Developers - S. Kucherenko, O. Zaccheus

SobolGSA is general purpose GUI driven global sensitivity analysis and metamodeling software. It can be used to compute various sensitivity measures and/or to develop metamodels.

There is a choice of three different metamodeling techniques, including Quasi Random Sampling-High dimensional model representation (QRS-HDMR) method both with regression and projection methods and radial basis function method. The improved QRS-HDMR method is described in [1-3]. SobolGSA can be used to construct metamodels either from explicitly known models or directly from data produced by "black-box" models. SobolGSA can be applied to both static and time-dependent problems. It can handle several outputs for analysis; each output can be time-dependent. Developed metamodels are produced in a form of self-contained MATLAB or C# files which can be used as accurate, reliable and very fast surrogates of the original CPU-expensive full scale models.

SobolGSA is a tool for global sensitivity analysis (GSA). GSA methods evaluate the effect of a factor while all other factors are varied as well and thus they account for interactions between variables and do not depend on the choice of a nominal point like local sensitivity analysis methods. The set of available global sensitivity analysis techniques includes screening methods (Morris measure), variance (Sobol’ indices, FAST) and derivative based sensitivity measures [4-8].

Sensitivity measures can be computed directly using MC/QMC techniques or by building metamodels first and then computing sensitivity measures using metamodels. The second approach is a much cheaper in terms of required function evaluations.

All techniques implemented in SobolGSA make use of Quasi Monte Carlo sampling based on Sobol sequences [9,10]. The software has a user friendly interface for inputs and presenting results. SobolGSA can be linked to MATLAB and other packages. Software includes detailed manuals, case studies and a set of test problems with descriptions for benchmarking and training.

Comments and questions can be sent to s.kucherenko@imperial.ac.uk.

References

  1. Kucherenko S. SobolHDMR: a general-purpose modeling software Methods Mol Biol. (2013) 1073:191-224. doi: 10.1007/978-1-62703-625-2_16
  2. Feil B., Kucherenko S., Shah N. Comparison of Monte Carlo and Quasi Monte Carlo Sampling Methods in High Dimensional Model Representation, SIMUL 2009, Porto, Portugal.
  3. Zuniga M., Kucherenko S., Shah N. Metamodelling with independent and dependent inputs, Computer Physics Communications, 184, 6 (2013) 1570-1580.
  4. Sobol’ I., Kucherenko S. Global Sensitivity Indices for Nonlinear Mathematical Models. Review, Wilmott, 1 (2005) 56-61.
  5. Kucherenko S., Feil B., Shah N., Mauntz W. The identification of model effective dimensions using global sensitivity analysis Reliability Engineering and System Safety 96 (2011) 440–449.
  6. Kucherenko S., Tarantola S., Annoni P. Estimation of global sensitivity indices for models with dependent variables, Computer Physics Communications, 183 (2012) 937–946.
  7. Kucherenko S., Rodriguez-Fernandez M., Pantelides C., Shah N. Monte Carlo evaluation of derivative based global sensitivity measures. Reliability Engineering and System Safety 94, 7 (2009) 1135-1148.
  8. Sobol’ I.M., Kucherenko S. Derivative based Global Sensitivity Measures and their link with global sensitivity indices, Mathematics and Computers in Simulation, 79, 10 (2009) 3009-3017.
  9. Sobol’ I.M., Asotsky D., Kreinin A., Kucherenko S. Construction and Comparison of High-Dimensional Sobol’ Generators, 2011, Wilmott Journal (2012) 64-79.
  10. Kucherenko S., Albrecht D., Saltelli A. Exploring multi-dimensional spaces: a Comparison of Latin Hypercube and Quasi Monte Carlo Sampling Techniques, to be published in Reliability Engineering and System Safety, 2016 arXiv:1505.02350.

Download software

Please fill out the form with your contact details and we will process your download request.

SUBSCRIBE: Join our mailing list

Imperial College logo

UCL logo