Sensitivity Analysis and Related Analyses : A Survey of Statistical Techniques

This paper reviews the state of art in five related types of analysis, namely (i) sensitivity or what-if analysis, (ii) uncertainty or risk analysis, (iii) screening, (iv) validation, and (v) optimization. The main question is: when should which type of analysis be applied; which statistical techniques may then be used? To answer this question, the paper distinguishes the following five stages in the analysis of a simulation model. 1) Validation: the availability of data on the real system determines which type of statistical technique to use for validation. 2) Screening: in the simulation's pilot phase the really important inputs can be identified through a novel technique, called sequential bifurcation, which uses aggregation and sequential experimentation. 3) Sensitivity analysis: the really important inputs should be subjected to a more detailed analysis, which includes interactions between these inputs; relevant statistical techniques are design of experiments (DOE) and regression analysis. 4) Uncertainty analysis: the important environmental inputs may have values that are not precisely known, so the uncertainties of the model outputs that result from the uncertainties in these model inputs should be quantified; relevant techniques are the Monte Carlo method and Latin hypercube sampling. 5) Optimization: the controllable inputs should be steered; a relevant technique is Response Surface Methodology (RSM), which combines DOE, regression analysis, and steepest-ascent hill-climbing. This approach with its five stages implies that sensitivityanalysis should precede uncertainty analysis. This paper briefly discusses several case studies for each phase