Unbiased guess : A concept to cope with fuzzy and random parameters?Funded under: FP2-RADWASTOM 4C
This work is concerned with the problem of making assertions on propositions which contain different kinds of uncertainty. It is shown how the uncertainties coming from statistical data, expert knowledge and linguistic inaccuracies can be represented by probability densities, possibility distribution functions and membership functions, which define random variables, fuzzy variables and fuzzy sets. Based on these representations, two suggestions have been worked out to handle the problem of mixed uncertainties, which do not allow a purely probabilistic or possibilistic treatment. The most important result of these investigations is a sequence of inequalities between formal probabilities, possibilities, necessities and some mixed measures. The inequalities are a consequence of a probability/possibility transformation derived within this project, which maps probability densities to possibility distribution functions and vice versa. Initial implementation of the corresponding numerical algorithms on a computer and the application to a very simple hypothetical geochemical transport problem are reported.
Bibliographic Reference: EUR 16199 EN (1995) 42 pp., FS, ECU 7
ISBN: ISBN 92-827-0154-9
Record Number: 199510499 / Last updated on: 1995-03-22
Original language: en
Available languages: en