Analysing the efficiency of quantitative measures of importance: the improved FAST
The improved FAST innovative technique is very close to the ideal tool for sensitivity analysis (SA) and may be quite reasonably termed as a global importance measure. Use of the term global is used to underline the contrast with the class of local methods, those based on Taylor expansion, otherwise termed methods for differential sensitivity analysis. A global measure allows each uncertain factor to vary within a non negligible region of interest and to take into account any possible probability distribution. When estimating a sensitivity index for a given factor, in a global measure all the other factors are allowed to vary simultaneously, while in a local analysis they are kept constant at a given central (nominal) value. A global measure is therefore closer to the reality, given that uncertainties affect the factors simultaneously, and not one at a time.
Bibliographic Reference: Paper presented: SAMO 98, Venice (IT), April 19-22, 1998
Availability: Available from (1) as Paper EN 41270 ORA
Record Number: 199810475 / Last updated on: 1998-05-05
Original language: en
Available languages: en