Global sensitivity analysis of stochastic computer models with joint metamodels

The global sensitivity analysis method used to quantify the influence of uncertain input variables on the variability in numerical model responses has already been applied to deterministic computer codes; deterministic means here that the same set of input variables always gives the same output value. This paper proposes a global sensitivity analysis methodology for stochastic computer codes, for which the result of each code run is itself random. The framework of the joint modeling of the mean and dispersion of heteroscedastic data is used. To deal with the complexity of computer experiment outputs, nonparametric joint models are discussed and a new Gaussian process-based joint model is proposed. The relevance of these models is analyzed based upon two case studies. Results show that the joint modeling approach yields accurate sensitivity index estimators even when heteroscedasticity is strong.

[1]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[2]  Irène Gijbels,et al.  Nonparametric estimation of mean and dispersion functions in extended generalized linear models , 2008 .

[3]  James O. Berger,et al.  A Framework for Validation of Computer Models , 2007, Technometrics.

[4]  Victor Picheny,et al.  Quantile-Based Optimization of Noisy Computer Experiments With Tunable Precision , 2013, Technometrics.

[5]  Serhat Yesilyurt,et al.  Bayesian-Validated Surrogates for Noisy Computer Simulations; Application to Random Media , 1996, SIAM J. Sci. Comput..

[6]  Emmanuel Manceau,et al.  Combination of Experimental Design and Joint Modeling Methods for Quantifying the Risk Associated With Deterministic and Stochastic Uncertainties - An Integrated Test Study , 2001 .

[7]  J. Nelder,et al.  An extended quasi-likelihood function , 1987 .

[8]  Dan Cornford,et al.  Learning Heteroscedastic Gaussian Processes for Complex Datasets , 2009 .

[9]  Douglas C. Montgomery,et al.  Response Surface Methodology: Process and Product Optimization Using Designed Experiments , 1995 .

[10]  Emmanuel Manceau,et al.  Uncertainty management: From geological scenarios to production scheme optimization , 2004 .

[11]  Jon C. Helton,et al.  Implementation and evaluation of nonparametric regression procedures for sensitivity analysis of computationally demanding models , 2009, Reliab. Eng. Syst. Saf..

[12]  S. Pope.  Lagrangian PDF Methods for Turbulent Flows , 1994 .

[13]  Jack P. C. Kleijnen,et al.  Robust Optimization in Simulation: Taguchi and Response Surface Methodology , 2008 .

[14]  R. A. Rigby,et al.  A semi-parametric additive model for variance heterogeneity , 1996, Stat. Comput..

[15]  A. O'Hagan,et al.  Bayesian calibration of computer models , 2001 .

[16]  Jon C. Helton,et al.  Survey of sampling-based methods for uncertainty and sensitivity analysis , 2006, Reliab. Eng. Syst. Saf..

[17]  Ron S. Kenett,et al.  Achieving Robust Design from Computer Simulations , 2006 .

[18]  John A. Nelder,et al.  Robust Design via Generalized Linear Models , 2003 .

[19]  Bertrand Iooss,et al.  An efficient methodology for modeling complex computer codes with Gaussian processes , 2008, Comput. Stat. Data Anal..

[20]  S. Pope.  PDF methods for turbulent reactive flows , 1985 .

[21]  Paola Annoni,et al.  Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index , 2010, Comput. Phys. Commun..

[22]  Connie M. Borror,et al.  Response surface design evaluation and comparison , 2009 .

[23]  M. Fuentes,et al.  Variable selection for high dimensional Bayesian density estimation: application to human exposure simulation , 2012 .

[24]  T. Simpson,et al.  Use of Kriging Models to Approximate Deterministic Computer Models , 2005 .

[25]  Russell R. Barton,et al.  A review on design, modeling and applications of computer experiments , 2006 .

[26]  David M. Steinberg,et al.  Screening Experiments for Dispersion Effects , 2006 .

[27]  S. Wood,et al.  GAMs with integrated model selection using penalized regression splines and applications to environmental modelling , 2002 .

[28]  Sonja Kuhnt,et al.  Design and analysis of computer experiments , 2010 .

[29]  Stefano Tarantola,et al.  Hydrocarbon exploration risk evaluation through uncertainty and sensitivity analyses techniques , 2006, Reliab. Eng. Syst. Saf..

[30]  Henry P. Wynn,et al.  Screening, predicting, and computer experiments , 1992 .

[31]  Yann Richet,et al.  Kriging with Heterogeneous Nugget Effect for the Approximation of Noisy Simulators with Tunable Fidelity (Krigeage avec effet de pépite hétérogène pour l'approximation de simulateurs bruités à fidélité réglable) , 2008 .

