Data-Driven Optimization of Aluminium Blanking Parameters: A Hybrid Taguchi and Regression Approach

The blanking process is an essential manufacturing technique in the sheet metal industry, widely employed for producing precision components. Increasing global market competitiveness necessitates high-quality output with minimal defects. Small-scale industries often struggle to predict optimal punch-die clearance due to limited skilled operators and the high costs associated with Computer-Aided Design/Computer-Aided Manufacturing systems. An accurate simulation of the shearing mechanism in blanking is complex due to the narrow shear band and challenges in fracture criteria. This study successfully employs a hybrid approach combining Taguchi’s Design of Experiments and multiple regression analysis to optimize punch-die clearance and minimize burr height in aluminium blanking. Taguchi’s method identifies key parameters influencing burr height, while regression analysis provides a predictive model for estimating burr height based on varying punch thickness and clearance levels. The findings reveal that a punch thickness of 1.5 mm and a clearance of 5% result in the minimum burr height of 0.0611 mm, with a 95% prediction interval ensuring the robustness of the solution. Contour plot analysis confirms that higher clearance values significantly increase burr height, particularly in thinner sheets, reinforcing the need for precise clearance control. Interaction analysis demonstrates a linear relationship between burr height and clearance across all thickness levels, with the 1.5 mm punch thickness consistently yielding optimal results. This study provides a cost-effective and systematic approach to enhancing blanking performance by integrating statistical optimization and predictive modeling. The proposed methodology offers practical insights for small-scale industries, enabling improved precision, reduced defects, and enhanced efficiency in aluminium sheet blanking.