Traditionally, high-precision components — for example, mirror optics — are mass-produced in a time-intensive and manufactory-like process and then individually characterized using complex measurement technology. As the demand for such components grows, manufacturing via conventional processes leads to enormous machine costs, longer processing times, and limitations in the transition to high-quality serial components. New manufacturing approaches are needed to meet the increased demand and at the same time enable economical as well as qualitatively excellent production with increasing requirements and high quantities. With the help of SWAP-IT, industrial manufacturing processes are to become more flexible, efficient, and cost-saving in the future, in this case the high-precision manufacturing of mirrors. Courtesy of Fraunhofer IOF. Working under the umbrella of the Fraunhofer flagship project SWAP, which aims to make these production processes in the factory of the future more flexible and more individual while simultaneously achieving the highest quality and reduced costs, researchers at the Fraunhofer Institute for Applied Optics and Precision Engineering IOF (Fraunhofer IOF) and nine additional Fraunhofer institutes have devised a production architecture called SWAP-IT. The architecture works with artificial intelligence (AI) and can be used in a variety of application areas, including the production of high-precision components. The SWAP project is based on a concept of heterogeneous, workload-optimized robot teams and production architectures. The basic idea behind the SWAP-IT solution is to create a scalable cyber-physical production system that is very lean and can be flexibly applied to a wide variety of production processes. SWAP-IT relies on workload-optimized manufacturing, which makes it possible to decouple the scaling of the number of workpieces, size, and accuracy from the individual performance of the processing machines. This involves combining various operating resources and additive and subtractive machining processes, functionalization, and characterization, as well as measuring and handling processes, analyzing their workflow directly on site and the linking of multiple cooperating robot stations as a result. The workspace of the novel machine environment is divided into subsegments, which enables the use of compact, precise, and economical processing stations. As a result, individually adaptable components are manufactured for both medium and large quantities with consistently high optical surface quality. For mirror optics manufacturing, this efficiency enables a reduction in manufacturing time of 30% per mirror, combined with an improvement in the quality characteristics of the mirror surface of up to a factor of two. The high-precision, parallelized manufacturing solution opens up individual component sizes and surface accuracies in the micrometer to nanometer range, supported by the latest AI methods. The modular structure of the SWAP-IT production architecture breaks up the static structures and schematic processes of classic production facilities and makes the work steps more flexible. At the same time, a uniform and semantically simplified description language for machines, processes, and products enables the integration of operating resources such as machines, robots, or autonomous transport systems. The partner institutes are working on four industry-relevant use cases in the context of manufacturing and automation to apply this architectural solution and demonstrate added value. SWAP and its production architecture will be presented at the Hannover Messe from April 17 to 21.