2.6 Need for Alternative Sensitivity/Uncertainty Analysis Methods

Traditional sampling methods for sensitivity and uncertainty analysis, such as the Monte Carlo and Latin Hypercube Sampling, require a substantial number of model runs to obtain a good approximation of the output pdfs, especially for cases involving a several inputs. On the other hand, analytical methods require the information about the mathematical equations of a model, and often are restricted in their applicability to cases where the uncertainties are small. Therefore there is a need for a computationally efficient method for uncertainty propagation that is robust and also applicable to a wide range of complex models. The following chapter describes the development of the Stochastic Response Surface Method (SRSM), which is a computationally efficient uncertainty analysis method developed as part of this work.