Mathematical modelling of plastic injection molding process using Central Composite Design (CCD) sampling and PRSM

Injection Molding is the most widely used plastic manufacturing process for producing outputs from a mold, particularly in polymer processing. This process is highly recommended to deliver high-quality product with low-cost production in a brief time. A Design of Experiment (DoE) approach is adapted to generate the experimental samples and the sampling method proposed to be studied in this research is the Central Composite Design (CCD). The study presents the application on Central Composite Design (CCD) and Response Surface Methodology (RSM) for the generation of the mathematical model. Four experimental parameters were chosen as the input variables which are pigmentation percentage, barrel temperature, barrel pressure and cooling time, while shrinkage is the output response parameter studied. The quadratic polynomial model is selected to significantly predict the interactions of the parameters and the response factor. The evaluation of the model is done through the analysis of variance (ANOVA) and 3D response surface plots are generated to illustrate the cross factors' relationships. From the ANOVA results obtained, the reduced model offers a better fit with a PRESS RMSE of 6.89 and PRESS R² of 71.2% in comparison to the full model. It can be concluded that CCD sampling is able to soundly predict the response of the parameters studied in the injection moulding process based on the statistical diagnostics obtained. It is also noted in both models that the pigmentation percentage (x1) is most influential factor in minimizing shrinkage in the plastic injection moulding process.