Closed loop dimensional quality control for an assembly system entails controlling process parameters based ondimensional quality measurement data to ensure that products conform to quality requirements. Effective closed-loopquality control reduces machine downtime and increases productivity, as well as enables efficient predictive maintenanceand continuous improvement of product quality. Accurate estimation of dimensional variations on the final part is a keyrequirement, in order to detect and correct process faults, for effective closed-loop quality control. Nowadays, this is oftendone by experienced process engineers, using a trial-and-error approach, which is time-consuming and can be unreliable.In this paper, a novel model to estimate process parameters error variations using high-density cloud-of-point measurementdata captured by 3D optical scanners is proposed. The proposed model termed as PointDevNet uses 3D convolutionalneural networks (CNN) that leverage the deviations of key nodes and their local neighbourhood to estimate the processparameter variations. These process parameters variation estimates are leveraged for root cause isolation as a necessarybut currently missing step needed for the development of closed-loop quality control framework. The proposed model iscompared with an existing state-of-the-art linear model under different scenarios such as a single and multiple root causes,and the presence of measurement noise. The state-of-the-art model is evaluated under different point selections and resultsare compared to the proposed model with consideration to an industrial case study involving a sheet metal part, i.e. windowreinforcement panel.
展开▼
