In many industrial processes, the resulting mechanical properties of produced steel parts are directly influenced by the thermo-physical properties, which affect the heat treatment significantly. The quality of application-oriented simulations is strongly dependent on the input quantities, which are often generated by regression analysis or simple extrapolations. The aim of this paper is to demonstrate the influence of the thermo-physical properties on such a process simulation referring to the hot stamping. Hot stamping is an established process in the automotive industry to produce ultra-high strength parts. A typical material used for this application is the low-alloyed steel 22MnB5. The thermal conductivity of this steel was investigated referring to the temperature-dependent microstructural changes during the hot-stamping process, particularly the gamma to alpha' transformation. In terms of the dynamic measuring method, the specific heat capacity, the thermal expansion coefficient, the density and the thermal diffusivity for the different temperature-dependent microstructures of the steel 22MnB5 were determined. The thermal conductivity for the complete temperature range of the hot-stamping process was generated, referring to measured and extrapolated data. To account for the fast gamma-I +/-' transformation kinetics, a novel characterization and extrapolation method was applied. The heat capacity and the thermal diffusivity have a major impact on the thermal conductivity compared to the subordinated influence of the density. The metastable austenitic condition (T a parts per thousand yen 900 A degrees C) was compared to the martensitic condition (T a parts per thousand currency sign 400 A degrees C). The dependent thermal conductivity is significantly dependent on the crystallographic orientation of the lattice. The face-centred cubic lattice (austenite) has referring to the body-centred cubic lattice (martensite), a proportionally low thermal conductivity. During the transformation from austenite to martensite, the development is not linear but based on complex interactions. The results reveal that the temperature-dependent thermal conductivity has to be considered for reliable process simulations.