Prediction of Solder Joint Quality Using a Data Mining Methodology for Surface Mounted Technology Process

In the application of automated optical inspection (AOI) machine for printed circuit boards (PCBs), the accurate inspection on solder joint defects of the electronic component become more critical and difficult due to electronics device assembly with miniaturization of components and denser packing of boards. Moreover, the inspected PCB with several false judgments detected by AOI will increase checking time by human operators, along with the decrement of working efficiency in surface mounted technology (SMT) process. In order to solve this problem, a new procedure structure for SMT process is proposed to enhance the judgment ability of AOI. Furthermore, the data mining methodology contained proposesorthogonal and clustering analysis is added. The experimental results showed that the proposed methods are not only more efficient, but also provide an accurate prediction ability in SMT process.