In recent years, due to the rapid growth of electronic products, the quality and productivity requirements for electronic product manufacturing have become increasingly strict and complex. On the one hand, the product engineering team requires the use of small and precise electronic components to design printed circuit boards. On the other hand, end consumers have higher requirements for product quality and stability throughout the entire product lifecycle. For mass-produced electronic products, printed circuit board assemblies (PCBA) are the key manufacturing design for polychlorinated biphenyl surface mount technology (SMT). The effective use of SMT is crucial for the yield performance and product quality of finished polychlorinated biphenyls, especially in handling small component package sizes (such as 0201). The general production process of the polychlorinated biphenyls authority is shown in Figure 1, where template printers, placement machines, and furnaces are the basic equipment for producing polychlorinated biphenyls. To prevent malfunctions, there are several programs on the production line that prevent malfunctions, such as solder paste inspection, automatic optical inspection, and circuit testing. However, existing fault prevention procedures are responsive to defects that may result in certain production waste and additional effort, as they require remanufacturing. Therefore, recent research is attempting to change these processes to be proactive and automated, in line with the goals of intelligent manufacturing. In a typical PCBA process, there are two main challenges that provide room for improvement. Firstly, the parameter settings of the PCBA machine are now set by domain experts, so the entire process is highly dependent on their expert knowledge and knowledge of identifying key factors, and cannot be effectively transferred to others. This complicates the PCBA process and makes it difficult to improve process performance. Secondly, utilize the existing inspection process to study the yield loss throughout the entire PCBA process, without considering the impact on machine parameter settings. Therefore, optimizing process performance is difficult to achieve, and yield performance depends entirely on domain experts.