The pick-and-place process is one of the manufacturing processes in the production line using surface mounting technology (SMT). The precision of the pick-and-place machines (also known as “mounter”) is critical to the accuracy of the component placement. Then it influences the overall quality of the SMT products. Thus, the diagnostic module is significantly important for the mounters. From the industry and our previous studies, multiple issues can be detected from the placing pattern of the mounter, including air pressure issues, gantry issues, nozzle issues, etc. In this study, the experiments (DOEs) were conducted on several different mounter issues. Based on the placement patterns, a defect mounter placement pattern simulation model was proposed with the help of theoretical estimation and polynomial regression-based simulation. The mounter issue can then be detected and classified using a convolutional neural network (CNN) based classification model. This detection model can offer maintenance suggestions to users with over 70% accuracy for various defect categories. Furthermore, the placement pattern can be simulated for augmentation with an error of 10 µm using the proposed simulation model to build the digital twin and generate data for future studies with reduced experimental waste materials.