Professional Documents
Culture Documents
Reliability Engineering and System Safety: Sciencedirect
Reliability Engineering and System Safety: Sciencedirect
Reliability Engineering and System Safety: Sciencedirect
Engineering Risk Analysis Group, Technische Universität München, München, 80333, Germany
Keywords: In reliability analysis with numerical models, one is often interested in the sensitivity of the probability of failure
Reliability-Oriented sensitivity analysis estimate to changes in the model input. In the context of multi-uncertainty, one whishes to separate the effect of
Multi-Uncertainty different types of uncertainties. A common distinction is between aleatory (irreducible) and epistemic (re-
Surrogate modelling ducible) uncertainty, but more generally one can consider any classification of the uncertain model inputs in two
Rare event simulation
subgroups, type A and type B. We propose a new sensitivity measure for the probability of failure conditional on
Decision support
type B inputs. On this basis, we outline a framework for multi-uncertainty-driven reliability sensitivity analysis.
A bi-level surrogate modelling strategy is designed to efficiently compute the new conditional reliability sen-
sitivity measures. In the first level, a surrogate is constructed for the model response to circumvent possibly
expensive evaluations of the numerical model. By solving a sequence of reliability problems conditional on
samples of type B random variables, we construct a level 2-surrogate for the logarithm of the conditional
probability of failure, using polynomial bases which allow to directly evaluate the variance-based sensitivities.
The new sensitivity measure and its computation are demonstrated through two engineering examples.
1. Introduction sampling methods (e.g., importance sampling [4–6], line sampling [7],
subset simulation [8] and sequential importance sampling [9,10]), both
Reliability analysis is concerned with the evaluation of the prob- of which can be combined with surrogate modelling (e.g., [11,12]).
ability of failure of an engineering system. The system can be described To account for different types of uncertainties (i.e., multi-un-
probabilistically in terms of the input random vector Θ with joint cu- certainty), one may consider a segmentation of the random input vector
mulative distribution function (CDF) FΘ and a deterministic model in two disjunct subsets = [ A , B ]T . The variables ΘB can be inter-
mapping each Θ to an output Y = ( ) . The performance of the system preted as epistemic (sometimes also termed reducible) and the variables
can be assessed in terms of its limit-state function g (LSF). The limit- in ΘA as aleatory (irreducible). Irrespective of the interpretation of ΘB,
state function defines the failure modes of a system and by convention we aim at explicitly quantifying its influence on (F ) . This is, for in-
takes values below 0 in the failure domain, which is a subset of the stance, useful to identify those inputs in ΘB whose uncertainty should
entire input variable space ΩΘ. The system probability of failure is be reduced by means of additional data in order to increase the accu-
given by [1] racy of the reliability analysis.
Fig. 1 illustrates our framework for reliability analysis and updating
(F ) = [I(g ( ) 0)] = I[g ( ) 0] ( )d ,
(1) in the presence of multi-uncertainty. To obtain information on the
potential influence of each component of ΘB on the reliability analysis,
where the indicator function I equals 1 on the failure domain {θ ∈ Ωθ: g
we consider the probability of failure conditional on ΘB [13]:
(Θ) ≤ 0} and 0 on its complement and πΘ is the joint probability
density function (PDF) of Θ. PF ( B ) = (F | B = B)
Standard Monte Carlo methods fail to efficiently solve Eq. (1) if = A [I(g ( A, B) 0)| B = B]
(F ) is very small, which is typically the case for failure probabilities.
Methods which are specifically designed to solve Eq. (1) for very small = I[g ( A, B) 0] A | B ( A | B )d A .
A (2)
(F ) are referred to as structural reliability methods (SRM). SRM can be
categorized into approximation (e.g., FORM, SORM [2,3]) and Note, that PF(θB) is a scalar function of the outcome of a random vector
⁎
Corresponding author.
E-mail address: max.ehre@tum.de (M. Ehre).
https://doi.org/10.1016/j.ress.2019.106726
Received 31 January 2019; Received in revised form 23 August 2019; Accepted 31 October 2019
Available online 08 November 2019
0951-8320/ © 2019 Elsevier Ltd. All rights reserved.
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Fig. 1. Framework for reliability sensitivity analysis in the presence of multi-uncertainty. The model is subject to irreducible (ΘA) and reducible (ΘB) probabilistic
inputs. Estimates of the distribution of the probability of failure conditional on ΘB, ^PF , may be obtained through repeated reliability analyses fixed at different θB.
Thus, ^PF is subject to change when updating components of ΘB. The proposed reliability sensitivities quantify the obtainable variability reduction in PF resulting
from uncertainty reduction (e.g., through Bayesian updating) in the components of ΘB. In this way, they allow for an informed selection of data acquisition measures.
and hence is itself a random variable. A similar way of handling un- indices for a probability of failure conditional on distribution para-
certainty separation is presented under the term ”second-order prob- meters using importance sampling. These indices bear some similarity
ability” in the program package Dakota [14] and is also discussed by to the definition of sensitivity indices used in the context of stochastic
other authors [13,15]. black boxes or stochastic emulators by, e.g., [34]. Here, a fraction of the
In the literature, a variety of metrics for quantifying sensitivities model uncertainty corresponding to ΘA is considered model-inherent or
with respect to different quantities of interest can be found. They can be latent and thus Sobol’ indices are defined for a conditional mean, which
grouped according to scope (local [16] vs. global importance [17,18]), retains only a dependence in ΘB and marginalizes over ΘA. Marrel et al.
