A Review on Global Sensitivity Analysis Methods

By Bertrand Iooss, Paul Lemaître. 2014

1,039 citations recorded by the source corpus.

This chapter makes a review, in a complete methodological framework, of various global sensitivity analysis methods of model output. Numerous statistical and probabilistic tools (regression, smoothing, tests, statistical learning, Monte Carlo, …) aim at determining the model input variables which mostly contribute to an interest quantity depending on model output. This quantity can be for instance the variance of an output variable. Three kinds of methods are distinguished: the screening (coarse sorting of the most influential inputs among a large number), the measures of importance (quantitative sensitivity indices) and the deep exploration of the model behaviour (measuring the effects of inputs on their all variation range). A progressive application methodology is illustrated on a scholar application. A synthesis is given to place every method according to several axes, mainly the cost in number of model evaluations, the model complexity and the nature of brought information.

Subfields: Civil and Structural Engineering, Mechanics of Materials, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/8a8c20df3ed43b936b5d17ded137c03ee04ef823

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Civil and Structural Engineering

In Mechanics of Materials

In Statistics and Probability

Citations in the corpus

Roadkill risk and population vulnerability in European birds and mammals (2020, methodology): “Because the MaxRoad model is non-linear but monotonic, we ran a ranked linear regression model with the input variables of MaxRoad to measure the effects of the parameters across their entire domains (Iooss and Lemaître 2015).”
SHAPLEY EFFECTS FOR SENSITIVITY ANALYSIS WITH CORRELATED INPUTS: COMPARISONS WITH SOBOL' INDICES, NUMERICAL ESTIMATION AND APPLICATIONS (2017, neutral): “As for the Sobol’ indices [1, 16], one important issue in practice is the numerical cost in terms of number of model evaluations required to estimate the Shapley eﬀects.”
The Future of Sensitivity Analysis: An essential discipline for systems modeling and policy support (2020, positive): “Tremendous advances in both theory and application of SA have been accomplished, as documented in the recent reviews by Norton (2015), Iooss and Lemaître (2015), Wei et al. (2015), Razavi and Gupta (2015), Borgonovo and Plischke (2016), Pianosi et al. (2016), Borgonovo et al. (2017), Borgonovo…”
Locally Interpretable Models and Effects based on Supervised Partitioning (LIME-SUP) (2018, neutral): “…(Friedman & Popescu, 2005), related ones such as ICE (Goldstein, Kapelner, Bleich, & Pitkin, 2013) and ALE plots (Apley, 2016), derivative-based analyses (Kucherenko, 2010; Sobol & Kucherenko, 2009), Sobol indices (Iooss & Lemaitre, 2014), Shapley analyses (Lundberg & Lee, 2017), and so on.”
A survey of adaptive sampling for global metamodeling in support of simulation-based complex engineering design (2017, background): “The SA is a very broad topic and has been widely reviewed in different fields, see (Kleijnen 2005; Saltelli et al. 2008; Tian 2013; Janouchová and Ku č erová 2013; Iooss and Lemaître 2015; Borgonovo and Plischke 2016).”
Learning Global Additive Explanations for Neural Nets Using Model Distillation (2018, background): “Examples include permutation-based measures [5], gradients/saliency (see [31] or [2] for a review), and measures based on variance decomposition [24], game theory [12, 30], etc.”