Parameter analysis of multi-objective optimization for energy efficiency and multiple quality aspects in injection molding

Injection molding is widely used to produce various components and products. An attempt to achieve a certain quality necessitates sacrificing other qualities and consuming excessive energy, causing significant carbon emissions. Although many studies have been conducted on optimizing injection molding parameters, they typically focus on optimizing a single quality attribute. Multi-objective optimization is the solution to address all relevant qualities and energy consumption simultaneously. Taguchi analysis can achieve single-objective optimization with fewer experiments and simple calculations. It provides an efficient way to optimize specific quality attributes. In contrast, response surface methodology (RSM) explores and optimizes the complex relationships within multivariable systems. RSM establishes a mathematical model for optimizing multivariable systems. However, RSM requires more experiments and involves more complex calculations. Single-variable experiments and Taguchi analysis were first used to address energy consumption and identify the key factors. This study focused on plastic screws, aiming not only to explore energy consumption but also to optimize the quality in terms of diameter, weight, and tensile strength. To construct mathematical models, this study applied Taguchi analysis and RSM. Using the desirability function, the RSM model was optimized, and the applicability of both approaches in multi-objective optimization for balancing energy efficiency and product quality was evaluated. The research results indicate that holding and cooling time significantly impact energy consumption. Both the Taguchi method and RSM optimization results show that the optimal combination is a melt temperature of 230 degrees C, holding pressure of 40 MPa, holding time of 3 s, and cooling time of 3 s. This optimal combination can reduce energy consumption by 29 similar to 33% while maintaining a certain level of product quality. However, compared to the Taguchi method, which relies on subjective judgment to select the optimal conditions, RSM can achieve more precise and reliable optimization results through regression analysis and model evaluation.