Community Research and Development Information Service - CORDIS


A sampling scheme is described that is used to perform sensitivity analysis (SA) of model output. In general, SA is conducted by the following steps:

- Defining a model which approximates economic, engineering, environmental, physical or social phenomenon of various levels of complexity, and its input parameters and output variable(s);
- Assigning probability distributions to the input parameters;
- Generating input values via an appropriate random sampling method and evaluating the output;
- Assessing the influence or relative importance of each input parameter on the output variable.

The paper is devoted to the third and fourth steps. Here, it is assumed that a model output, Y, is depended on k input parameters, say x(1), x(2), ... x(k), namely
y=f(x(1), x(2), ..... x(k)). x(k)).

Additional information

Authors: HONTAÑÓN J, CIEMAT, Dpto de Fusión Nuclear, Madrid (ES);ARIAS R, CIEMAT, Dpto de Fusión Nuclear, Madrid (ES);DE LOS REYES A, CIEMAT, Dpto de Fusión Nuclear, Madrid (ES);LOPEZ-JIMENEZ J, CIEMAT, Dpto de Fusión Nuclear, Madrid (ES);AREIA-CAPITAO J, JRC Ispra (IT);DE SANTI G, JRC Ispra (IT);DE LOS REYES A, Consejo de Saguridad, Madrid (ES);HONTA?ON J, CIEMAT, Dpto de Fusion Nuclear, Madrid (ES)
Bibliographic Reference: Paper presented: TOP SAFE '98, Valencia (ES), April 15-17, 1998
Availability: Available from (1) as Paper EN 41267 ORA
Record Number: 199810472 / Last updated on: 1998-05-05
Original language: en
Available languages: en