-
Results:
OPENZDM solutions for AI quality assessment and decision support aim for data driven defect prediction and estimation of machine components degradation through drifts in data distribution. The main goal is to predict defects such as broken paper, dust or glued paper so that changes in machinery/parameters can be performed and the defect avoided and also to estimate degradation based on slight and smooth changes in the process start occurring due to component wear-out / degradation and, if data is representative enough, it may be different data distributions in time.
-
Results:
-
Results:
OPENZDM is expected to contribute to the SONAE Arauco use case in the aspects of proactive quality control towards zero-defect manufacturing through the digital twin enabled machine learning approaches for data analytics and quality assessment.
The vision of the future process with the contribution of the openZDM solutions targets an improved decision-making at the plant level to reduce defects towards a zero defects paradigm, that will facilitate defect prediction and recommendation of production recipes to adjust machine parameters according to predicted defects. In turn, improved products can be developed using the data-driven knowledge acquired, resulting in less wasted material. Towards this direction, the openZDM project performed an LCA at the beginning ot he project to compare it with the final one and also it supports integration through its platform to an LCA tool for on-demand LCA.