Design methodology of plastic injection process using approximate solution techniques

In this study, Moldflow analysis will be performed in accordance with the set of design of experiment by using statistical experimental design methods for optimization of injection parameters. In order to evaluate the results, firstly a proper regression model will be created by making statistical analysis by means of Mini Tab software.Variance analysis (Anova) will be performed to determine the F test of the accuracy of the model coefficients and the effects of the parameters. Secondly, the regression model will be defined as the fitting function and the optimum parameters will be worked on to be found by using the Genetic Algorithm with the constraints of the injection process. Thirdly, the process will be optimized by creating variable based on a hypothesis by means of Monte Carlo simulation. Monte Carlo simulation is used to make calculations based on assumption in case of indeterminating propagation in traditional analytical models and analytical techniques do not work properly. In this study, the process parameters such as mould temperature, cooling water temperature, injection pressure are never constant value. To remove this uncertainty, the process will be optimized by generating the variable based on hypothesis. It will be evaluated which method gives more accurate results by comparing to results being obtained by different optimization method in the final phase. Keywords: Injection parameters, Desing of experiment , Genetic alghorithm