moment-dependence (e.g., variance-based [17,19–21] vs. distribution- [34] compute Sobol’ indices for the response of such an inherently
based [18,22]) and considered inputs (deterministic parameters [23] stochastic model by constructing surrogates jointly for the model re-
vs. random variables [17,19–21]). A comprehensive overview over sponse mean and variance. Combining this idea with that of [22] - i.e.
global sensitivity measures is given in [24] and [25]. to analyze the indicator function - yields the reliability-oriented Sobol’
Most often, these measures are computed for the outcome of the indices discussed in [32,33]. Alternative sensitivity measures for rare
numerical model. In contrast, reliability-oriented sensitivity analysis is event probabilities include the use of quantiles [35] or perturbation of
concerned with determining the sensitivity of the output of a reliability input densities [36] to globally quantify influence of model inputs on
analysis to the model inputs. Many approaches dedicated to reliability- rare event probabilities.
oriented sensitivities are concerned with determining local derivative- In this contribution, we introduce a reliability-oriented sensitivity
based sensitivity measures for deterministic model parameters in the measure that is based on the variance decomposition of the logarithm of
limit-state function [23] or deterministic distribution parameters of the the conditional probability of failure, defined in Eq. (2). The proposed
input [15,26–28]. Alternatively, [29] computes failure probability measure enables the quantification of the influence of the type B (re-
histograms conditional on design parameters (LSF or distribution ducible inputs), which may consist of either input variables or uncertain
parameters) by artificially considering them as random variables. These distribution parameters thereof, on the magnitude of the probability of
histograms can be used to compute global sensitivity measures for the failure (Fig. 1). To efficiently compute the proposed reliability sensi-
parameters. When using approximative reliability methods like FORM/ tivities, we introduce a 2-level surrogate based approach. In the first
SORM, reliability sensitivities are obtained from the analysis in form of level, a surrogate of the model response is constructed and is further
the component (or α-) values of the design point (or most probable used to compute the probability of failure conditional on samples of the
point of failure) [3,30]. These can be interpreted as global, variance- type B-variables. The resulting conditional probability estimates are
based sensitivity indices of the first-(FORM) and second-(SORM) order further used to construct a polynomial basis surrogate model of the
Taylor expansions of g around the design point [30] in standard-normal logarithm of the conditional probability of failure. The coefficients of
space. Recently, [31] extended this idea to account for a mixture of the level-2 surrogate can be post-processed to compute an estimate of
several significant points and regions along the hypersurface g = 0 by the sought sensitivity indices.
means of a Gaussian mixture (GM) model through defining a partici- The structure of the paper is as follows: in Section 2, the basics of
pation factor for each of the regional design point contributions of the global sensitivity analysis are discussed and the new reliability sensi-
GM components. [22] compute variance-based sensitivities for the in- tivity index is introduced. Section 3 provides an introduction to poly-
dicator function I(g ≤ 0) and show, that this is equivalent to computing nomial basis surrogate modelling (polynomial chaos expansions (PCEs)
the moment-independent sensitivity measure proposed in [18] for the and low-rank approximations (LRAs)) and their relation to global sen-
probability of failure. They use a surrogate model to relax the compu- sitivity measures. Section 4 outlines a two-level framework for the
tational cost induced by the sampling-based sensitivity computation surrogate-driven estimation of the novel reliability sensitivities, which
approach. Along the same lines, [32] also use a surrogate (Kriging) is subsequently tested on two numerical examples in Section 5. The
modelling approach when computing variance-based sensitivities for studied examples consist of a 12-dimensional elastic truss structure and
the probability of failure conditional on uncertain distribution para- a 87-dimensional wind turbine monopile foundation in plastically be-
meters of the input variables. [33] compute variance-based sensitivity having soil. Section 6 contains a discussion of the presented method and
2
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
an outlook on possible extensions and improvements of the introduced reliability of a system one has to choose Q appropriately. In reliability
framework. analysis, the QOI is the failure event F and the associated probability of
failure. Since F is defined via the indicator function of the failure do-
2. Global reliability sensitivity analysis main, Li et al. [22] propose to compute importance rankings through
the variance decomposition of the indicator function I(g ≤ 0). They do
The sensitivity measures for reliability proposed here belong to the so by means of a surrogate modelling technique to avoid the slow
realm of variance-based sensitivity methods. After a brief introduction convergence of standard MC-estimators in cases where the QOI is a rare
of the underlying idea of variance-based sensitivity indices, their ap- event.
plication within reliability analysis is discussed. We discuss the com- Within the multi-uncertainty setting, we propose to perform the
putational cost of these newly introduced indices when estimated with variance decomposition of the conditional probability of failure defined
classical Monte Carlo-based (MC) approaches and thereby motivate a in Eq. (2). More precisely, we propose using
surrogate-driven computational approach, which is discussed in the Q = log PF ( B ), (7)
next section.
as the QOI, where log indicates the logarithm to the base 10. Q is
2.1. Variance-based sensitivity analysis measure for the magnitude of the conditional probability of failure. In
this way, we focus the sensitivity analysis on possibly substantial/
Consider a d-dimensional standard uniform random vector magnitude-altering changes in the estimate of P(ΘB). Note that in the
U d (0, 1), representing the input uncertainty of some model, and an mono-uncertainty case, in which ΘA is empty and all uncertainty is
output quantity of interest (QOI) Q = f (U ), with f : [0, 1]d . reducible, Q takes values 0 or . This obstructs the computation of
Variance-based sensitivity indices rely on estimating the contribution of the proposed index in such case. The concepts introduced in the fol-
a single (first-order index) or a combination of the inputs U (n-th order lowing may be generalized to any QOI that can be cast in terms of the
index, n being the number of variables considered in the combination) expectation of a functional (e.g., any moment of Y). Consider now an
to the output variance of the QOI. These fractions can be found by arbitrary subset of ΘB, which is denoted by , and its complement
projecting the QOI on an orthogonal functional basis, which leads to a such that B = { , } . The novel sensitivity indices for the variable
unique representation of f, namely the Sobol’–Hoeffding decomposition subset are given by
[17] [ [log PF | ]]
Slog PF , = ,
d d d [log PF ] (8)
f (U ) = f0 + fi (Ui ) + fij (Ui, Uj ) + +f12 d (U ).
i=1 i=1 j=i +1 (3) [ [log PF | ]]
T
Slog PF , = 1 .
