Multi-Objective Optimization of Injection Molding Using Taguchi, Fuzzy Methods, and GA

The objective of this research is to optimize the injection molding process of an automotive window regulator bracket by improving the moldability index while minimizing key defects. To achieve this, a multi-objective framework is developed that combines the Taguchi method with Fuzzy Analytic Hierarchy Process (FAHP) and Fuzzy-TOPSIS. Five critical processing parameters—melt temperature, mold temperature, filling time, holding pressure time, and cooling time—were investigated, with polypropylene as the base material. A Taguchi L25 orthogonal array was employed to reduce the number of experimental trials from 3,125 to just 25, thereby saving resources while maintaining reliability. The evaluation considered warpage, residual stress, and shear stress, which are the most influential defects affecting part performance. Finite Element Analysis (FEA) was incorporated to validate the accuracy of the results, while a hybrid ANFIS-GA predictive model was applied to forecast the moldability index, demonstrating an improvement of about 1% over conventional optimization methods. The optimized settings resulted in minimized warpage (1.8122 mm), residual stress (43.03 MPa), and shear stress (0.08 MPa). The novelty of this work lies in integrating Taguchi with FAHP and Fuzzy-TOPSIS for a single-objective transformation, offering a systematic and efficient approach for multi-objective optimization in injection molding applications.