This report describes a case study in which JMP was used to revise the correction method and evaluate its introduction into mass production, with the aim of improving the accuracy of the frequency adjustment process for ceramic oscillators. Previously, adjustment values were set for each section within the equipment, but this presented a challenge: the more detailed the section, the fewer usable data points were available, leading to unstable adjustment values. Therefore, JMP was used to analyze the effects of equipment structure, processing location, and changes over time, and factors considered to have a significant impact were incorporated into the model as hypotheses. Furthermore, a cycle of reviewing the factor composition and data exclusion conditions was repeated while confirming the regression results and predicted output through simulation. Statistical judgment was introduced for outliers to suppress the influence of data that deviates from the norm. In addition, differences in product specifications and the equipment condition immediately before processing were reflected in the correction, creating a system to continuously incorporate recent equipment trends.
