Leveraging Synergies of Methods from the Artificial Intelligence Domain to Produce New Solutions for Industry 4.0

Complex problems in industry 4.0 are often concerned with systems consisting of many entities that interact and communicate with each other, adapt to changes, make decisions, and try to solve tasks or achieve a certain performance. A specific application scenario is scheduling in production: A high mixture of product diversity together with usually historical growth of the industrial plant further induces complexity leading to extremely hard scheduling (and other optimization) problems. Methods from the AI domain based on either machine learning or swarm intelligence have shown to be successful within certain but different problem domains where each of the approaches has its advantages.

In this project, we want to combine these conceptually different AI methods (namely, machine learning and swarm intelligence) to complement each other, find synergies and produce better results regarding a specific application scenario. A major goal is also to create a new common field of applied research which will lay the ground for long-term collaboration between Fraunhofer KI for Life (KI4L) and Lakeside Labs (LS). To show the synergies and possibilities from these AI methods in industry 4.0, we will jointly work already in this project on a portfolio solver which dynamically selects the best swarm algorithm for a situation by applying a machine learning method.

Particular goals of the project are: i) to gain a clear picture on features, limits, advantages and disadvantages of both approaches; ii) with the insights gained we will elaborate on how to combine the two different AI methods, let them interact, and work complementary to each other which could lead to a new methodology based, e.g., on the aforementioned portfolio solver that solves cuttingedge problems, e.g., in the optimization of extremely complex/large production plants; iii) we will identify and sketch the frame conditions for a new field for (applied) research that combines the AI (swarms and machine learning) advantages also for further fields of applications. Roadmaps for further development including the sketch of a follow-up project as well as a strategic plan will be elaborated. The strategic plan will describe how the stakeholders (project partners, funding authority and additional partners) will engage to anchor and further develop the novel field of R&D in the region and create added value for it.

The benefit and potentials are considerable for a number of different stakeholders: The target users (industrial plants) will profit from a reduction of errors, disturbances and failures (improved OEE), more efficient use of resources, reduction of maintenance costs and increased profit. The benefit for the research organization is the opportunity to lay the ground for a new area of applied research which will strengthen the organization’s standing and visibility in the scientific community. It is also a great opportunity to start and intensify the collaboration of two well-established research organizations in Carinthia located in the Lakeside Science & Technology Park. The synergies in these two approaches might lead to a new, innovative solution that can be used in international projects and lead to more visibility of the research campus in Klagenfurt and might become a hub for AI. SMEs might benefit from this new approach through cooperation with the project DIH-Süd and helps them to be more competitive. Emerging products and services that will stem from our collaboration face a dramatically rising market for solutions supporting Industry 4.0 and associated further digital transformation of the production sector.

As the project will secure jobs for young highly skilled talents already in the project and create new ones in the long run, it will considerably contribute to the corresponding programme indicator.

This project is funded by:

Further information about EFRE funding at can be found at efre.gv.at.