Application of statistical methods to reduce the defect of assembly process for plastic ball grid array (PBGA) component
Abstract
The manufacturing of printed circuit board assembly (PCBA) is facing many problems
especially in ball grid array (BGA) assembly. Therefore, it is important to focus on
manufacturing process. Process improvement in this study is conducted in the company
‘S’ on the printed circuit board assembly (PCBA) production line. The objectives of the
study is to collect and analyze data of selected printed circuit board assembly (PCBA)
process parameters in order to reduce the number of defects of assembly process for
plastic ball grid array (PBGA) component. There are four important variables in this
study which is factor A-print speed, factor B-solder paste thickness, factor C-semi auto
connecter jig force, and factor D-separation speed. Using design of experiment (DOE),
the experiments are designed with one-half fraction of a 24 design or 24-1 design method for screening experiments. Twenty numbers of experiments are run in production line in the factory. Finally, experiments are designed with central composite design (CCD) to optimal concentration significant factors that obtain from screening experiments. Seventeen numbers of experiments are run in production line in company ‘S’. Total thirty-seven numbers of experiments had been conducted. Software Design Expert is used for simulation and data analysis. ANOVA (analysis of variance) is performed to know interactive effect of the variables on the responses which is to reduce number of defects. It is suggested that the optimal setting of the parameters can reduce number of defects and company can save cost.