<p>In the present work, a hybrid model including the machine learning and theoretical approach has been developed to study a trade-off between the life time of the machined material and energy consumption of ultrasonic assisted milling (UAM) process in cutting of Inconel 718. To do so, firstly, a series of experiments were carried out using full factorial experimental design, taking into account process’ main factors viz. ultrasonic vibration amplitude, feed per tooth, and cutting velocity as input, and machinability indicators viz. cutting forces, roughness, hardness, and surface residual stress as responses. Then, an adaptive neuro-fuzzy inference system (ANFIS) was employed to correlate between the input and responses through a cross-validation approach using five different folds of training and testing. After finding the most accurate ANFIS model of machinability indicators, they were used as input to predict the energy consumption and fatigue Life of machined components using theoretical principles of machining process and macromechanical stress approach, respectively. The developed hybrid model has been verified through comparing the calculated values of fatigue with experimental fatigue Life values of 12 samples, which were machined under different UAM conditions and applied fatigue loads using plane-bending fatigue test. It was found that the developed model is accurate enough based on previous studies, where the mean absolute percentage error is 13% for those 12 sets of experiments. It was also found that the application of ultrasonic vibration significantly meets the sustainability by improving the Life cycle of the material up to 300% and energy efficiency of more than 40%.</p>