Thermo-elastic effects are one of the major reasons for positioning errors in machine tools. Next to friction and waste heat from drives, the heat exchange with the machine's surroundings influences the temperature field inside the machine tool significantly. The thermal parameters necessary to describe this heat transfer can be obtained through computational fluid dynamics (CFD) simulations. This paper presents a new method aimed at decoupling these CFD simulations from the thermo-elastic simulations in order to provide the heat transfer parameters quickly and efficiently for transient environmental conditions. This is done by defining a suitable set of load scenarios for the CFD simulations, clustering the resulting parameters with radial basis functions and interpolating them using characteristic diagrams.