AI for product creation – AI Day celebrates successful premiere
How does artificial intelligence revolutionise product creation? This was the question asked by the 50 or so participants at the AI Day offered by it’s OWL and prostep ivip as part of the AI Marketplace on 18 November 2021. And they received exciting answers – from science as well as from concrete examples from practice.
For example, Prof. Dr. Rainer Stark, Head of the Department of Industrial Information Technology at the Technical University of Berlin, underlined the relevance of AI for product creation with a variety of different application examples – from AI-based success measurement in projects to intelligent CAD modelling of components to AI-supported classification of collisions. He also made it clear that AI knowledge must be combined with domain knowledge, as is happening in the AI marketplace.
Prof. Dr. Barbara Hammer, Head of the Chair of Machine Learning at Bielefeld University, explained how the “black box artificial intelligence” can be opened up with the help of Explainable AI technologies and how trust in machine learning can be promoted. Because the decisions of an artificial intelligence are not always comprehensible to humans. On the one hand, it is necessary to create this comprehensibility in order to react to errors in the AI. On the other hand, knowledge should also be integrated into the AI system and be able to be extracted from it. This would significantly increase the user-friendliness of AI.
After a workshop that gave participants an insight into the AI Marketplace platform, two practical projects impressively demonstrated the added value of AI.
Project examples show added value in practice
In the project “Data Preparation for Data Analytics and AI”, a standardised approach to data preparation was developed that considerably facilitates the application of AI. After all, 50-70% of the effort in a data project goes into data preparation. In this respect, there is enormous potential here for increasing efficiency, which has already been raised in the use case with the company Aibus.
The pilot project of the company düspohl together with Fraunhofer IEM showed how artificial intelligence can support the set-up of complex production machines. Here, the set-up process of a profile wrapping machine is automated with the help of AI. The machine automatically recognises the profile to be wrapped and adapts the robot arms automatically. In this way, the company can significantly reduce the duration of the set-up process.
Dr Alain Pfouga, Managing Director prostep ivip, drew positive feedback: “We succeeded in demonstrating the importance of our mission to bring AI into product creation from a scientific and economic point of view.” Next year, prostep ivip and it’s OWL plan to offer another AI Day to bring the AI community and industry together.