Variance-based sensitivity analysis provides a quantitative measure of how uncertainty in a model input contributes to uncertainty in the model output. Such sensitivity analyses arise in a wide variety of applications and are typically computed using Monte Carlo estimation, but the many samples required for Monte Carlo to be sufficiently accurate can make these analyses intractable when the model is expensive. This work presents a multifidelity approach for estimating sensitivity indices that leverages cheaper low-fidelity models to reduce the cost of sensitivity analysis while retaining accuracy guarantees via recourse to the original, expensive model. This paper develops new multifidelity estimators for variance and for the Sobol' main and total effect sensitivity indices. We discuss strategies for dividing limited computational resources among models and specify a recommended strategy. Results are presented for the Ishigami function and a convection-diffusion-reaction model that demonstrate up to $10\times$ speedups for fixed convergence levels. For the problems tested, the multifidelity approach allows inputs to be definitively ranked in importance when Monte Carlo alone fails to do so.

  • 1.  D. Allaire and  K. Willcox , Surrogate modeling for uncertainty assessment with application to aviation environmental system models , AIAA J. , 48 ( 2010 ), pp. 1791 -- 1803 . CrossrefGoogle Scholar

  • 2.  C. Bierig and  A. Chernov , Convergence analysis of multilevel Monte Carlo variance estimators and application for random obstacle problems , Numer. Math. , 130 ( 2015 ), pp. 579 -- 613 . CrossrefISIGoogle Scholar

  • 3.  E. Borgonovo , Measuring uncertainty importance: Investigation and comparison of alternative approaches , Risk Anal. , 26 ( 2006 ), pp. 1349 -- 1361 . CrossrefISIGoogle Scholar

  • 4.  E. Borgonovo , A new uncertainty importance measure , Reliab. Engrg. Syst. Safety , 92 ( 2007 ), pp. 771 -- 784 . CrossrefISIGoogle Scholar

  • 5.  M. Buffoni and  K. Willcox , Projection-based model reduction for reacting flows , in 40th Fluid Dynamics Conference and Exhibit, AIAA (American Institute of Aeronautics & Astronautics) , 2010 , p. 5008 . Google Scholar

  • 6.  S. Chaturantabut and  D. Sorensen , Nonlinear model reduction via discrete empirical interpolation , SIAM J. Sci. Comput. , 32 ( 2010 ), pp. 2737 -- 2764 . LinkISIGoogle Scholar

  • 7.  W. Chen R. Jin and  A. Sudjianto , Analytical global sensitivity analysis and uncertainty propagation for robust design , J. Qual. Technol. , 38 ( 2006 ), p. 333 . CrossrefISIGoogle Scholar

  • 8.  M. Chun S. Han and  N. Tak , An uncertainty importance measure using a distance metric for the change in a cumulative distribution function , Reliab. Engrg. Syst. Safety , 70 ( 2000 ), pp. 313 -- 321 . CrossrefISIGoogle Scholar

  • 9.  W. Cochran , Sampling Techniques , John Wiley & Sons , New York , 2007 . Google Scholar

  • 10.  B. Cuenot and  T. Poinsot , Asymptotic and numerical study of diffusion flames with variable Lewis number and finite rate chemistry , Combust. Flame , 104 ( 1996 ), pp. 111 -- 137 . CrossrefISIGoogle Scholar

  • 11.  R. I. Cukier C. M. Fortuin K. E. Shuler A. G. Petschek and  J. H. Schaibly , Study of the sensitivity of coupled reaction systems to uncertainties in rate coefficients. I theory , J. Chem. Phys. , 59 ( 1973 ), pp. 3873 -- 3878 . CrossrefISIGoogle Scholar

  • 12.  M. S. Eldred L. W. T. Ng M. F. Barone and  S. P. Domino , Multifidelity uncertainty quantification using spectral stochastic discrepancy models , in Handbook of Uncertainty Quantification , R. Ghanem, D. Higdon, and H. Owhadi, eds., Springer International Publishing , Cham , 2016 , pp. 1 -- 45 . Google Scholar

  • 13.  M. Giles Carlo path simulation , Oper. Res. , 56 ( 2008 ), pp. 607 -- 617 . CrossrefISIGoogle Scholar

  • 14.  J. W. Hall S. A. Boyce Y. Wang R. J. Dawson S. Tarantola and  A. Saltelli , Sensitivity analysis for hydraulic models , J. Hydraulic Engrg. , 135 ( 2009 ), pp. 959 -- 969 . CrossrefISIGoogle Scholar

  • 15.  J. Hart A. Alexanderian and  P. Gremaud , Efficient computation of Sobol' indices for stochastic models , SIAM J. Sci. Comput. , 39 ( 2017 ), pp. A1514 -- A1530 , https://doi.org/10.1137/16M106193X. LinkISIGoogle Scholar

