In injection molding of rubbers, the temperature of the dosed rubber compound is crucial in order to achieve stable processes. Temperature peaks during this process must be minimized to prevent a start of the crosslinking reaction at a too early stage. Within the scope of this work, the complex dosing process is scientifically examined. Based on experimental findings, a process model for calculating the transient mass temperature of the dosed rubber compound is developed. In order to evaluate the process model, a test stand equipped with sensors was manufactured from the components of the injection unit of a rubber injection molding machine. As a result, it was possible to investigate the transient mass temperature as well as the flow behavior in the screw channel in a manner very close to the actual process. The key influences on the resulting mass temperature level, the temperature setting of the plasticizing cylinder, followed by screw speed, and back pressure could be examined. Thus, practice-oriented process parameters led to temperature rises of about 20 K during the dosing of a volume of 150 cm³, compared to the target setting. This rise in mass temperature is not negligible for chemically active systems. For the mathematical description of the flow conditions in the screw channel, it was essential to verify whether cross-flows occur during the transient dosing process or not. Visualization studies carried out with two-colour rubber compounds have shown that cross-flows are present in the screw channel. Accurate characterization of industrial rubber compounds provided the material data required for the numerical calculation of the mass temperature. Moreover, four test methods for the characterization of thermal conductivity were compared. Differences of up to 100 % in the determined values were detected. As a result, the stationary “Guarded Heat Flow Meter-method” was considered to be the most suitable method. Both experimentally determined and calculated mass temperature profiles were introduced and compared. With an appropriate selection of transient boundary conditions, deviations between calculated and measured mass temperatures of less than 6 K could be achieved. Effects of different mass temperatures on part quality were investigated. However, minor influences on the achieved part quality was detected. Based on the scientific elaboration of the dosing process, the complete rubber injection molding process can now be simulated without restrictions.