Tool wear prediction during machining is of significant interest for the development of intelligent functionalities in the manufacturing industry. A data driven Bayesian Monte-Carlo based probabilistic approach is used for predicting the wear of TiAlN coated carbide inserts during the end milling of Ti-6Al-4V alloy. A series of slot milling passes at varying combinations of speed, feed and depth of cut were conducted and wear was measured after each pass. Each insert was used for successive passes at a particular cutting condition until the flank wear crosses the failure threshold of 0.3 mm of average flank wear. The wear estimation from the model is good at tracking wear growth for the unknown data sets which can provide a timely tool change command before the tool failure. This model thus leads to the formulation of an adaptive control strategy for timely replacement of cutting tools for optimal machining.