In this thesis the production process of press hardening in ski production is investigated in order to find out if there is an economic and efficient way to improve the effort and cost. The current process uses oil in order to heat up moulds containing the layers of a ski combined with an adhesive in between. The layered and glued ski is pressed, heated and, after a certain holding period, cooled down again. The energy for heating and holding is provided by hot oil while the cooling happens with water pumped through the heating elements which results in the need of two separate cycles. The goal of this thesis is to build up a FEM model which is able to simulate this process by analyzing the current process and the methods of how the heat transfer takes place in combination with measurements of the temperature progression along the heating elements as well as measurements of both the inlet temperatures and outlet temperatures. In addition to that the FEM model must be able to simulate an alternative process with the use of oil instead of water during the cooling phase and therefore lay the base for giving a recommendation whether an adaption of the process in this way could be possible or not. The FEM model is created with ABAQUS® while the 3D-model is built out of a combination of already existing and new models. An uncoupled heat transfer analysis is used to calculate the process. The heat transfer coefficient is controlled and calculated during the simulation by the use of a subroutine implemented in ABAQUS®. The simulation results are then compared with the measurements of the real process and the simulation is calibrated until the accuracy is sufficient. Using this FEM model, the material parameters of the fluid during the cooling phase are changed within the subroutine to calculate the expected results for the use of oil instead of water for this step. This new cooling process is simulated in two ways. With the use of the current oil and varying volume flow as well as additional cooling time and with alternative oils, which have a lower viscosity, and increased volume flow. The results indicate that increasing the volume flow for the current oil has little effect in the range that could technically be achieved as with a certain viscosity the flow profile remains laminar and therefore the heat transfer is very low and takes a lot of time. However, by using a less viscous oil it is possible to remain turbulent even at room temperatures and thus, the cooling is sufficient to achieve the necessary end temperature in the same time as with water as a coolant. Finally, the simulation results show that there are two ways to achieve the necessary cooling temperature with oil as a coolant. The first is to increase the cooling time, which is not very efficient, while the second is the use of a low viscosity oil.