Here, all basis functions have zero mean except for f0 = [Q]. If Q [log PF ] (9)
depends on pairwise independent inputs with arbitrary distribution
Substituting Eq. (2) in the above, one arrives at the following expres-
with known marginal CDFs F i , an isoprobabilistic transformation can
sions for the novel sensitivity indices:
be used to generalize Eq. (3). This transformation is given by
T: i F i ( i ), 1 i d, and the transformed decomposition is ob- [ [log{ A [I(g 0)| B ]}| ]]
Slog PF , =
tained by setting U = T ( ) in Eq. (3). By virtue of the orthogonality B [log{ A [I(g 0)| B ]}] (10)
property, the variance of f is merely the sum of the partial variances of
all the basis functions. The partial variance corresponding to a set of T [ [log{ A [I(g 0)| B ]}| ]]
Slog PF , = 1 .
inputs indicated by the index set {1, 2, , d} is found as the var- B [log{ A [I(g 0)| B ]}] (11)
iance of f . The corresponding Sobol’ index is then defined by the ratio
While these expressions appear cumbersome, they exhibit key features
[f ] of the new indices:
SQ, = .
[f ] (4)
The Sobol’ index measures the variance fraction that originates from a 1. The variance decomposition of the total variance contributed by ΘB
particular combination of variables exclusively without considering rather than Θ is performed, which is reflected by the normalizing
interaction of with , where ∼ denotes the complement of constants in Eqs. (10) and (11).
over the total set {1, 2, …, d} . Homma and Saltelli [19] defined the total- 2. Due to the expectation A, the employed QOI is smooth on B . In
effect indices, which measure the partial variances of all basis functions particular, it is non-binary as opposed to the QOI underlying the
that include : indices proposed by Li et al. [22].
3. The log-transformation of the conditional probability of failure fo-
[f ] cuses the sensitivity analysis on the probability of failure magnitude.
SQT, = .
[f ] (5)
A key motivation for defining the sensitivity metric based on log PF
Alternatively, both Sobol’ and total-effect indices can be cast in terms of is that the probability of failure and the resulting risk estimates are
variances of conditional expectations as [25]: typically interpreted in terms of their orders of magnitude [37,38].
[ [Q| ]] Additionally, the distribution of PF is heavily right-skewed and its
SQ, = , support covers multiple orders of magnitude, as opposed to the dis-
[Q]
[ [Q| ]] tribution of log PF which is supported on a single order of magnitude.
SQT, = This renders the estimation of the associated sensitivity indices more
[Q]
robust for the log-transformed QOI.1
[ [Q| ]]
= 1 . Given that the two measures will attribute different importance to
[Q] (6) the same variable, it is worthwhile discussing which QOI is the more
3
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
3. Polynomial basis surrogate modelling lection = {X , Y } is referred to as the training set. The data matrix Ψ
collects the evaluation of each of the multi-dimensional basis poly-
Let Θ be a random vector on the outcome space d with joint CDF nomials at each point in X
FΘ whose elements are mutually independent and Y = ( ), with = j (x
(i) ), 1 i ns , 1 j P, (20)
: d . If Y has finite mean-square, i.e., [ ( )2] < , then the
(i)
function belongs in a Hilbert space on which an inner product of where x is the i-th row of X. The vector of all P PCE-coefficients a is
any two functions g , h is defined as then obtained by regressing Y on Ψ which gives
4
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
a=( T ) 1 TY .
(21) r r
aa
j =1 j j
( z ij,0 z ij ,0)( (
mi
z z ))
k = 1 i jk i j k
S^Q,
j=1 i i
= r r d mi d
Eq. (18) indicates a fast growth of the OLS problem size with increasing j=1
aa
j =1 j j
(( i=1
( z z ))
k = 1 ijk ij k
( z z ))
i = 1 ij,0 ij ,0
dimension d. This motivates the use of sparse PCE methods, which are r r
aa (
d m
( k =i 1 z ijk z ij k ))
j =1 j =1 j j i=1
also applied in this work. Sparse PCE reduces P by penalizing the r r d mi d
,
number of terms in the PCE through solving a regularized least-squares j=1
a a (( i = 1 (
j =1 j j
z z )) ( i = 1 z ij,0 z ij ,0))
k = 1 ijk ij k
problem [41]. In this way, the method elicits a minimal number of basis (24)
functions such as to best explain the output variance.
