The increasing demand for precise and efficientCondition monitoring condition monitoringMonitoring of milling toolsMilling tool in operation demands the development of advanced diagnostic methods. Accurate fault identification, for rake and flank wear in milling tool is crucial for optimizing performance and minimizing downtime. This manuscript presents a novel approach for automated fault identification in milling toolsMilling tool specifically rake and flank wear using vibration signals acquired with the help of triaxial accelerometer during operation. Vibration signals are chosen as they provide sensitive, non-intrusive, and comprehensive information about dynamic behaviour. These signals are further decomposed usingTunable Q-wavelet transform Tunable Q-wavelet transform into different sub-bands. Followed to this, eleven statistical features are extracted from each sub-bands of the signal. Few of them includes mean, standard deviation, kurtosis, skewness etc. Subsequently, supervisedMachine learning machine learning classifiers, Artificial Neural Network (ANN) is trained to classify the working condition of the milling toolMilling tool either as healthy, rake or flank wear. The study shows that ANN shows 100% accuracy in identifying the condition of the tool.