"There are already many examples of the successful integration of artificial intelligence in production systems, but the effort required to integrate these technologies is too high and usually not feasible," says 'KOARCH' project manager Professor Dr. Oliver Niggemann from the Institute for Industrial Information Technology (inIT) at OWL University of Applied Sciences, describing the problem. Special solutions are required from the experts in these processes - be it in commissioning, converting machines or optimizing them. 'KOARCH' (Cognitive Architecture for Cyber-physical Production Systems and Industry 4.0) aims to reduce manual effort in order to make artificial intelligence technologies widely available.

There are still a few hurdles to overcome in order to achieve this goal. From data acquisition and model learning to the generation of instructions, existing solutions need to be combined generically. Although the necessary technologies are already available, there is still a lack of interaction between the various components. The researchers in Lemgo can draw on the results and findings of many projects. "We are working on specific use cases from various industries in order to identify a suitable architecture," explains Professor Niggemann.

In the 'KOARCH' project, we are working together with industrial companies and research institutions on a new automation technology that is to be made more efficient using artificial intelligence. The overarching goal of those involved in the project is for different Industry 4.0 devices and components to work together, regardless of manufacturer, and jointly access data and information such as anomalies or optimization targets. An exchange of algorithms and solution strategies is also to be established so that a network of cooperating production systems is created.

In the first stage of the project, a customizable reference architecture will be developed to enable the practical networking of different plants and systems. The resulting standardized interfaces and modular analysis algorithms will be tested in practice at the demonstrators of the SmartFactoryOWL, a facility of Fraunhofer IOSB-INA and the OWL University of Applied Sciences, as well as at the Big Data Lab of the TH Köln. In the long term, the aim is to implement the system for industrial use.

In addition to the Institute for Industrial Information Technology (inIT) at OWL University and Cologne University of Applied Sciences, partners from industry are also involved in the research project, including Deutsche Telekom, Innovations Laboratories (T-Labs), Telexiom and Opitz Consulting. KOARCH is being funded by the Federal Ministry of Education and Research (BMBF) with a total of 1.5 million euros over a period of four years.