MONITORING OF THE STENCIL PRINTING PROCESS USING A MODIFIED REGRESSION RESIDUAL CONTROL CHART: AN EMPIRICAL STUDY

Abstract

This paper aims at developing a regression residual control chart to economically detect the abnormal patterns of the stencil printing process (SPP), so as to predict significant deviations that might result in nonconforming products. The SPP is widely recognized as the main contributor of soldering defects in a surface mount assembly (SMA). The earlier those abnormal conditions can be detected in the SPP stage, the less expensive the defect correction costs. Shewhart control chart is frequently used to monitor the amount of solder paste volume. However, its results can be error-prone since the solder paste volume is significantly affected by other process factors. For developing the proposed control chart, a 3^8-3 experimental design was first conducted and validated to formulate the relationship between the control variables and the SPP response. Eight process factors for SPP were initially defined, including stencil thickness, component pitch, aperture area, snap-off height, squeegee speed, squeegee pressure, solder paste viscosity, and solder paste type. The control variables of the SPP can be expressed as a linear regression function, and a regression residual control chart can then be constructed using the significant variables through the results of ANOVA analysis. Finally, the proposed control chart is employed to detect out-of-control conditions of the SPP. A Monte-Carlo simulation and an empirical evaluation were also carried out to demonstrate the effectiveness of the proposed methodology. The empirical evaluation shows that the proposed regression residual control chart provides approximately 90% of detection accuracy for the SPP.