[32]  Emmanuel Manceau,et al.  A NEW APPROACH FOR QUANTIFYING THE IMPACT OF GEOSTATISTICAL UNCERTAINTY ON PRODUCTION FORECASTS : THE JOINT MODELING METHOD IAMG , Cancun , September 6-12 , 2001 , 2001 .

[33]  Barry L. Nelson,et al.  Stochastic kriging for simulation metamodeling , 2008, 2008 Winter Simulation Conference.

[34]  Gordon K. Smyth,et al.  Generalized linear models with varying dispersion , 1989 .

[35]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[36]  Jack P. C. Kleijnen,et al.  Sensitivity analysis and related analyses: A review of some statistical techniques , 1997 .

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

[38]  R. Tibshirani,et al.  Generalized Additive Models , 1986 .

[39]  Wolfram Burgard,et al.  Most likely heteroscedastic Gaussian process regression , 2007, ICML '07.

[40]  Jon Craig Helton,et al.  Conceptual and computational basis for the quantification of margins and uncertainty. , 2009 .

[41]  D. Shahsavani,et al.  Variance-based sensitivity analysis of model outputs using surrogate models , 2011, Environ. Model. Softw..

[42]  Runze Li,et al.  Design and Modeling for Computer Experiments , 2005 .

[43]  Jerome Sacks,et al.  Choosing the Sample Size of a Computer Experiment: A Practical Guide , 2009, Technometrics.

[44]  Thomas J. Santner,et al.  The Design and Analysis of Computer Experiments , 2003, Springer Series in Statistics.

[45]  Bertrand Iooss,et al.  Response surfaces and sensitivity analyses for an environmental model of dose calculations , 2006, Reliab. Eng. Syst. Saf..

[46]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[47]  A. Saltelli,et al.  Making best use of model evaluations to compute sensitivity indices , 2002 .

[48]  M. J. Bayarri,et al.  Computer model validation with functional output , 2007, 0711.3271.

[49]  G. Geoffrey Vining,et al.  Combining Taguchi and Response Surface Philosophies: A Dual Response Approach , 1990 .

[50]  Stefano Tarantola,et al.  Uncertainty in Industrial Practice , 2008 .

[51]  Jack P. C. Kleijnen,et al.  Robustness of Kriging when interpolating in random simulation with heterogeneous variances: Some experiments , 2005, Eur. J. Oper. Res..

[52]  Ilya M. Sobol,et al.  Sensitivity Estimates for Nonlinear Mathematical Models , 1993 .

[53]  Madhan Shridhar Phadke,et al.  Quality Engineering Using Robust Design , 1989 .

[54]  Peer-Olaf Siebers,et al.  Discrete-event simulation is dead, long live agent-based simulation! , 2010, J. Simulation.

[55]  B. Iooss,et al.  Global sensitivity analysis for a numerical model of radionuclide migration from the RRC “Kurchatov Institute” radwaste disposal site , 2008 .

[56]  Bertrand Iooss,et al.  Global sensitivity analysis of computer models with functional inputs , 2008, Reliab. Eng. Syst. Saf..

[57]  G. Wahba.  Spline models for observational data , 1990 .

[58]  J A Nelder.  A large class of models derived from generalized linear models. , 1998, Statistics in medicine.

[59]  A. Forrester,et al.  Design and analysis of 'noisy' computer experiments , 2006 .

[60]  Robert B. Gramacy,et al.  Ja n 20 08 Bayesian Treed Gaussian Process Models with an Application to Computer Modeling , 2009 .

[61]  J. Chilès,et al.  Geostatistics: Modeling Spatial Uncertainty , 1999 .

[62]  A. Saltelli,et al.  A quantitative model-independent method for global sensitivity analysis of model output , 1999 .

[63]  Robert E. Shannon,et al.  Design and analysis of simulation experiments , 1978, WSC '78.

[64]  J. P. Dejean,et al.  Prediction and Density Estimation of a Horizontal Well Productivity Index Using Generalized Linear Models , 1998 .

[65]  Stanley H. Cohen,et al.  Design and Analysis , 2010 .

[66]  Randall P. Sadowski,et al.  Simulation with Arena , 1998 .

[67]  Joseph A. C. Delaney.  Sensitivity analysis , 2018, The African Continental Free Trade Area: Economic and Distributional Effects.

[68]  Runze Li,et al.  Design and Modeling for Computer Experiments (Computer Science & Data Analysis) , 2005 .

[69]  P. McCullagh,et al.  Generalized Linear Models , 1984 .

[70]  Jeffrey B. Birch,et al.  A semi-parametric approach to dual modeling when no replication exists , 2010 .

[71]  A. Saltelli,et al.  Importance measures in global sensitivity analysis of nonlinear models , 1996 .

[72]  Jean-Camille Chassaing,et al.  A discontinuous galerkin approach for the numerical simulation of transit-time ultrasonic flowmeters , 2014, 2014 IEEE International Ultrasonics Symposium.