Study on the Quality Improvement of Injection Molding in LED Packaging Processes Based on DOE and Data Mining

The LED (light emitting diode) packaging is a very important process between semiconductor manufacturers and the electric product manufacturers. And the injection molding process in LED packaging is critical to the quality of the final products. Based on the analysis of the injection molding processes, the main quality problems and their possible causes are studied. Then the RSM (response surface methodology) of DOE (design of experiment) is used for the optimization of the producing parameters. The optimized parameters, the mold temperature, the warm-up temperature, the screw pressure and the screw time, are found with DOE. In the running process, the CTQ (critical to quality) is controlled with SPC (Statistical process control) by different dimensions like product, product catalog, time and devices. Finally, a model of continuous quality improvement based on data mining is put forward. The association analysis is used for the parameter optimization in the running process.