T
S^Q, = 1
r r mi
aa ( z ij,0 z ij ,0)( ( z z ))
3.2. Canonical decomposition j =1 j =1 j j i i k = 1 i jk i j k
r r d mi d
j=1
aa
j =1 j j
(( i=1
( k=1
z ijk z ij k )) ( z z ))
i = 1 ij,0 ij ,0
Low-rank approximations (LRA) have been introduced originally to r r d m
aa ( ( k =i 1 z ijk z ij k ))
represent high-dimensional tensors by means of lower-dimensional j=1 j =1 j j i=1
.
r r d mi d
tensors [42]. A specific format of such approximations are canonical j=1
a a (( i = 1 (
j =1 j j
z z )) ( i = 1 z ij,0 z ij ,0))
k = 1 ijk ij k
decompositions, in which tensors are approximated by means of a
(25)
linear combination of products of one-dimensional tensors [43]. The
idea extends to continuous spaces where a multivariate function is In summary, Eqs. (23)–(25) provide a way of computing the var-
approximated by a linear combination of products of univariate func- iance fraction in a QOI caused by a subset of surrogate model inputs
tions: identified by . One can compute the effect of the variables in ex-
clusively (and call this the Sobol’ index S^ ) or including any interactions
^ LRA ( ) =
r d mi
(i )
of these variables with the remainder of the inputs (and call this the
aj z ijk ( i), 0 k mi , 1 i d. T
j=1 i=1 k=0
k
(22) total-effect index S^ ).
Therein, an additional set of coefficients z appears, which can be effi- 4. Conditional surrogate-based reliability sensitivities
ciently determined by solving reduced, univariate least squares pro-
blems over the directions i = 1, …, d repeatedly (while keeping all re- The computation of sensitivity indices via polynomial surrogates
maining directions constant in each step; this is often referred to as requires the QOI to be sufficiently smooth. In particular, any attempts
alternating least squares). In a second step, the coefficients a are de- to obtain surrogate-based indices of the indicator function of the failure
termined via OLS. A detailed description of the procedure is given in domain I(g ≤ 0) directly in such a manner must fail due to the dis-
[44] and [45]. The number of coefficients in the canonical decom- continuity in I(g ≤ 0). However, the log-transformed conditional
position is ((m + 1)· d· r ) assuming a constant polynomial degree m in probability of failure is continuous in the space of ΘB so that one may
all dimensions. This linearity in d is remarkable and explains the ad- compute the proposed sensitivity indices with polynomial surrogates.
vantage this format offers over classical PCE where the coefficients To this end, we devise a two-level surrogate modelling procedure.
grow factorially with the dimension as described in Eq. (18). Building a surrogate of log PF (level 2) requires an experimental design
that consists of samples of ΘB and the associated probabilities of failure
given each of these samples. That is, one has to solve n2 reliability
3.3. Surrogate-based sensitivity indices problems, where n2 is the experimental design size for the final surro-
gate. An auxiliary (level 1) surrogate is built for the actual model ,
Both PCEs and LRAs can be used to infer first-order and total sen- based on which the reliability computations can be conducted. This
sitivity indices directly from the computed model coefficients. Rather approach is referred to as global in the following. Alternatively, a local
than searching estimates of the expressions in Eqs. (6), the similarity of surrogate model may be constructed at each sample of ΘB for solving its
the underlying orthogonal Sobol’ decomposition in Eq. (3) with Eqs. associated reliability problem. The local surrogate has lower-dimen-
(19) and (22) is exploited and the expressions (4) and (5) can be sional input compared to the global surrogate because the input space is
computed directly[46]. showed that the Sobol’ decomposition of the reduced from ΩΘ to A , and is therefore cheaper to determine; how-
PCE is readily obtained by collecting any multi-dimensional orthogonal ever, it has to be recomputed for each ΘB-sample. This approach will be
polynomials depending on identical variable subsets into f ( ) . referred to as local. Prior to construction, an isoprobabilistic transfor-
Therefore, computing the partial variance of the PCE model associated mation to an independent standard-normal space is used, such that both
with a subset of variables amounts to summing the squared coef- PCEs and LRAs can be consistently constructed using the orthogonal
ficients of the respective multi-dimensional basis polynomials in which polynomial basis with respect to the standard normal probability
the elements of occur (exclusively for Sobol’ indices and collectively measure, which is the Hermite polynomial basis [39].
for total indices). The same concept can be applied to LRAs even though
the compressed format (product) renders the evaluation somewhat 4.1. Level 1
more tedious. For a given subset of the input variables denoted by the
index set , {0, 1}d is a boolean index vector s.t. i = 0 if i On the first level, the goal is to construct a surrogate model for the
and i = 1 if i . In the same way, for the j-th row of αj, i j = 0 if original model ( ), describing the engineering system.
ij = 0 and i = 1 if αij > 0. Then, the PCE-based sensitivity indices
j
5
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
sensitivity estimates mostly depends on the quality of the surrogate and obtain larger experimental designs at no additional computational cost
the applied SRM in level 1. Thus, tuning the method in this component by sampling [Θ1, M, Σ] from iso-Θ1-surfaces, as done in [50]. Then, the
will yield the most substantial improvements in estimating the relia- PCE will still approximate the exact relationship amongst Θ1, M and Σ,
bility indices. Echard et al. [11] introduced a Kriging-driven Monte- though at considerably better accuracy through the larger training set.
Carlo sampling approach, which enriches the Kriging experimental The predictive CDF of Θ1, F 1, can be computed via numerical in-
design according to a learning function that favors large model un- tegration, e.g., Monte-Carlo integration:
certainty close to the limit-state face g = 0 and [48] applied a similar
idea to PCEs. A recently introduced surrogate-driven sequential sam- F 1 ( 1) = F 1| M , ( 1 |m , ) M (m ) ( )dmd
pling approach for reliability analysis explores the failure domain se-
quentially and reconstructs a surrogate model at each intermediate step = [F 1|M , ( 1 |M , ) ]
in the sampling procedure [49]. All of these approaches have been N
1
shown to substantially improve the unconditional (predictive) relia- F 1 |M , ( 1 |mi , i ), mi M,
i .
