Robust Design Optimization
Traditional optimization techniques tend to "over-optimize", producing solutions that perform well at the design point but may have poor off-design characteristics. It is important, therefore, that the designer ensures robustness of the solution, defined as the system’s insensitivity to any variation of the input parameters. It is quite possible that the optimal solution will not be the most stable solution.
In the example shown here there is a global optimum at point A, and a local optimum at point B. However, any small variations in the input parameters will cause the performance to drop off markedly around A. The performance of B may not be as good in absolute terms, but it is much more robust, since small changes in the input values do not cause drastic performance degradation.
modeFRONTIER allows the user to perform a robust design analysis to check on the system’s sensitivity to manufacturing tolerances or small changes in operating conditions through its MORDO (Multiobjective Robust Design Optimization) module.