Efficient Optimization of the Multi-Response Problem in the Taguchi Method Through Advanced Data Envelopment Analysis (DEA) Formulations Integration

This research presents a novel approach for effectively addressing the multi-response problem in the Taguchi Method, a pervasive issue in the production process and productivity index improvement efforts. We achieved this by integrating the Taguchi Method with DEA, transforming experimental trials into Decision Making Units (DMUs) and response variables into inputs and outputs. Subsequently, advanced DEA formulations, like simultaneous DEA and super-efficiency models, were employed to compute the relative and cross efficiencies of the DMUs, leading to a specific combination of multiple control factors at different levels. This method preserves the benefits of the Taguchi Design of Experiments (DoE) while providing an optimal solution for improving multiple quality responses. The Addictive Model for Factor Effects confirmed that our proposed combination delivers superior improvement results when compared to empirical and previous studies' solutions, while also reducing computational effort. The study's findings offer significant implications for advanced manufacturing practices, especially those requiring multi-response optimization