly evaluate thousands of design variants, reducing design exploration time from days to minutes. It also allows for near-instant predictions on new geometries within the simulation environment, with built-in validation against high-fidelity CFD results to support faster, more confident decisions. Additionally, Simcenter PhysicsAI empowers engineers to reuse existing CFD results as training data, facilitating the creation of predictive models more efficiently.
- https://www.siemens.com/en-us/products/simcenter/engineering-data-science-ai/physicsai/ – Simcenter PhysicsAI is a software solution that accelerates simulation workflows by transforming historical data into reliable, high-fidelity predictive models. Powered by cutting-edge geometric deep learning, it uncovers relationships between shape and performance across various physics domains. This tool enables teams to evaluate more concepts and make confident design decisions, delivering data-driven insights up to 1000 times faster than traditional solver simulations. It integrates machine learning for simulation across CAE workflows, predicting product KPIs directly on mesh or CAD models, thereby reducing simulation time from hours to seconds.
- https://www.siemens.com/en-us/products/simcenter/engineering-data-science-ai/ – Siemens’ Engineering Data Science and AI solutions aim to accelerate product innovation by embedding AI and data science into engineering workflows. These solutions include AI-powered surrogate models, reduced-order modeling, and intelligent design exploration, enabling engineers to quickly transform simulation and test data into actionable insights at every stage of product design. The offerings allow for the incorporation of AI and machine learning into CAD, CAE, and CFD workflows, training, validating, and deploying AI surrogate models with intuitive no-code workflows, and predicting scalar, vector, and full-field results up to 1000 times faster than traditional solvers.
- https://plm.sw.siemens.com/en-US/simcenter/integration-solutions/reduced-order-modeling/ – Simcenter Reduced Order Modeling is a platform for building, validating, and exporting reduced-order models (ROMs) from simulation and test data. It offers an open platform for ROM, allowing the use of high-fidelity simulation or test data from any source to make super-fast predictions, enabling faster decision-making, co-simulation, and real-time applications including control and monitoring. The platform is easy to use for both experts and non-experts, providing a comprehensive range of best-in-class data reduction methods, from control theory to neural networks and Artificial Intelligence (AI).
- https://arxiv.org/abs/2012.07938 – The paper presents SimNet, an AI-driven multi-physics simulation framework designed to accelerate simulations across a wide range of disciplines in science and engineering. Unlike traditional numerical solvers, SimNet addresses various use cases, including coupled forward simulations without any training data, inverse and data assimilation problems. It offers fast turnaround times by enabling parameterized system representation that solves for multiple configurations simultaneously. SimNet is integrated with parameterized constructive solid geometry as well as STL modules to generate point clouds and is customizable with APIs that enable user extensions to geometry, physics, and network architecture.
- https://arxiv.org/abs/2404.02482 – This study investigates the compute speed, power consumption, and service unit cost considerations for the GPU acceleration of Computational Fluid Dynamics (CFD) workflows on High-Performance Computing (HPC) architectures. Using ANSYS Fluent, a leading commercial code in CFD, the research assesses different CPU architectures and GPUs. The findings indicate that while GPU compute speed is faster, the initialization speed, power, and cost performance are less clear-cut. The study suggests that in situations where more than one GPU is required, their adoption may not be beneficial from a power or cost perspective.
Noah Fact Check Pro
The draft above was created using the information available at the time the story first
emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed
below. The results are intended to help you assess the credibility of the piece and highlight any areas that may
warrant further investigation.
Freshness check
Score:
8
Notes:
The article reports on Siemens’ recent announcement of the Simcenter PhysicsAI add-on, dated May 27, 2026. ([news.siemens.com](https://news.siemens.com/pl-pl/siemens-simcenter-physicsai/?utm_source=openai)) This aligns with the publication date of the source article, indicating freshness. However, the article’s content closely mirrors the press release, suggesting potential recycling of information. ([automation.com](https://www.automation.com/article/siemens-simcenter-physicsai-add-on-ai-powered-cfd-design-exploration?utm_source=openai))
Quotes check
Score:
7
Notes:
The article includes a direct quote from Sam Mahalingam, executive vice president for Simulation, HPC and AI at Siemens Digital Industries Software. ([news.siemens.com](https://news.siemens.com/pl-pl/siemens-simcenter-physicsai/?utm_source=openai)) A search for this quote reveals it is present in the press release dated May 27, 2026. ([news.siemens.com](https://news.siemens.com/pl-pl/siemens-simcenter-physicsai/?utm_source=openai)) This suggests the quote is directly sourced from the press release, raising concerns about originality and potential reuse of content.
Source reliability
Score:
9
Notes:
The article originates from Aerospace Manufacturing and Design, a publication focused on aerospace industry news. While it is a niche publication, it is reputable within its field. However, the article’s content closely mirrors Siemens’ press release, indicating a lack of independent reporting and potential reliance on a single source.
Plausibility check
Score:
8
Notes:
The claims about Simcenter PhysicsAI’s capabilities, such as reducing design exploration time from days to minutes and providing predictions up to 1,000 times faster than conventional simulations, are plausible and align with Siemens’ press release. ([news.siemens.com](https://news.siemens.com/pl-pl/siemens-simcenter-physicsai/?utm_source=openai)) However, the absence of independent verification or additional sources raises concerns about the accuracy and potential exaggeration of these claims.
Overall assessment
Verdict (FAIL, OPEN, PASS): FAIL
Confidence (LOW, MEDIUM, HIGH): HIGH
Summary:
The article closely mirrors Siemens’ press release, with direct quotes and content suggesting a lack of independent reporting. The absence of additional sources or verification raises significant concerns about the originality and reliability of the information presented. ([news.siemens.com](https://news.siemens.com/pl-pl/siemens-simcenter-physicsai/?utm_source=openai))
