Variance gradients and uncertainty budgets for nonlinear measurement functions with independent inputs

A novel variance-based measure for global sensitivity analysis, termed a variance gradient (VG), is presented for constructing uncertainty budgets under the Guide to the Expression of Uncertainty in Measurement (GUM) framework for nonlinear measurement functions with independent inputs. The motivation behind VGs is the desire of metrologists to understand which inputs' variance reductions would most effectively reduce the variance of the measurand. VGs are particularly useful when the application of the first supplement to the GUM is indicated because of the inadequacy of measurement function linearization. However, VGs reduce to a commonly understood variance decomposition in the case of a linear(ized) measurement function with independent inputs for which the original GUM readily applies. The usefulness of VGs is illustrated by application to an example from the first supplement to the GUM, as well as to the benchmark Ishigami function. A comparison of VGs to other available sensitivity measures is made.

[1]  E. Borgonovo Measuring Uncertainty Importance: Investigation and Comparison of Alternative Approaches , 2006, Risk analysis : an official publication of the Society for Risk Analysis.

[2]  Nicolas Fischer,et al.  SENSITIVITY ANALYSIS IN METROLOGY: STUDY AND COMPARISON ON DIFFERENT INDICES FOR MEASUREMENT UNCERTAINTY , 2009 .

[3]  René Dybkaer,et al.  Revision of the ‘Guide to the Expression of Uncertainty in Measurement’. Why and how , 2012 .

[4]  Maurice G. Cox,et al.  AUTOMATIC DIFFERENTIATION AND ITS APPLICATION IN METROLOGY , 2004 .

[5]  Raghu N. Kacker,et al.  Evolution of modern approaches to express uncertainty in measurement , 2007 .

[6]  Raghu N. Kacker,et al.  On use of Bayesian statistics to make the Guide to the Expression of Uncertainty in Measurement consistent , 2003 .

[7]  K. Weise,et al.  A Bayesian theory of measurement uncertainty , 1993 .

[8]  Saltelli Andrea,et al.  Global Sensitivity Analysis: The Primer , 2008 .

[9]  B. Toman,et al.  Contribution to a conversation about the Supplement 1 to the GUM , 2009 .

[10]  Andrew McCulloch,et al.  Sensitivity Analysis in Practice: a Guide to Assessing Scientific Models , 2005 .

[11]  P M Harris,et al.  Software Support for Metrology Best Practice Guide No. 6. Uncertainty evaluation. , 2006 .

[12]  Clemens Elster,et al.  A two-stage procedure for determining the number of trials in the application of a Monte Carlo method for uncertainty evaluation , 2010 .

[13]  K. Sommer,et al.  Systematic approach to the modelling of measurements for uncertainty evaluation , 2006 .

[14]  Walter Bich,et al.  How to revise the GUM? , 2008 .

[15]  Francesca Pennecchi,et al.  Non-linear models and best estimates in the GUM , 2006 .

[16]  Stefano Tarantola,et al.  Sensitivity Analysis in Practice: A Guide to Assessing Scientific Models , 2004 .

[17]  E. Iso,et al.  Measurement Uncertainty and Probability: Guide to the Expression of Uncertainty in Measurement , 1995 .

[18]  R. Bartle The elements of integration and Lebesgue measure , 1995 .

[19]  M. D. McKay,et al.  A comparison of three methods for selecting values of input variables in the analysis of output from a computer code , 2000 .

[20]  Clemens Elster,et al.  Draft GUM Supplement 1 and Bayesian analysis , 2007 .

[21]  Clemens Elster,et al.  Bayesian uncertainty analysis under prior ignorance of the measurand versus analysis using the Supplement 1 to the Guide: a comparison , 2009 .

[22]  Raghu N. Kacker,et al.  Comparison of ISO-GUM, draft GUM Supplement 1 and Bayesian statistics using simple linear calibration , 2006 .

[23]  Andrew L. Rukhin,et al.  Three Statistical Paradigms for the Assessment and Interpretation of Measurement Uncertainty , 2009 .

[24]  J. Rosenthal A First Look at Rigorous Probability Theory , 2000 .

[25]  P M Harris,et al.  Software specifications for uncertainty evaluation. , 2006 .

[26]  Antonio Possolo,et al.  Copulas for uncertainty analysis , 2010 .

[27]  Blaza Toman,et al.  Assessment of measurement uncertainty via observation equations , 2007 .

[28]  Raghu N. Kacker,et al.  Coefficient of contribution to the combined standard uncertainty , 2006 .

[29]  Roger Frost,et al.  International Organization for Standardization (ISO) , 2004 .

[30]  R. Steiner History and progress on accurate measurements of the Planck constant , 2013, Reports on progress in physics. Physical Society.

[31]  Ignacio Lira,et al.  Comparison between the conventional and Bayesian approaches to evaluate measurement data , 2006 .

[32]  Raghu N. Kacker,et al.  A Test of Linearity Using Covering Arrays for Evaluating Uncertainty in Measurement | NIST , 2008 .

[33]  Maurice G. Cox,et al.  The use of a Monte Carlo method for evaluating uncertainty and expanded uncertainty , 2006 .

[34]  G. Wübbeler,et al.  Evaluation of measurement uncertainty and its numerical calculation by a Monte Carlo method , 2008 .

[35]  M. Rosenblatt Remarks on a Multivariate Transformation , 1952 .

[36]  JCGM 102 : 2011 Evaluation of measurement data – Supplement 2 to the “ Guide to the expression of uncertainty in measurement ” – Extension to any number of output quantities , 2011 .

[37]  F. Pennecchi,et al.  Why always seek the expected value? A discussion relating to the Lp norm , 2007 .