skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Statistical model selection for better prediction and discovering science mechanisms that affect reliability

Journal Article · · Systems
 [1];  [1];  [2]
  1. Los Alamos National Lab. (LANL), Los Alamos, NM (United States). Statistical Sciences Group
  2. ARDEC, Picatinny Arsenal, Township, NJ (United States)
+ Show Author Affiliations

Understanding the impact of production, environmental exposure and age characteristics on the reliability of a population is frequently based on underlying science and empirical assessment. When there is incomplete science to prescribe which inputs should be included in a model of reliability to predict future trends, statistical model/variable selection techniques can be leveraged on a stockpile or population of units to improve reliability predictions as well as suggest new mechanisms affecting reliability to explore. We describe a five-step process for exploring relationships between available summaries of age, usage and environmental exposure and reliability. The process involves first identifying potential candidate inputs, then second organizing data for the analysis. Third, a variety of models with different combinations of the inputs are estimated, and fourth, flexible metrics are used to compare them. As a result, plots of the predicted relationships are examined to distill leading model contenders into a prioritized list for subject matter experts to understand and compare. The complexity of the model, quality of prediction and cost of future data collection are all factors to be considered by the subject matter experts when selecting a final model.

Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Sponsoring Organization:
USDOE
Grant/Contract Number:
AC52-06NA25396
OSTI ID:
1224056
Report Number(s):
LA-UR-15-21723; PII: systems3030109
Journal Information:
Systems, Vol. 3, Issue 3; ISSN 2079-8954
Publisher:
MDPICopyright Statement
Country of Publication:
United States
Language:
English

References (11)

System Health Assessment journal March 2011
A new look at the statistical model identification journal December 1974
Estimating the Dimension of a Model journal March 1978
Bayesian measures of model complexity and fit
  • Spiegelhalter, David J.; Best, Nicola G.; Carlin, Bradley P.
  • Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 64, Issue 4 https://doi.org/10.1111/1467-9868.00353
journal October 2002
Optimal predictive model selection journal June 2004
Model Selection for Good Estimation and Prediction over a User-Specified Covariate Distribution for Linear Models under the Frequentist Paradigm journal October 2011
Optimization of Designed Experiments Based on Multiple Criteria Utilizing a Pareto Frontier journal November 2011
Incorporating response variability and estimation uncertainty into Pareto front optimization journal October 2014
Least angle regression journal April 2004
QQ Models: Joint Modeling for Quantitative and Qualitative Quality Responses in Manufacturing Systems journal April 2015
Optimal predictive model selection text January 2004

Cited By (3)

Comparing the Reliability of Related Populations With the Probability of Agreement journal July 2016
Comparing the Reliability of Related Populations With the Probability of Agreement dataset April 2017
Comparing the Reliability of Related Populations With the Probability of Agreement [Supplemental Data] dataset April 2017

Figures / Tables (17)

Figure 1(p. 4)figure Figure 1
Table 1(p. 8)table Table 1
Table 2(p. 8)table Table 2
Table 3(p. 8)table Table 3
Table 4(p. 8)table Table 4
Figure 2(p. 9)figure Figure 2
Figure 3(p. 10)figure Figure 3
Figure 4(p. 14)figure Figure 4
Table 5(p. 16)table Table 5
Figure 5(p. 17)figure Figure 5
Figure 6(p. 18)figure Figure 6
Figure 7(p. 19)figure Figure 7
Figure 8(p. 20)figure Figure 8
Figure 9(p. 21)figure Figure 9
Figure 10(p. 22)figure Figure 10
Figure 11(p. 23)figure Figure 11
Table 6(p. 23)table Table 6

Similar Records

Related Subjects

97 MATHEMATICS AND COMPUTING
automated model evaluation
variable selection
environmental exposure
system usage
advancing underlying theory