N (28)
bility estimate. However, using such an active learning-strategy for the i=1
lowing hierarchical sampling strategy yields a sample from Θ1 based on For each ΘB-sample, use a structural reliability method and the
global level 1-surrogate-based limit-state function g^ to compute
a sample u = [u1, u2 , u3]T from the standard-uniform distribution.
(i )
m = FM1 (u1), P^F = A [I(g
^( A,
(i )
B ) 0)| (i )
B ]
I(g^ ( A ,
1 (i ) (i )
= F (u2 )), = B ) 0) A| B ( A | B )d A .
1 1 A
1 = F 1 |M ,
(u3 |m , ) = F 1 |M ,
(u3 |FM1 (u1), F 1
(u2)). (26)
Since the surrogate is constructed in standard-normal space, an iso- b) Local
probabilistic transformation is used to compute the corresponding For each ΘB-sample, sample n1 ΘA-samples and evaluate the
standard-normal sample v1: model at these samples. Based on these, build the local level 1-
(i ) ^ (i )
v1 = 1 [F
1 ( 1)], (27) surrogate over A conditional on B , | B and its associated
^
local limit-state function g^ ( A) to compute | B(i)
where Φ( · ) is the standard-normal CDF. Eqs. (26) and (27) facilitate the i
6
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Fig. 2. Flow diagram of 2-level surrogate-based conditional reliability sensitivities with global surrogate in level 1, where the dashed boundary marks the starting
point.
Fig. 3. Flow diagram of 2-level surrogate-based conditional reliability sensitivities with local surrogates in level 1, where the dashed boundary marks the starting
point.
log^PF ( B).
the local surrogate strategy, the cost in level 1 is given by
^ n1, local = 2(dA + 1) per ΘB-sample. The total cost thus reads
5. Obtain variance-based sensitivity indices of log PF by means of the
nlocal = n1, local ·n2 . When using no level 1-surrogate at all, n2 reliability
model coefficients.
problems are solved using the original model . In order to estimate
the computational cost in the surrogate-free approach, the number of
The procedure outlined above is also sketched in Figs. 2 (global) and 3
required -evaluations to solve a dA-dimensional reliability problem,
(local).
n1, , has to be estimated. Therein, dA influences which method should
be used to solve the problem. Table 1 provides a qualitative relationship
4.4. Computational cost between the dimensionality and computational cost of a dA-dimensional
reliability problem (n1, in Fig. 4).
Fig. 4 shows the behaviour of computational cost in both levels The total cost in this approach is computed as n = n1, · n2 .
when using either a global or a local or no surrogate strategy at all in Fig. 4 (right-hand side) shows, that both the local surrogate and
level 1. The computational cost is measured in terms of number of surrogate-free strategy yield the highest overall computational cost
-evaluations. To proceed, we make the assumption that the number of when dB and dA have similar size, i.e., neither reducible nor irreducible
samples required to learn a surrogate model in input dimension d is uncertainties dominate the model input. This is due to two counter-
n = 2·(d + 1) . Note that this is a rather crude assumption and is accu- acting effects: On the one hand, as the effective local surrogate model
rate only for mildly nonlinear models . The number of samples re- dimension equals dA, the local strategy requires less samples per sur-
quired in level 2, n2 = 2(dB + 1), is the same in all three versions, rogate when dA is small. Similarly, if not using a surrogate at all in level
though for the global surrogate model, this does not influence the 1, the reliability analysis performed with the original model will require
computational cost. At fixed model dimension d, the global surrogate less model evaluations due to the decreasing dA. On the other hand,
modelling costs are n1, global = nglobal = 2(d + 1), i.e., they are constant, decreasing dA implies an increasing reducible space dimension dB. Thus,
depend only on level 1 and the total dimension and not on dA or dB. For
7
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Fig. 4. Comparison of global (global) vs. local (local) vs. no ( ) surrogate strategy in level 1 in terms of computational cost (number of -calls) over reducible space
dimension dB. Left: cost per level, right: total cost.
8
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Table 2 should depend on one another reciprocally; the larger the index the
Input variable definitions of the elastic truss. more accurate it is estimated based on a given set of samples. Indeed,
Quantity Distribution Mean μ CoV δ relative to their respective magnitudes, the indices of the most im-
portant indices aP and bP are estimated more accurately.
2
A1 [m ] Log-Normal 2·10 3 0.1
A2 [m2] Log-Normal 1·10 3 0.1
5.2. Monopile foundation
E1, E2 [Pa] Log-Normal 2.1 · 1011 0.1
P1 - P6[N] Gumbel µP|aP, bP P|aP , bP
aP [N] Log-Normal 46624 0.2 In this example, we consider a finite element model for the inter-
bP [N] Log-Normal 3375 0.2 action of a monopile foundation of an offshore wind turbine (Fig. 8)
with stiff, plastic soil. Deterministic parameters of the monopile are its
depth L = 30 m, diameter D = 6 m, wall thickness t = 0.07 m, Poisson
repeated 20 times redrawing random level 1-experimental designs ratio = 0.3 and Young’s modulus E = 2.1·105 MPa. The uncertain in-
which yields the estimator statistics provided in Figs. 6 and 7. These are puts comprise the lateral load H as well as the undrained shear strength
computed for the difference of the surrogate-based estimator from the s of the soil and hyperparameters of both quantities. The engineering
direct Monte-Carlo (DMC) reference solution, i.e., model setup follows [53] and the probabilistic model considered there
has been modified following [54].
