The goal of the project is to build upon a sensor instrumented fibre production line, by textile producer, and provide predictive quality information in a trustful way to operators an decision makers throughout the production pipeline so that humans can change the process immediately and avoid inferior quality outcomes. Through this optimisation energy loss and waste are avoided, as no costly upcycling needs to be performed on inferior goods. To achieve this AI based instrumentation of the production line, we want to close (i) causality gaps and (ii) measurement data gaps by using historic information. In a second step we want to build (iii) trustworthy embedded energy-efficient explainable AI tools into the process and (iv) provide counterfactual-based explanations for what-if analysis of the overall process. Finally, this should showcase, to others, how sustainability goals and traditional cooperate goals are not mutually exclusive and can be enabled by using advanced AI methods.
|Effective start/end date||1/04/22 → 31/03/25|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.