  • 16.  J. C. Helton and  F. J. Davis , Latin hypercube sampling and the propagation of uncertainty in analyses of complex systems , Reliab. Engrg. Syst. Safety , 81 ( 2003 ), pp. 23 -- 69 . CrossrefISIGoogle Scholar

  • 17.  W. Hoeffding class of statistics with asymptotically normal distribution , Ann. Math. Statist. , ( 1948 ), pp. 293 -- 325 . Google Scholar

  • 18.  T. Homma and  A. Saltelli , Importance measures in global sensitivity analysis of nonlinear models , Reliab. Engrg. Syst. Safety , 52 ( 1996 ), pp. 1 -- 17 . CrossrefISIGoogle Scholar

  • 19.  B. Iooss F. V. Dorpe and  N. Devictor , Response surfaces and sensitivity analyses for an environmental model of dose calculations , Reliab. Engrg. Syst. Safety , 91 ( 2006 ), pp. 1241 -- 1251 . CrossrefISIGoogle Scholar

  • 20.  B. Iooss and  P. Lemaître , A review on global sensitivity analysis methods , in Uncertainty Management in Simulation-Optimization of Complex Systems , G. Dellino and C. Meloni, eds., Operations Research/Computer Science Interfaces Series 59, Springer , New York, 2015 , pp. 101 -- 122 , https://doi.org/10.1007/978-1-4899-7547-8. Google Scholar

  • 21.  T. Ishigami and  T. Homma , An importance quantification technique in uncertainty analysis for computer models , in Proceedings of the First International Symposium on Uncertainty Modeling and Analysis,, IEEE , New York , 1990 , pp. 398 -- 403 . Google Scholar

  • 22.  A. Janon T. Klein A. Lagnoux M. Nodet and  C. Prieur , Asymptotic normality and efficiency of two Sobol index estimators , ESAIM Probab. Stat. , 18 ( 2014 ), pp. 342 -- 364 . CrossrefGoogle Scholar

  • 23.  A. Janon M. Nodet and  C. Prieur , Uncertainties assessment in global sensitivity indices estimation from metamodels , Int. J. Uncertain. Quantif. , 4 ( 2014 ), pp. 21 -- 36 . CrossrefISIGoogle Scholar

  • 24.  M. Kendall and  A. Stuart , The Advanced Theory of Statistics , 4 th ed. Macmillan New York , 1979 . Google Scholar

  • 25.  B. Krzykacz : A Computer Program for the Derivation of Empirical Sensitivity Measures of Results from Large Computer Models, Technical report GRS-A-1700, Gesellschaft für Reaktorsicherheit , Garching , Germany , 1990 . Google Scholar

  • 26.  S. Kucherenko B. Depuech B. Iooss and  S. Tarantola , Application of the control variate technique to estimation of total sensitivity indices , Reliab. Engrf. Syst. Safety , 134 ( 2015 ), pp. 251 -- 259 , https://doi.org/10.1016/j.ress.2014.07.008. CrossrefISIGoogle Scholar

  • 27.  L. LeGratiet C. Cannamela and  B. Iooss , A Bayesian approach for global sensitivity analysis of (multifidelity) computer codes , SIAM/ASA J. Uncertain. Quantif. , 2 ( 2014 ), pp. 336 -- 363 . LinkGoogle Scholar

  • 28.  H. Liu W. Chen and  A. Sudjianto , Relative entropy based method for global and regional sensitivity analysis in probabilistic design , ASME J. Mech. Des. , 128 ( 2006 ), pp. 1 -- 11 . CrossrefISIGoogle Scholar

  • 29.  W. Mauntz , Global Sensitivity Analysis of General Nonlinear Systems , Master's thesis , Imperial College London , CPSE , 2002 . Google Scholar

  • 30.  M. D. Morris , Factorial sampling plans for preliminary computational experiments , Technometrics , 33 ( 1991 ), pp. 161 -- 174 . CrossrefISIGoogle Scholar

  • 31.  A. Narayan C. Gittelson and  D. Xiu , A stochastic collocation algorithm with multifidelity models , SIAM J. Sci. Comput. , 36 ( 2014 ), pp. A495 -- A521 . LinkISIGoogle Scholar

  • 32.  L. Ng and  K. Willcox , Multifidelity approaches for optimization under uncertainty , Int. J. Numer. Methods Engrg. , 100 ( 2014 ), pp. 746 -- 772 , https://doi.org/10.1002/nme.4761. CrossrefISIGoogle Scholar

  • 33.  L. Ng and  K. Willcox , Monte Carlo information-reuse approach to aircraft conceptual design optimization under uncertainty , J. Aircraft , 53 ( 2016 ), pp. 427 -- 438 , https://doi.org/10.2514/1.C033352. CrossrefISIGoogle Scholar

  • 34.  M. Opgenoord D. L. Allaire and  K. E. Willcox , Variance-based sensitivity analysis to support simulation-based design under uncertainty , J. Mech. Des. , 138 ( 2016 ), p. 111410 . CrossrefISIGoogle Scholar