Q =Q QDMC . (31) s is considered both uncertain and increasing in mean with soil
depth z. It is thus modelled by a random field with linear mean drift
All reliability analyses haven been performed with the first-order re-
along the soil depth coordinate z. Given an underlying stationary
liability method (FORM). For the reference solution, nDMC = 2·105
Gaussian random field s˜ (z, )
samples have been used implying the solution of 8 · 105 reliability
problems of dimension 6 (since the type B-variable space has dimension {˜s (z ): 0 z L} (0, s̃ ),
6, see Section 2.3).
the non-stationary random field representing the shear strength of the
While both PCE and LRA-based approaches capture the variable
soil can be expressed as
importance ranking correctly, the LRA-based approach performs
slightly but consistently better in the mean compared to the PCE-based s (z , ) = s0 + s1 (z )exp{˜s (z, )}
approach. The least important variables E2, A2 are estimated with the = s0 + s1 z exp{˜s (z , )},
smallest error mean and variance, which, however, is due to their small
true magnitude. Intuitively, estimation accuracy and index magnitude where γ is the soil unit weight, (z ) = z is the effective vertical stress,
Fig. 6. log PF first-order Sobol’ indices: mean estimates and errors (n1 = 200, n2 = 1000 ).
9
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Fig. 7. log PF Total-effect indices: mean estimates and errors (n1 = 200, n2 = 1000 ).
s0 is the undrained shear strength at ground level and s1 is the drift 2|z z |
˜˜ (z , z ) = exp ,
parameter governing the mean increase of s with increasing soil depth. ss
s˜
s˜ (z, ) models the intra-site variability. That is, at a given site with
known deterministic s0 and s1, it describes the inherent variability of with vertical soil scale of fluctuation s̃ = 1.9 m [55] and standard
the undrained shear strength. In order to describe the inter-site un- deviation s̃ = 0.3 [55,56]. We assume the soil to be stiff and plastic
certainty in s, the parameters s0 and s1 are modeled probabilistically as according to the classification provided in [57]. There, the specific soil
well. The stationary RF s̃ is taken to be correlated with exponential-type weight range is given with 17 19 kN/m3, whence we set
= 18 kN/m3 . The mean cohesion range is given with 20 50 kN/m2
by Rackwitz [57] while Cao et al. [58] lists the mean range of the
H undrained shear strength ratio su/σ′ as 0.23 1.4 . We fit log -normal
distributions for s0 and s1 by setting the 10% and 90% quantiles of the
z=0 distributions equal to the lower and upper bounds of these ranges. The
resulting parameters are detailed in Table 3 along with uncertain
parameters for the load H, namely μH and σH. The mean and coefficient
p of variation of the load Gumbel distribution in Table 3 are conditional
on the parameters aH (location parameter) and bH (scale parameter)
D, t, L, ν, E according to Eqs. (29) and (30).
s̃ is simulated by means of the midpoint method. That is, the spatial
y
Table 3
Input variable definitions of the monopile foundation.
10
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
domain [0, L] is discretized with n spatial elements and s̃ is represented solved using a bisection- or Newton-procedure due to the monotonicity
by means of n random variables with joint distribution (0 , ss˜˜ ) . The of the limit-state-function the same SRM is applied for the reference
random variables represent the random field values at the element solution with the original model ). Accurate estimates of the condi-
midpoints. Thus, the covariance matrix ss ˜˜ is computed by evaluating tional reliability sensitivities are achieved with an overall 2000 eva-
˜˜ (z , z ) at the element midpoints. The number of elements is luations of the limit-state-function for the second example with the
2
s˜ ss
chosen such that 95% of the inherent RF variability is captured by the introduced framework. This is a conservative choice and satisfying
RF discretization, leading to n = 82 in this example. accuracy may be achieved with a considerably lower number of model
Therefore, the total input dimension is d = 87 . As the surrogate evaluations as is evident from Figs. 11 and 12.
modeling approach requires independent inputs, the midpoint random The algorithm is run 20 times redrawing level 2-experimental de-
variables are transformed to independent standard normal random signs to compute errors ϵ arising from exploring B randomly. These
variables, denoted as ξ, by means of the Nataf transform. The model errors are measured against a DMC-based reference solution according
output Y = ( ) is the maximum occurring stress in the foundation. to Eq. (31). The reference solution is based on ns = 20000 independent
The limit-state function is given by samples which yields a total of 1.64 · 106 reliability problems since
dB = 86. A single reference solution (fixed σcrit) obtained with a Matlab
g( ) = crit ( ),
implementation took approximately 48 h to complete on 10 Intel Xeon
where three different stress thresholds crit = {80MPa, 100MPa, 120MPa} E5-2697 v3 14-core nodes, which emphasizes the need for surrogate
are considered with corresponding system failure probabilities modelling-based estimators. In order to assess the random field influ-
(F ) = {3.0·10 3, 3.6·10 4, 8.0·10 5} (estimated with MC and ence, the sum of all first-order indices of the elements of ξ are reported
CoV = {0.0409, 0.1187, 0.2500} ). For this example, we consider a sce- for both the reference solution and the surrogate-based approach. This
nario in which uncertainties about all the input hyperparameters as is motivated by the impossibility of efficiently estimating a sensitivity
well as the inherent variability of the shear strength can be reduced index of order 86 with the MC-methods of [20,21]. However, the sur-
through additional data, i.e., B = [ , aH , bH , s0, s1]. This leaves the rogate-based estimators indicate negligible interaction of the elements
inherent load variability as the only remaining aleatory input, i.e., of ξ (all Sobol’ and total-effect indices are virtually identical) thus jus-
A = H . This example has proven extremely challenging for common tifying this approximation.
