Design of Experiments (DoE)

The goal of the design of experiments (DoE) is the model-like description of unknown systems based on measuring data. The methodology of the DoE includes the creation of the experiment plan according to statistical aspects, the creation of models and the optimization of modeled systems.

The experiment planning is performed with the purpose of minimizing the measuring effort – for the design of experiments, several variables are modified for each measurement. For the model training, mathematical approximation methods are used.

Fig. 52: Model training with data from the DoE plan

For example, one application is the description of the behavior of an internal combustion engine by a model. The maps of the ECU are not being optimized using the real system, but a model.