A role for sensitivity analysis in presenting the results from MCDA studies to decision makers

Abstract

The aim of sensitivity analysis (SA) is to ascertain how much the uncertainty in the output of a model is influenced by the uncertainty in its input factors. An SA can be performed using different methods, which are classified according to various criteria. One possible classification is that in which global and local approaches are identified. This paper strengthens the role of global SA methods and suggests their use in the context of MCDA. Useful applications of global SA already exist in a variety of fields where numerical models are considered, e.g. in economics, engineering and chemistry. Global SA is quite different in its formulation and application from local SA, which is seen more frequently in the literature. The global sensitivity indices adopted are derived from a variance decomposition scheme: they can be estimated by alternative computational strategies, such as the extended FAST or the method proposed by Sobol'. A truly global SA is capable of apportioning the output uncertainty according to any subgroup of input factors. Hence, the output uncertainty of an MCDA can, for instance, be decomposed into a part due to uncertain model inputs and a part due to poorly defined (or variable) weights attached to the criteria. This information could be useful to the decision maker (DM) since it explains synthetically how much the assessment of an MCDA study is biased by the assessor judgements. Alternative regrouping of the uncertain input elements might shed light on other features of the problem addressed by the DM.

A decision analytic model to guide early‐stage government regulatory action: Applications for synthetic biology

“…However, the secure labs only (0.64), IRB pre‐research (0.61), and ERB post‐research (0.59) alternatives also performed well. We will further discuss quantitative reasons why the alternatives score as they do, and perform sensitivity analysis to determine which conditions the existing optimality may shift (Saltelli et al ). Such an assessment allows for an understanding of how resilient and dependable the selection of ERB pre‐research is as the highest scoring governance option for SB cyberplasm research and development (Saltelli et al ; Stewart ).…”

“…Specifically, MCDA integrates qualitative and quantitative information alike, allowing experts to voice their beliefs and opinions on a subject with limited field data or formal scientific agreement (Linkov & Moberg ; Linkov et al ). Further, MCDA tools can be used to review any disagreements or uncertainties signaled by subject experts through the use of sensitivity analysis, making MCDA all the more helpful for regulators that have to address shifting political, economic, social, and environmental factors as more information about specific products becomes available (Saltelli et al ; Linkov et al ). Given these factors, MCDA serves as an appropriate method to analyze information derived from the Delphi process and rank governance options for cyberplasm based upon relevant decision criteria.…”

Natural Hazard Susceptibility Assessment for Road Planning Using Spatial Multi-Criteria Analysis

“…A comparison of the expert scoring in AHP was undertaken. The results of the comparison explained how much the assessment of the SMCA study was influenced by the assessor judgments, which was reflected by the weights considered (Saltelli et al 1999 ). In this study the comparison was undertaken by aggregating and dividing the expert scoring into the three weighting sets AHP Q1, Q2, and Q3 (Tables 5 – 7 ).…”

Use of multi-criteria decision analysis to identify potentially dangerous glacial lakes

“…In MCDA, sensitivity analysis serves to determine how much the uncertainty of the results of a model are influenced by the uncertainty of its input criteria (Saltelli et al, 1999). Sensitivity analysis can be performed using different methods.…”