ETAS ASCMO-STATIC supports a variety of use cases, offers a wide range of functions including appropriate visualizations and model-evaluation tools, and provides powerful optimization algorithms. It enables users to carry out data-based modeling and run model-based applications quickly and easily.


Display of the behavior of a direct-injection gasoline engine

Black lines: Model prediction.
Red lines: Model accuracy.
Y-axis, top-down: Prediction of the output parameters “fuel consumption”,” running smoothness”, “soot”, and “NOx emissions”.
X-axis, left to right: Interactive adjustment of torque and load-point and of the relevant calibration parameters (blue lines).

Automated optimization of characteristic maps. The optimization criteria in this example are minimum fuel consumption, as well as compliance to emissions and running smoothness limits.

ETAS ASCMO-STATIC models make it possible, for instance, to automatically or manually optimize and accurately predict fuel consumption and emission values for complex combustion engines according to engine speed, load, or any controlled engine variable. Based on the progression of the manipulated variables, the relevant engine control characteristic maps can then be precalibrated to achieve the best compromise between emissions and fuel consumption during engine operation.

Design of experiments (DoE)

Design of experiments (DoE), an essential element of ETAS ASCMO-STATIC, enables maximum model accuracy to be achieved with minimal measuring effort.

ETAS ASCMO-STATIC models are parameterized on the basis of measurement results collected from the test bench in a real system environment. The DoE test planning module allows the underlying measurements to be planned in a simple manner, and DoE methodology is used to prepare the number and positions of points at which measurements are taken. The variation range of the measuring points can be reduced in up to three dimensions graphically or numerically, and can be limited in any dimension using a formula editor. Test automation requirements, such as the setting of operation points, are taken into account in a practice-oriented manner. Intelligent splitting of the designed experiment into blocks makes it possible to adjust the measurement effort to the desired degree of accuracy. ETAS ASCMO’s design of experiments module selects the position of measuring points for creating models in such a way that maximum model accuracy is achieved with minimum measurement effort.


Using measurement data based on Gaussian processes, the statistical learning procedure automatically determines the specific mathematical function that best maps the real system behavior. To avoid overfitting, measurement noise is implicitly taken into account. The method is robust to measurement outliers. A spatially resolved model variance gives the user a measure of the reliability of the model prediction. The model quality is indicated by easy-to-understand graphs and figures.

Visualization and systems analysis

Interactive intersection plots enable multidimensional dependencies to be simply displayed, which allows easy analysis of the influence and interaction of input variables. With the powerful ETAS ASCMO-STATIC optimization algorithms, users can resolve calibration target conflicts and reduce fuel consumption and emissions in the driving cycle.

Optimization in multiple dimensions

To optimize the input variables that control an engine, for example, it is possible to define different criteria such as minimization/maximization, upper/lower limits, and target values for output variables. During optimization, either these criteria can be weighted or the full trade-off (or Pareto) curve can be calculated. Users can select the appropriate compromise from the trade-off curve. For engine calibration, it is possible to carry out global optimization across the entire operating range. As a result, optimum calibration of the characteristic maps related to the input variables is achieved in no time. The characteristic maps can also be changed manually. Here, the user is always able to directly observe the impact of changes they have made on the output values. This is true both for the individual output value at the current operating point and for the sum value relative to the entire driving cycle.


Analogous to ETAS INCA, a working and reference page concept makes it possible to compare alternative calibrations. The results of the model-based calibration can be exported in the DCM or CSV export format and read into INCA.

Virtual test bench

Hundreds of thousands of measurements can be generated artificially from an ETAS ASCMO-STATIC model within a few seconds. For this purpose, any grid size for the input variables can be defined in terms of step size, step number, and free nodes directly in the tool or by importing an Excel list. The measurement data generated by the model can be used just like real test bench measurements, analyzed graphically, and exported as a CSV file for further processing in Excel.

Open and flexible

ETAS ASCMO-STATIC is open and flexible. The tool supports all data formats relevant to test bench and calibration functions. ETAS ASCMO-STATIC is capable of exporting existing models in a variety of formats. With the aid of the MATLAB® interface, customer-specific functions and models can be easily integrated and script-based sequences automated. It is also possible to connect a test bench automation system.