global surrogate models (polynomial basis surrogates, adaptive kriging The sensitivities depicted in Fig. 9 and 10 summarize the effect of
surrogates). Therefore, the local surrogate modelling strategy is chosen, the inherent random field variability in the order-n-indices of the
which, in this case, is an efficient choice since the limit-state function to random vector ξ.
be approximated at each ΘB-sample is one-dimensional. As little as four The PCE and LRA level 2-surrogates yield similar estimates for the
training points are necessary per ΘB-sample. Based on the local surro- reliability sensitivities (Figs. 9 and 10): The load dispersion parameter
gate, the corresponding one-dimensional reliability problem can be bH is identified as the most influential input to the monopile reliability
Fig. 9. log PF first-order Sobol’ indices: mean estimates and errors (n1 = 4 (local), n2 = 2000 ).
11
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Fig. 10. log PF Total-effect Sobol’ indices: mean estimates and errors (n1 = 4 (local), n2 = 2000 ).
analysis across all investigated scenarios. The dominance of bH becomes categories, e.g., aleatory and epistemic. The interest then lies in ex-
more pronounced at higher critical stress levels. PCE-based sensitivity pressing the sensitivity of the probability of failure-estimate to the
estimate means are in slightly better agreement with the reference so- epistemic inputs. We devise a flexible two-level surrogate modelling
lution than LRA-based estimates. Moreover, PCE-based estimates ex- approach which allows for a cost-efficient estimation of the proposed
hibit consistently smaller variability resulting from the random level 2- index. The approach relies on either building a global surrogate model
experimental design. Error means and variabilities are of comparable over the entirety of input variables once and for all (global strategy) or
magnitude for all computed indices in this example. However, as dis- on repeatedly recomputing cheaper, lower-dimensional surrogates
cussed for the truss example, indices of larger magnitude should be (local strategy). An analysis of computational cost for both local and
estimated more accurately at a given amount of information. This is global surrogate modelling strategy has been carried out in dependence
true when considering the estimation error relative to the index mag- on how the model input is divided in reducible and irreducible.
nitudes. Then, estimates of the most influential variables aH and bH We demonstrate the novel approach by means of two examples. In
exhibit comparably small variability due to the random level 2-design the elastic truss example, the sensitivity indices were estimated using a
of experiments while the less important variables’ estimators prove global low-rank approximation in level 1 at a total cost of 200 original
more sensitive in this respect. In Figs. 11 and 12, the evolution of the model evaluations. The LRAs are more suitable for global surrogate-
error means and variances at different level 2-experimental design sizes driven reliability computations (level 1) compared to the PCEs, as
n2 are depicted. The error converges in mean and standard deviation as discussed in [45]. For the geotechnical example, namely a monopile
n2 increases. The error mean apparently is bounded from below which foundation, a local surrogate modelling approach based on PCEs has
is likely due to a bias in the reference solution which arises as the MC- been adopted in level 1 yielding a total cost of 8000 original model
estimates are not fully converged at nc = 20000 independent samples. evaluations. In level 2, the sparse PCEs perform slightly better than the
Ultimately, if a decision-maker were to choose whether to acquire data LRA. Sensitivity estimates for both examples were validated with
on either of the uncertain model inputs and on which input in parti- sampling-based reference estimates showing that the important input’s
cular, they should choose bH according to the introduced reliability indices are estimated accurately by the method. While in this paper
sensitivity framework. only two possible choices for the level 1 surrogate are discussed, vir-
tually any surrogate modelling technique could be applied here.
6. Concluding remarks
6.1. Discussion
In this paper we describe a framework for reliability sensitivity
analysis whose core is a novel, variance-based sensitivity index tailored The contribution of the proposed sensitivity index is threefold. First,
to reliability analysis in the presence of multi-uncertainty. Multi-un- it focusses the sensitivity analysis on a variable subset that is of interest
certainty refers to a separation of uncertain model inputs in different in the presence of multi-uncertainty and data assimilation applications.
12
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
Fig. 11. Sobol’ (left column) and total-effect indices (right column) error mean vs. level 2 experimental design size n2 for PCE- (upper row) and LRA-based (lower
row) at crit = 100 MPa.
Fig. 12. Sobol’ (left column) and total-effect indices (right column) error standard deviation vs. level 2 experimental design size n2 for PCE- (upper row) and LRA-
based (lower row) at crit = 100 MPa.
Second, it represents a direct sensitivity measure for the probability of numerical examples.
failure magnitude as opposed to indices based on the indicator function The problem dimension, which can be handled by the approach is
that rank influence on the failure hypersurface shape. Finally, the new guided by the surrogate modelling techniques chosen in both levels.
index facilitates the entirely surrogate-driven computation of reliability Using an arbitrary surrogate model that is capable of addressing high-
sensitivities by smoothening the indicator function discontinuity dimensional problems in level 1 and LRAs or the recently introduced
through an integral formulation. PLS-driven PCEs [59] in level 2, it is applicable up to several thousand
In accordance with intuition, estimators of sensitivity indices with input variables.
small magnitude exhibit smaller errors compared to those of large At a fixed accuracy (with respect to -model output), the cost of the
magnitude. Relative to the magnitude, the opposite is true: at fixed global surrogate modelling strategy remains constant irrespective of the
amount of information (here: the level 1-training set size), the sensi- fraction of reducible and irreducible uncertainty in the model input.
