Community Research and Development Information Service - CORDIS


When taking decisions concerning uncertainties, making use of the output from mathematical or computational models, the decision maker might be helped if the uncertainty in model predictions is broken down in a quantitative, rather than qualitative fashion, apportioning uncertainty according to source. For complex models, such a break down of the uncertainty into constituent elements could be impractical as such, due to the large number of parameters involved. If instead parameters could be grouped into logical subsets, then the analysis could be more useful, also because the decision maker may have different perceptions (and degree of acceptance) for different kinds of uncertainty. One might imagine that a partition of the uncertainty could be sought between stochastic (or aleatory) and subjective (or epistemic) uncertainty. The research shows how to compute rigorous break down of the output's variance with grouped parameters, and how this approach may be beneficial for the efficiency and transparency of the analysis.

Additional information

Authors: SALTELLI A, JRC Ispra (IT);TARANTOLA S, JRC Ispra (IT);CHAN K, JRC Ispra (IT)
Bibliographic Reference: Article: Risk Analysis
Record Number: 199711706 / Last updated on: 1998-01-20
Original language: en
Available languages: en