Integration of ANN and ANFIS Models to Predict Quality, Cost and Energy During Machining Alloy 2017A

The global demand for energy is experiencing rapid growth, propelled by factors such as population expansion and economic advancements, particularly in emerging market economies. However, amidst this greater prosperity, the escalating demand for energy presents novel challenges. This study delves into the scientific insights concerning the impact of strategies, machining sequences, and cut-ting parameters on surface quality, machining cost, and energy consumption (QCE) by leveraging both neural network (ANN) and ANFIS models. Research findings underscore the significant influence of machining strategies and sequences on energy consumption and cost. Notably, the {3–10-3} architecture, coupled with the Bayesian regularization (BR) algorithm, emerges as the optimal neural architecture, yielding a global mean square error (MSE) of 2.74 × 10–3. Furthermore, the results highlight the efficacy of the Bayesian regularization (BR) algorithm in conjunction with a multi-criteria output response, showcasing its favorable outcomes compared to the adaptive neuro-fuzzy inference system (ANFIS).