tivity indices with larger magnitude are estimated more reliably. The same is not true for the local strategy where two counteracting
Moreover, the variable importance ranking is accurately captured in all effects lead to the local strategy being most efficient if either of the
13
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
uncertainty types dominates the model input. The local modelling computational cost. Next to the choice of surrogate, the level 1-per-
strategy is more flexible and expected to perform well even for strongly formance would likely benefit from a more guided selection of experi-
nonlinear models, when the global modelling strategy may deteriorate. mental design points in level 1, which addresses the method’s re-
In such case, the required number of samples for a sufficiently accurate quirement of surrogates to remain accurate close to g = 0 rather than
global surrogate model may overcompensate the strategy-based savings minimize global error measures.
the global modelling strategy offers and render the local strategy the Moreover, the introduced sensitivity measure may be connected to
better choice. Therefore, the model under consideration should govern decision-oriented sensitivity analysis and in particular the concept of
the decision which of either approach is used. Generally, we re- expected value of partial perfect information (EVPPI). It is known that
commend to apply the local surrogate modelling strategy when the the classical first-order variance-based sensitivity measure coincides
model under consideration exhibits significant nonlinear behaviour. with the EVPPI for a special decision context (with a quadratic loss
function) [60]. Based on this, the proposed framework may be adapted
6.2. Outlook to facilitate computation of more relevant loss functions associated
within decision analysis, which in turn promotes its applicability for
The experimental design for the level 1-surrogate dominates the decision support.
computational cost of the approach. The accuracy of the overall method
is controlled by the capability of the level 1-surrogate to capture the tail Acknowledgment
behaviour of the model response and that of the level 2-surrogate to
capture the second moment of its derived response (the conditional This work was supported by the Deutsche Forschungsgemeinschaft
probability of failure). Thus, the level 1-surrogate represents the most (DFG) in the framework of the priority programme SPP 1886 by grant
crucial element of the framework both with respect to accuracy and STR 1140/6-1.
Appendix A. Connecting the proposed Sobol’ index to its indicator function-based counterpart
Here, we discuss the connection of the novel sensitivity indices to the measures proposed by Li et al. [22]. In particular, we show that the Sobol’
index of the condititonal probability of failure (without logarithmic transformation) is identical to the Sobol’ index of the indicator function I(g ≤ 0)
up to a normalizing constant. In this case, the QOI reads
Q = PF ( B ) = A [I(g ( A, B) 0)| B = B ],
while in [22], Q = I(g ( ) 0) is used. Consider now an arbitrary subset of ΘB which is denoted by and its complement such that
B = { , } . The Sobol’ index of Q = PF for the variable subset is given by
[ [ A [I(g 0)| B ]| ]]
SPF , = ,
B[ A [I(g 0)| B ]]
1
SPF , = ( | ) I(g ( A , , ) 0) ( A| , )d Ad
[PF ] A
1
= I(g ( A, , ) 0) ( A, | )d A,
[PF ] A,
14
M. Ehre, et al. Reliability Engineering and System Safety 195 (2020) 106726
optimization using bridge importance sampling. Probab Eng Mech 2013;34:48–57. [37] Calman SKC, Royston G. Personal paper: risk language and dialects. BMJ
[11] Echard B, Gayton N, Lemaire M. AK-MCS: an active learning reliability method 1997;315(7113):939–42. https://doi.org/10.1136/bmj.315.7113.939.
combining kriging and Monte Carlo simulation. Struct Saf 2011;33(2):145–54. [38] Levine E. Improving risk matrices: the advantages of logarithmically scaled axes. J
https://doi.org/10.1016/j.strusafe.2011.01.002. http://www.sciencedirect.com/ Risk Res 2012;15(2):209–22. https://doi.org/10.1080/13669877.2011.634514.
science/article/pii/S0167473011000038 [39] Xiu D, Karniadakis GE. The wiener–askey polynomial chaos for stochastic differ-
[12] Sudret B. Meta-models for structural reliability and uncertainty quantification. In: ential equations. SIAM J Scientif Comput 2002;24(2):619–44.
Phoon KK, Beer M, Quek ST, Pang SD, editors. Proc. Fifth Asian-Pacific Symposium [40] Berveiller M, Sudret B, Lemaire M. Stochastic finite element: a non intrusive ap-
on Structural Reliability and its Applications (5APSSRA), Singapore, 2012. 2012. proach by regression. Eur J Comput Mech 2006;15(1–3):81–92.
[13] Der Kiureghian A, Ditlevsen O. Aleatory or epistemic? does it matter? Struct Saf [41] Blatman G, Sudret B. Adaptive sparse polynomial chaos expansion based on least-
2009;31(2):105–12. angle regression. J Comput Phys 2011;230(6):2345–67. https://doi.org/10.1016/j.
[14] Adams B., Bauman L.E., Bohnhoff W., Dalbey K., Ebeida M., et al. Dakota, a mul- jcp.2010.12.021.
tilevel parallel object-oriented framework for design optimization, parameter esti- [42] Grasedyck L, Kressner D, Tobler C. A literature survey of low rank tensor approx-
mation, uncertainty quantification, and sensitivity analysis: user’s manual, imation techniques. GAMM-Mitteilungen 2013;36(1):53–78.
SAND2014-4633. Sandia National Laboratories; November 2015. [43] Hitchcock FL. The expression of a tensor or a polyadic as a sum of products. J Math
[15] Chabridon V, Balesdent M, Bourinet J-M, Morio J, