The thermal error of CNC machine tool spindle is one of the main factors affecting the machining accuracy of the machine tool.The optimization of the spindle thermal error temperature measurement point is of great significance for accurately establishing the thermal error model of the machine tool spindle and improving the precision of the machine tool.A method for optimizing the temperature measurement point of machine tools was proposed based on the combination of fuzzy clustering and grey theory.By fuzzy clustering analysis of the spindle temperature measurement points,the temperature points were classified into several categories according to the Xie-Beni effectiveness index.Then,the gray correlation analysis between the temperature measurement point and the spindle thermal error after fuzzy clustering was carried out,and the temperature measurement point of the machine tool spindle was further optimized.The feasibility and effectiveness of the method are verified by experiments.