MINIMIZING DEFECTS IN INJECTION MOLDING USING THE TAGUCHI METHOD

Despite being one of the most widely applied net-shape manufacturing processes, plastic injection molding is extremely sensitive to the values of process parameters, and poor setting often results in dimensional tolerances which are out of specification, such as shrinkage and warpage, which increase scrap and rework costs. In this paper, the Taguchi method is used to determine a set of four experimental conditions, namely, melt temperature, mold temperature, injection pressure and cooling time, which can be used to minimize shrinkage and warpage in an injection-molded flat plaque. A 3 27 L9(3⁴) orthogonal array design was employed to decrease the number of experiments needed from 81 to 9. For each response, signal to noise (S/N) ratios under the smaller the better criterion were calculated, and then the percentage contribution of each factor was computed using analysis of variance (ANOVA). It was concluded that the shrinkage and warpage had the highest contribution of injection pressure (67.25%) and melt temperature (28.72%), respectively, followed by mold temperature (8.01% and 14.19%, respectively). Both single-response optima were found to be the same combination of parameters (200 °C melt temperature, 30 °C mold temperature, 120 MPa injection pressure and 20 s cooling time); the directly tested Trial 3 had the highest grey relational grade (GRG = 1.000), which led to a compromise setting. It is expected that the shrinkage will be reduced from the average of 1.55% to 0.64%, and the warpage will be decreased from 0.85 mm to 0.32 mm from the baseline. The results have shown that Taguchi method using the ANOVA and GRA can be a statistically sound and experimentally efficient approach to the simultaneous multi-response defect minimization of injection molding processes.