| dc.description.abstract | The R&D and application technology of the plastic molding industry has stabilized. However, with the variety of products, the new technology of plastic molding need to be explored and developed. In the traditional injection molding process, preliminary testing and parameter adjustment are manually performed, and most of the molded products are manually inspected for quality confirmation and defect detection, and the parameters are manually recorded after production. However, most of research are related to appearance, such as dimensions, deformation, or product bonding lines.
However, in the past, the topic of plastic molding has lacked the ability to consider issues other than appearance, such as the lateral force and stiffness of studs, and the use of digital methods to improve process parameters. The research of plastic molding using conventional statistical methods usually requires some assumptions and limitations such as normality, large dataset and not considered the factors into model. Consequently, this study investigated model which are Grey Relation Analysis model, Genetic Programming and Genetic Algorithms. First, the grey relation analysis model get the high relational group. Then use algorithm to create mathematic forms and get parameters. In the experimental section, three plastic molding plants will be used as examples to compare the process capability index and standard deviation with each other using the hybrid optimization method proposed in this study and the unused method, considering the combination of different influencing factors of the experiments, to confirm the value of this method in the application of practical problems. | en_US |