Design of Experiments (DoE) is the unique approach that is able to describe the physical relationship between input and output variables of technical systems, when all analytical approaches fail. The description is done by using a model (transfer function), which is parameterized based on experimental data. However, DoE requires normally distributed residuals. In case of lifetime tests, the residuals are not normally distributed. Therefore, the use of the DoE methodology is not valid. In order to counteract the restrictions of DoE, Lifetime-DoE (L-DoE) was developed. Four different concepts (Box-Cox (BC)) approach, Generalized Linear Logarithmic model, Proportional Hazard model and Lifetime Regression) are the core of L-DoE. In order to investigate the performance of the modelling concepts, a simulation study is carried out. For this purpose, Weibull distributed data are generated by a Monte Carlo (MC) simulation using given parameters. The effects are estimated and are then used to develop a reliability model. Finally, this model is compared with the reliability function obtained for the given parameters. For statistical validation of the results, a 95%-MC confidence interval for the reliability function is determined. While with the last three approaches the given effects are estimated correctly, reliability modelling with Box-Cox is associated with an inaccuracy. This is due to the fact that the effects are blurred when the data is transformed.