The detection and control of the temperature variation at the nano-scale level of thermo-mechanical materials during a compression process have been challenging issues. In this paper, an empirical method is proposed to predict the temperature at the nano-scale level during the solid-state phase transition phenomenon in NiTi shape memory alloys. Isothermal data was used as a reference to determine the temperature change at different loading rates. The temperature of the phase transformed zone underneath the tip increased by ∼3 to 40 °C as the loading rate increased. The temperature approached a constant with further increase in indentation depth. A few layers of graphene were used to enhance the cooling process at different loading rates. Due to the presence of graphene layers the temperature beneath the tip decreased by a further ∼3 to 10 °C depending on the loading rate. Compared with highly polished NiTi, deeper indentation depths were also observed during the solid-state phase transition, especially at the rate dependent zones. Larger superelastic deformations confirmed that the latent heat transfer through the deposited graphene layers allowed a larger phase transition volume and, therefore, more stress relaxation and penetration depth.