Temperature on the machined surface is critical for surface integrity and the performance of a precision component. However, the temperature of a machined surface is challenging for in-situ measurement. Furthermore, the individual contribution of tool/work friction and plastic deformation of work materials to surface temperature is very difficult to quantify because the measured temperature is always the resultant temperature. This lack of understanding on the temperature distribution blocks the design of effective cutting tool geometries and materials to minimize surface temperature. This study provides a finite element method based on a predictive model to decouple the contributions of tool/work friction and material plastic deformation to surface temperature in a dry cutting process. The study shows that the plastic deformation of work material contributes to the majority of surface temperature, whereas the tool/work friction contribution is secondary. High temperatures are produced when more materials are plowed under the cutting edge. A large tool/work friction leads to higher surface temperatures, and the use of a cutting tool with physical properties in process simulation significantly improves the accuracy of predicted surface temperatures. Residual stress reversal from subsurface maximum residual to surface maximum residual stress may occur when tool/work friction increases.