optiSLang (Adv. Optimization)

MOP/CoP

Kopfgrafik_OSL_eng_01

Metamodel of Optimal Prognosis (MOP) / Coefficient of Prognosis (CoP)

cop_24pt-01_01
CoP Chart
MOP Chart

Engineering tasks are affected by many parameters. With optiSLang and the MOP/CoP-methodology, you can exploit the full optimization potential without limiting to a few paramaters (CAE + CAD).

Running a sensitivity analysis, optiSLang identifies automatically the most important parameters and finds the best related model (metamodel) with the relevant input and output variables. Within the sensitivity module, the Coefficient of Prognosis (CoP) allows you to filter the relevant input parameters. According to the prognosis capability of the resulting values, optimal meta-models will be selected. These Metamodels of Optimal Prognosis (MOP) represent the most important correlations between parameter input variation and output results. They can be used as surrogate models for CAE-calculations in optimization procedures or robustness analyses. A predictable forecast quality of the simulation model is the key to efficiency. Thus, a “no run too much” philosophy can be implemented to minimize solver calls.

More information:

Sensitivity analysis using the MOP (Paper)

Recent advances in MOP (Paper)

Parameter identification with optiSLang and MOP (Webinar Video Tutorial)