SYMPOSIA PAPER Published: 14 December 2018

Assessment of Novel Techniques for Nuclear Data Evaluation


The quality of evaluated nuclear data can be impacted by, for example, the choice of the evaluation algorithm. The objective of this work is to compare the performance of the evaluation techniques generalized least squares (GLS), generalized nonlinear least squares in the parameter domain (GLS-P), and the Unified Monte Carlo evaluation algorithms B (UMC-B) and G (UMC-G), by using synthetic data. In particular, the effects of model defects are investigated. For small model defects, UMC-B and GLS-P are found to perform best, while these techniques yield the worst results for a significantly defective model; in particular, they seriously underestimate the uncertainties. If UMC-B is augmented with Gaussian processes, it performs distinctly better for a defective model but is more susceptible to an inadequate experimental covariance estimate.

Author Information

Helgesson, Petter
Uppsala University, Dept. of Physics and Astronomy, Uppsala, SE Nuclear Research and Consultancy Group NRG, Petten, NL
Neudecker, Denise
Los Alamos National Laboratory, Los Alamos, NM, US
Sjöstrand, Henrik
Uppsala University, Dept. of Physics and Astronomy, Uppsala, SE
Grosskopf, Michael
Simon Fraser University, Dept. of Statistics and Actuarial Science, Burnaby, CA
Smith, Donald
Argonne National Laboratory (retired), Coronado, CA, US
Capote, Roberto
Argonne National Laboratory (retired), Coronado, CA, US International Atomic Energy Agency, NAPC-Nuclear Data Section, Vienna, AT
Price: $25.00
Contact Sales
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Developed by Committee: E10
Pages: 105–116
DOI: 10.1520/STP160820170087
ISBN-EB: 978-0-8031-7662-1
ISBN-13: 978-0-8031-7661-4