In the quasi-dimensional modeling of the spark-ignition combustion process, the burn rate calculation depends, among other influences, on the laminar flame speed. Commonly used models of laminar flame speeds are usually developed on the basis of measurement data limited to boundary conditions outside of the engine operation range. This limitation is caused by flame instabilities and forces flame speed models to be extrapolated for the application in combustion process simulation. However, for the investigation of, for example, lean burn engine concepts, reliable flame speed values are needed to improve the quality and predictive ability of burn rate models. For this purpose, a reference fuel for gasoline is defined to perform reaction kinetics calculations of laminar flame speeds for a wide range of boundary conditions. In order to define a reference fuel as representative as possible for standard gasoline, the influence of octane number and hydrogen-to-carbon ratio on the laminar flame speed of a toluene reference fuel (TRF) is investigated. Furthermore, the necessity to use a TRF instead of a primary reference fuel is shown. The reaction kinetics calculation results are then investigated to give possible explanations for the influence of boundary conditions like temperature, pressure, fuel composition, residual exhaust gas and air-fuel ratio on the laminar flame speed. Additionally, they are used to identify and illustrate the shortcomings of the widely used Heywood-approach, which belongs to the extrapolating flame speed models mentioned above. In a next step, the calibration parameters of an existing, more advanced model are adapted to match the flame speeds calculated for the TRF. This model is then expanded to cover the admixture of ethanol and finally employed in engine simulation to exemplarily show the prediction of burn rate changes due to a variation of exhaust gas recirculation rate or air-fuel equivalence ratio.