Data driven approach to quality assessment of 3D printed electronic products

Quality issues are of utmost importance when it comes to 3D printing technology and its applications in the field of electronics manufacturing. This paper presents a data driven approach that enables the condition-based monitoring of 3D inkjet printing process and the preservation of important quality characteristics of the manufactured electronic products in relation to their design specification. The proposed assessment approach for 3D inkjet printing builds upon the capabilities of computational intelligence algorithms to recognize, and ultimately to predict, relationships between key process operational/environmental parameters and respective quality of fabricated structures. The use of neural network methods in predicting the quality of printed electronics structures in terms of their geometrical characterization and shape accuracy, assessed against the original specifications, is presented and demonstrated. Algorithm performance characteristics are also studied and reported.