Quantifizierende Bemerkungen zur Frage der Konservativität der "allgemeinen Berechnungsgrundlagen"
Quantifizierende Bemerkungen zur Frage der Konservativität der "allgemeinen Berechnungsgrundlagen"
Dose prediction models are always subject to uncertainties due to a number of factors including deficiencies in the model structure and uncertainties of the model input parameter values. In lieu of validation experiments the evaluation of these uncertainties is restricted to scientific judgement. Se...
Saved in:
Personal Name(s): | Brenk, H. D. |
---|---|
Vogt, K. J. | |
Contributing Institute: |
Publikationen vor 2000; PRE-2000; Retrocat |
Imprint: |
Jülich
Kernforschungsanlage Jülich, Verlag
1982
|
Physical Description: |
getr. Pag. |
Document Type: |
Report Book |
Research Program: |
ohne Topic |
Series Title: |
Berichte der Kernforschungsanlage Jülich
1821 |
Link: |
OpenAccess OpenAccess |
Publikationsportal JuSER |
Dose prediction models are always subject to uncertainties due to a number of factors including deficiencies in the model structure and uncertainties of the model input parameter values. In lieu of validation experiments the evaluation of these uncertainties is restricted to scientific judgement. Several attempts have been made in the literature to evaluate the uncertainties of the current dose assessment models resulting from uncertainties of the model input parameter values using $\underline{stochastic}$ $\underline{approaches}$. Less attention, however, has been paid to potential sources of systematic over- and underestimations of the predicted doses due to deficiencies in the model structure. The present study addresses this aspect with regard to dose assessment models currently used for regulatory purposes. The influence of a number of basic simplifications and conservative assumptions has been investigated. Our $\underline{systematic}$ $\underline{approach}$ is exemplified by a comparison of doses evaluated on the basis of the regulatory guide model and a more realistic model respectively. This is done for 3 critical exposure pathways. As a result of this comparison it can be concluded that the currently used regulatory-type models include significant safety factors resulting in a systematic o verprediction of dose to man up to two orders of magnitude. For this reason there are some indications that these models usually more than compensate the bulk of the stochastic uncertainties caused by the variability of the input parameter values. |