The SAFE (Sensitivity Analysis For Everybody) Toolbox provides a set of functions to perform Global Sensitivity Analysis. It implements several methods, including the Elementary Effects Test, Regional Sensitivity Analysis, Variance-Based (Sobol’) sensitivity analysis and the novel PAWN method. SAFE was originally developed for the Matlab/Octave environment but it is now available also in R and Python.

Pianosi, F., Sarrazin, F., Wagener, T. (2015), A Matlab toolbox for Global Sensitivity Analysis, Environmental Modelling & Software, 70, 80-85. (Open Access)

Why using SAFE

  • Modular structure to facilitate interactions with other computing environments
  • Set of functions to assess the robustness and convergence of sensitivity indices
  • Several visualization tools to investigate and communicate GSA results
  • Lots of comments in the code and workflow examples to get started

Want a playful introduction to the benefits of SAFE/GSA? You can try one of our interactive Jupyter Notebooks:
SAFE-Notebooks on myBinder

Who is using SAFE?

Since its first release in 2015, we have sent a copy of SAFE to >2000 students and researchers around the world and working across a variety of disciplines. More information about who is using SAFE and what they think of it, can be found in our new paper on a survey of SAFE users that we performed in late 2017:

Pianosi, F., Wagener, T., Sarrazin, F. (2020), How successfully is open-source research software adopted? Results and implications of surveying the users of a sensitivity analysis toolbox, Environmental Modelling & Software, 124.

Thanks to all SAFE users that took some time to fill in the survey questionnaire!

Who develops SAFE?

SAFE was originally developed by Francesca Pianosi, Fanny Sarrazin and Thorsten Wagener at the Department of Civil Engineering at the University of Bristol. Other contributors are:
Isabella Gollini (R version) (past contributor)
Valentina Noacco (R version)
Andres Penuela-Fernandez (Python Jupyter Notebooks)

Acknowledgements

The development of SAFE was originally supported by the UK Natural Environment Research Council through the Consortium on Risk in the Environment: Diagnostics, Integration, Benchmarking, Learning and Elicitation (CREDIBLE) [NE/J017450/1].
The further development of SAFE, including the implementation of the Python version, has been supported by the UK Engineering and Physical Sciences Research Council through a Living with Environmental Change Fellowship [EP/R007330/1] and the EPSRC Impact Acceleration Account.