  • 35.  A. B. Owen , Better estimation of small Sobol' sensitivity indices , ACM Trans. Model. Comput. Simul. , 23 ( 2013 ), pp. 11: 1 -- 11 , https://doi.org/10.1145/2457459.2457460. CrossrefISIGoogle Scholar

  • 36.  A. B. Owen , Variance components and generalized Sobol' indices , SIAM/ASA J. Uncertain. Quantif. , 1 ( 2013 ), pp. 19 -- 41 , https://doi.org/10.1137/120876782. LinkGoogle Scholar

  • 37.  P. Palar L. Zuhal K. Shimoyama and  T. Tsuchiya , Global sensitivity analysis via multi-fidelity polynomial chaos expansion , Reliab. Engrg. Syst. Safety , 170 ( 2017 ), pp. 175 -- 190 , https://doi.org/10.1016/j.ress.2017.10.013. CrossrefISIGoogle Scholar

  • 38.  B. Peherstorfer K. Willcox and  M. Gunzburger , Optimal model management for multifidelity Monte Carlo estimation , SIAM J. Sci. Comput. , 38 ( 2016 ), pp. A3163 -- A3194 . LinkISIGoogle Scholar

  • 39.  B. Peherstorfer K. Willcox and  M. Gunzburger , Survey of multifidelity methods in uncertainty propagation, inference, and optimization , SIAM Rev., to appear. Google Scholar

  • 40.  S. Rahman , A generalized anova dimensional decomposition for dependent probability measures , SIAM/ASA J. Uncertain. Quantif. , 2 ( 2014 ), pp. 670 -- 697 . LinkISIGoogle Scholar

  • 41.  S. Rahman , The f-sensitivity index , SIAM/ASA J. Uncertain. Quantif. , 4 ( 2016 ), pp. 130 -- 162 . LinkISIGoogle Scholar

  • 42.  M. Ratto A. Pagano and  P. C. Young , Non-parametric estimation of conditional moments for sensitivity analysis , Reliab. Engrg. Syst. Safety , 94 ( 2009 ), pp. 237 -- 243 . CrossrefISIGoogle Scholar

  • 43.  A. Saltelli , Making best use of model evaluations to compute sensitivity indices , Comput. Phys. Commun. , 145 ( 2002 ), pp. 280 -- 297 , https://doi.org/10.1016/S0010-4655(02)00280-1., http://www.sciencedirect.com/science/article/pii/S0010465502002801. CrossrefISIGoogle Scholar

  • 44.  A. Saltelli T. H. Andres and  T. Homma , Sensitivity analysis of model output: An investigation of new techniques , Comput. Statist. Data Anal. , 15 ( 1993 ), pp. 211 -- 238 . CrossrefISIGoogle Scholar

  • 45.  A. Saltelli and  R. Bolado , An alternative way to compute fourier amplitude sensitivity test (fast) , Comput. Statist. Data Anal. , 26 ( 1998 ), pp. 445 -- 460 . CrossrefISIGoogle Scholar

  • 46.  A. Saltelli and  J. Marivoet , Non-parametric statistics in sensitivity analysis for model output: a comparison of selected techniques , Reliab. Engrg. Syst. Safety , 28 ( 1990 ), pp. 229 -- 253 . CrossrefISIGoogle Scholar

  • 47.  A. Saltelli and  S. Tarantola , On the relative importance of input factors in mathematical models: Safety assessment for nuclear waste disposal , J. Amer. Statist. Assoc. , 97 ( 2002 ), pp. 702 -- 709 . CrossrefISIGoogle Scholar

  • 48.  A. Saltelli S. Tarantola F. Campolongo and  M. Ratto , Sensitivity Analysis in Practice: A Guide to Assessing Scientific Models , John Wiley & Sons , New York , 2004 . Google Scholar

  • 49.  T. J. Santner B. J. Williams and  W. I. Notz , The Design and Analysis of Computer Experiments , Springer Science & Business Media , New York , 2013 . Google Scholar

  • 50.  I. Sobol' , Sensitivity estimates for nonlinear mathematical models , Mat. Modelirovanie , 1 ( 1993 ), pp. 407 -- 414 . Google Scholar

  • 51.  C. B. Storlie L. P. Swiler J. C. Helton and  C. J. Sallaberry , Implementation and evaluation of nonparametric regression procedures for sensitivity analysis of computationally demanding models , Reliab. Engrg. Syst. Safety , 94 ( 2009 ), pp. 1735 -- 1763 . CrossrefISIGoogle Scholar

  • 52.  X. Zhu A. Narayan and  D. Xiu , Computational aspects of stochastic collocation with multifidelity models , SIAM/ASA J. Uncertain. Quantif. , 2 ( 2014 ), pp. 444 -- 463 . LinkISIGoogle Scholar