Siemens and NVIDIA Bring Industrial AI to Factory Planning

Digital twins and AI agents tested at Siemens’ Erlangen plant target faster production preparation and more flexible scheduling, linking industrial software with NVIDIA’s simulation and accelerated computing technologies.

Production planning and factory logistics are at the centre of new industrial AI applications developed by Siemens and NVIDIA at the Electronics Factory in Erlangen, Germany.

The partners have been testing the tools in live production since expanding their collaboration at CES in January. Their October announcements bring together virtual production preparation, inspection training and agent-based planning, with the Erlangen site serving as a reference for wider deployment.

For printed circuit board assemblies used in drives and frequency converters, AI generates realistic training images from digital twins. This allows inspection systems to be prepared before physical prototypes exist. Siemens reports potential reductions of up to 85% in engineering effort for virtual production preparation and up to 40% in time to market.

Planning agents generate production scenarios when orders arrive, weighing capacity, material availability and delivery dates. Proposed changes are validated against a digital twin of the factory before implementation. Siemens says this approach can enable cost savings of up to 20%.

A broader announcement also covers AI infrastructure. Siemens and Delta have developed prefabricated electrical power modules which, the company says, can reduce data-centre time to market by up to 50%. Digital twins support assessment of grid connections, cooling and noise before construction.