Webinar 1: AI-enabled Optimization of Production Processes
On this page, you'll find the info regarding the first webinar in the "How can AI leverage smart manufacturing?" series. This takes place on Thursday, 29 April 2021.
Industrial production processes are often very complex: they consist of multiple steps and the different parameters of each individual production step can highly influence the end result. Consequently, obtaining good process parameters which guarantee product quality and productivity, can be a cumbersome task, particularly in case measuring (intermediate) product quality is hard and experimentation is expensive. In additions, it is often necessary to make a trade-off between conflicting objectives, e.g. minimizing process time while maximizing product quality.
During this webinar, we introduce new approaches making use of AI-techniques to optimize process parameters. Firstly, we will illustrate by means of an industrial adhesive joining use case how Gaussian Processes can be used to retrieve cost-effective process parameters which guarantee structural strength of the end product. Secondly, we will discuss how AI-techniques can be used to realize anomaly detection systems and active learning methods can provide online quality control for additive manufacturing, able to cope with changing manufacturing and environmental conditions, and demonstrate the effectiveness of the techniques using first experimental results. Finally, Nicolas Deruytter, CEO of ML6, will explain how Machine Learning is already used today to improve efficiency and reduce the carbon footprint of production processes
|13h00||Introduction to the Flanders AI Research program||Sabine Demey, Program Director Flanders AI Research Program|
|13h15||Gaussian Processes to support data driven process parameters selection: an adhesive bonding case study||
dr.ing. Jeroen Jordens – Flanders Make
dr.ir. Ivo Couckuyt - UGent
Prof. dr. Inneke Van Nieuwenhuyse – UHasselt
|13h35||Monitoring solutions to allow adaptive process control for 3D metal printing||
Brian Booth - UGent-IPI
Dries Verhees - Flanders Make
|13h55||How ML can improves efficiency and reduces carbon footprint of production processes||Nicolas Deruytter - Managing Director – ML6|