Considering the prediction of thermal machine tool error data this paper proposes using GM (1,1) model of gray system to predict machine tool thermal error data. Through comparing with the calculated result between traditional GM(1,1) model and the optimized GM(1,1) model, more satisfactory results can be achieved, and indicating that this model can predict the distribution of machine tool thermal error data more accurately. So the optimized GM (1,1) model can be used to provide a more reliable methodology in controlling the digital machine tool.