Sebastian K. Crönert presents a new, automated process that makes it possible to obtain all the fuel properties required for combustion simulation. If necessary, these are then transferred - also automatically - into specially created correlation equations through which they are then made available again at simulation runtime. This method makes it possible to represent even more complex correlations and cross-influences on calculation variables in a resource-optimised way (memory requirements and access time) while maintaining the same accuracy. The procedure is validated using test bench measurement data for the pure fuels anisole and cyclopentanone and their blends with regular petrol (RON95E10). Additional validations include more established synthetic fuels and hydrogen. It is shown that an extraordinarily high prediction quality can be achieved for the model class.