Estimating alarm thresholds for process monitoring data under different assumptions about the data generating mechanism

Process monitoring (PM) for nuclear safeguards sometimes requires estimation of thresholds corresponding to small false alarm rates. Threshold estimation dates to the 1920s with the Shewhart control chart; however, because possible new roles for PM are being evaluated in nuclear safeguards, it is timely to consider modern model selection options in the context of threshold estimation. One of the possible new PM roles involves PM residuals, where a residual is defined as residual = data − prediction. This paper reviews alarm threshold estimation, introduces model selection options, and considers a range of assumptions regarding the data-generating mechanism for PM residuals. Two PM examples from nuclear safeguards are included to motivate the need for alarm threshold estimation. The first example involves mixtures of probability distributions that arise in solution monitoring, which is a common type of PM. The second example involves periodic partial cleanout of in-process inventory, leading to challenging structure in the time series of PM residuals.

[1]  D. Hunter,et al.  mixtools: An R Package for Analyzing Mixture Models , 2009 .

[2]  Adrian E. Raftery,et al.  Model-Based Clustering, Discriminant Analysis, and Density Estimation , 2002 .

[3]  Joseph J. Pignatiello,et al.  On Estimating X̄ Control Chart Limits , 2001 .

[4]  Tom Burr,et al.  Signal estimation and change detection in tank data for nuclear safeguards , 2011 .

[5]  Joseph J. Pignatiello,et al.  On Estimating X-bar Control Chart Limits , 2001 .

[6]  Marion R. Reynolds,et al.  Shewhart x-charts with estimated process variance , 1981 .

[7]  Tom Burr,et al.  Estimating alarm thresholds and the number of components in mixture distributions , 2012 .

[8]  Russell V. Lenth,et al.  Response-Surface Methods in R, Using rsm , 2009 .

[9]  Wilbert C.M. Kallenberg,et al.  Self-adapting control charts , 2004 .

[10]  Bradley Jones.  Bayesian Process Monitoring, Control and Optimization , 2007 .

[11]  Michael F. Simpson,et al.  Modeling the Pyrochemical Reduction of Spent UO2 Fuel in a Pilot-Scale Reactor , 2006 .

[12]  Charles W. Solbrig,et al.  Electrorefiner Liquid Cadmium Cathode Crucible Thermal Shock , 2006 .

[13]  Benjamin M. Adams,et al.  Alternative Designs of the Hodges-Lehmann Control Chart , 1996 .

[14]  Douglas C. Montgomery,et al.  Research Issues and Ideas in Statistical Process Control , 1999 .

[15]  Michaela M. Black,et al.  Maintaining the performance of a learned classifier under concept drift , 1999, Intell. Data Anal..

[16]  Wilbert C.M. Kallenberg,et al.  Parametric control charts , 2003 .

[17]  Subhabrata Chakraborti.  Run length, average run length and false alarm rate of shewhart x-bar chart: exact derivations by conditioning , 2000 .

[18]  Rob J Hyndman,et al.  Sample Quantiles in Statistical Packages , 1996 .

[19]  Wilbert C.M. Kallenberg,et al.  Are estimated control charts in control? , 2001 .

[20]  Wilbert C.M. Kallenberg,et al.  Improved data driven control charts , 2006 .

[21]  Bing-Yi Jing,et al.  Adjusted empirical likelihood method for quantiles , 2003 .

[22]  M. A. Williamson,et al.  PYROPROCESSING FLOWSHEETS FOR RECYCLING USED NUCLEAR FUEL , 2011 .

[23]  William H. Woodall,et al.  High breakdown estimation methods for Phase I multivariate control charts , 2007, Qual. Reliab. Eng. Int..

[24]  Tom Burr,et al.  Pattern Recognition Options to Combine Process Monitoring and Material Accounting Data in Nuclear Safeguards , 2012 .

[25]  Gemai Chen,et al.  THE MEAN AND STANDARD DEVIATION OF THE RUN LENGTH DISTRIBUTION OF X̄ CHARTS WHEN CONTROL LIMITS ARE ESTIMATED Gemai Chen , 2003 .

[26]  L. K. Chan,et al.  Robustness of mean E(X) and R charts , 1988 .

[27]  Jennifer L. Wadsworth,et al.  Likelihood-based procedures for threshold diagnostics and uncertainty in extreme value modelling , 2012 .

[28]  C. B. Roes,et al.  Shewhart-type Charts in Statistical Process Control , 1995 .

[29]  Wilbert C.M. Kallenberg,et al.  Alternative Shewhart-type charts for grouped observations , 2004 .

[30]  Willem Albers,et al.  New Corrections for Old Control Charts , 2003 .

[31]  Charles P. Quesenberry,et al.  The Effect of Sample Size on Estimated Limits for and X Control Charts , 1993 .

[32]  Tom Burr,et al.  Loss detection results on simulated tank data modified by realistic effects , 2012 .

[33]  L. de Haan,et al.  Estimating the probability of a rare event , 1999 .

[34]  Masato Hori,et al.  Study on Loss Detection Algorithms Using Tank Monitoring Data , 2009 .

[35]  L. Haan,et al.  On the Estimation of the Extreme-Value Index and Large Quantile Estimation , 1989 .

[36]  G. T. Cowell,et al.  Some Implications Of Applying NRTA To A Mox Facility With Significant Temporary Stores , 2010 .

[37]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[38]  Wilbert C.M. Kallenberg,et al.  Estimation in Shewhart control charts , 2000 .

[39]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[40]  D. Vaden.  Fuel Conditioning Facility Electrorefiner Process Model , 2006 .

[41]  Michael D. Conerly,et al.  Phase I control chart based on a kernel estimator of the quantile function , 2011, Qual. Reliab. Eng. Int..

[42]  Peter Hall,et al.  On the estimation of extreme tail probabilities , 1997 .

[43]  Tae-Sic Yoo,et al.  Development of Computational Models for the Mark-IV Electrorefiner—Effect of Uranium, Plutonium, and Zirconium Dissolution at the Fuel Basket-Salt Interface , 2010 .