A new approach to counting measurements: Addressing the problems with ISO11929
We present an alternative approach to making counting measurements of radioactivity which offers probabilistic interpretations of the measurements. Unlike the approach in the current international standard (ISO11929), our approach, which uses an assumed prior probability distribution of the true amount in the sample, is able to answer the question of interest for most users of the standard: “what is the probability distribution of the true amount in the sample, given the data?” The final interpretation of the measurement requires information not necessarily available at the measurement stage. However, we provide an analytical formula for what we term the “measurement strength” that depends only on measurementstage count quantities. Here, we show that, when the sources are rare, the posterior odds that the sample true value exceeds ε are the measurement strength times the prior odds, independently of ε, the prior odds, and the distribution of the calibration coefficient. We recommend that the measurement lab immediately followup on unusually high samples using an “action threshold” on the measurement strength which is similar to the decision threshold recommended by the current standard. Finally, we further recommend that the measurement lab perform large background studies in order to characterize non constancy of background, includingmore »
 Authors:

^{[1]}
;
^{[1]}
;
^{[2]}
 Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
 Santa Fe, NM (United States)
 Publication Date:
 Report Number(s):
 LAUR1724027
Journal ID: ISSN 01689002; TRN: US1800894
 Grant/Contract Number:
 AC5206NA25396
 Type:
 Accepted Manuscript
 Journal Name:
 Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment
 Additional Journal Information:
 Journal Volume: 892; Journal ID: ISSN 01689002
 Publisher:
 Elsevier
 Research Org:
 Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
 Sponsoring Org:
 USDOE
 Country of Publication:
 United States
 Language:
 English
 Subject:
 46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; Radiation Protection; Counting Measurements; Background Radiation; Bayesian Statistics; Odds Ratio
 OSTI Identifier:
 1416281
Klumpp, John Allan, Poudel, Deepesh, and Miller, Guthrie. A new approach to counting measurements: Addressing the problems with ISO11929. United States: N. p.,
Web. doi:10.1016/j.nima.2017.12.016.
Klumpp, John Allan, Poudel, Deepesh, & Miller, Guthrie. A new approach to counting measurements: Addressing the problems with ISO11929. United States. doi:10.1016/j.nima.2017.12.016.
Klumpp, John Allan, Poudel, Deepesh, and Miller, Guthrie. 2017.
"A new approach to counting measurements: Addressing the problems with ISO11929". United States.
doi:10.1016/j.nima.2017.12.016. https://www.osti.gov/servlets/purl/1416281.
@article{osti_1416281,
title = {A new approach to counting measurements: Addressing the problems with ISO11929},
author = {Klumpp, John Allan and Poudel, Deepesh and Miller, Guthrie},
abstractNote = {We present an alternative approach to making counting measurements of radioactivity which offers probabilistic interpretations of the measurements. Unlike the approach in the current international standard (ISO11929), our approach, which uses an assumed prior probability distribution of the true amount in the sample, is able to answer the question of interest for most users of the standard: “what is the probability distribution of the true amount in the sample, given the data?” The final interpretation of the measurement requires information not necessarily available at the measurement stage. However, we provide an analytical formula for what we term the “measurement strength” that depends only on measurementstage count quantities. Here, we show that, when the sources are rare, the posterior odds that the sample true value exceeds ε are the measurement strength times the prior odds, independently of ε, the prior odds, and the distribution of the calibration coefficient. We recommend that the measurement lab immediately followup on unusually high samples using an “action threshold” on the measurement strength which is similar to the decision threshold recommended by the current standard. Finally, we further recommend that the measurement lab perform large background studies in order to characterize non constancy of background, including possible time correlation of background.},
doi = {10.1016/j.nima.2017.12.016},
journal = {Nuclear Instruments and Methods in Physics Research. Section A, Accelerators, Spectrometers, Detectors and Associated Equipment},
number = ,
volume = 892,
place = {United States},
year = {2017},
month = {12}
}