UNIVERSITY OF BUCHAREST
FACULTY OF PHYSICS

Guest
2026-06-11 23:58
 HOME     CONFERENCES     SEARCH            LOGIN     NEW USER     IMAGES   


Conference: Bucharest University Faculty of Physics 2026 Meeting


Section: Nuclear and Elementary Particles Physics


Title:
Simple predictive models for evaluating the uncertainty associated with computed values of the full-energy peak efficiency provided by Monte Carlo simulation codes for high-purity germanium detectors


Authors:
D. GURAU, D. STANGA, E. IONESCU, L. DONE


Affiliation:
National Institute of R&D for Physics and Nuclear Engineering-Horia Hulubei (IFIN-HH)-Romania


E-mail
doru@nipne.ro


Keywords:
gamma spectrometry, Monte Carlo simulation, model calibration, uncertainty evaluation


Abstract:
It is important to estimate not only the true value of the computed FEP efficiency but also its standard uncertainty (JCGM-100, 2008). In this work three simple predictive models are described, which are based on a multiplicative model where the true value of the FEP efficiency is given by the product between a correction factor and the computed value of the FEP efficiency. Geometric and arithmetic mean models make use of the least squares estimator for predicting the correction factor. The third model takes a priori the predicted value of the efficiency correction factor equals one because this value is commonly used in practice. For geometric and arithmetic mean models, the prediction uncertainty is calculated via the variance of prediction errors while for the third model, it is calculated via the root of the mean squared prediction errors, which takes into account the bias errors. The predictive models can easily be applied in two steps: (i) generating of the dataset {x_i,w_i }_(i=1)^n and (ii) predicting the value of the efficiency correction factor and the associated uncertainty for a future computation of the FEP efficiency relating to a new vector x_new. The measured correction factors w_i are given by the ratio between measured and computed FEP efficiency (for optimal values of the detector parameters). In the second step, the predicted value of the efficiency correction factor and the associated uncertainty for a new vector x_new are calculated using one of the predictive models. These quantities are constant for all vectors belonging to a given target population. Once these values are available, the user of the MC simulation code can make use of them for estimating the true value of the FEP efficiency and the associated uncertainty for the new vector x_new. The predictive models were applied to two datasets taken from the literature.


References:

D. Gurau, D. Stanga*, O. Sima L. Done. Evaluating the uncertainty associated with computed values of the full-energy peak efficiency used in gamma spectrometry measurements, Appl. Radiat. Isot., 230 (2026).

D. Gurau, D. Stanga*, L. Done, O. Sima, G. Ilie. Calibrating GESPECOR model of computing the full-energy peak efficiency of coaxial high-purity germanium detectors by Monte Carlo simulation. Appl. Radiat. Isot., 204 (2024)

Acknowledgement:
This work was supported by NUCLEU Program within the National Research Development and Innovation Plan 2022–2027, carried out with the support of Management Center of Research, Development and Innovation (CMCDI), project code PN 23-21.