Fri, 21st Nov 2025 (Yesterday)

By Melvin Hipolito, Editor

Synopsys has introduced a simulation-based framework for real-time optimisation of manufacturing processes, developed with NVIDIA, Microsoft and Ansys. The digital twin solution, which has been deployed by packaging specialist Krones, reduces industrial simulation times from multiple hours to under five minutes, aiming to improve efficiency on factory floors.

Framework details

The system uses GPU acceleration with support from NVIDIA Omniverse, CUDA-X libraries, and OpenUSD, running high-fidelity Ansys Fluent simulations via Microsoft Azure. By enabling accurate virtual models of entire assembly lines, manufacturers can run near real-time scenario testing and adjust production variables, such as bottle shape and liquid viscosity, without disrupting live operations.

The deployment at Krones focused on bottle filling lines, where variables were tested and optimised digitally, leveraging cloud resources for scalability. Ansys Apex Channel Partner CADFEM Germany GmbH assisted in tailoring solver settings for Krones' requirements, further accelerating simulation performance. Visualisations and analytics were delivered on a digital twin representing the full factory environment.

Operational impact

The reduction in simulation time allows manufacturing teams to run rapid scenario comparisons and implement immediate production changes based on data-driven insight. The framework supports continuous decision-making and resource allocation, helping to reduce waste and improve operational efficiency. It also encourages closer collaboration between engineering, operations, and R&D departments through quicker feedback cycles.

The system is designed for custom, sector-specific deployment, reflecting distinct industrial requirements across different manufacturing processes. It intends to extend high-speed digital simulation benefits beyond manufacturing, with the platform forming the basis for other data-intensive, simulation-heavy use cases, including healthcare applications, according to Synopsys.

Collaborative approach

OpenUSD protocol is used as a common format for interoperability between computer-aided engineering tools, so teams can efficiently iterate simulation results and coordinate across different specialisations. The collaboration included SoftServe's expertise in integrating and deploying the solution to meet manufacturing standards.

"For manufacturers, the ability to optimise dynamic processes in real-time has been a massive challenge, as traditional, high-fidelity simulation is often too slow to be practical for immediate factory-floor decisions," said Rev Lebaredian, Vice President of Omniverse and Simulation Technology, NVIDIA. "By integrating Synopsys' accelerated physics solvers and NVIDIA Omniverse libraries on the Microsoft Azure cloud, this open framework allows manufacturers to simulate entire factory floors, test scenarios, and optimise operations in real time without disrupting physical systems."

"It's remarkable what can be accomplished when working with technology partners who share a relentless focus on solving customer challenges," said Prith Banerjee, Senior Vice President of Simulation & Analysis Incubation, Synopsys.

"Microsoft Azure provides the high-performance cloud foundation manufacturers need to unlock the full potential of AI and simulation," said Dayan Rodriguez, Corporate Vice President, Manufacturing and Mobility, Microsoft.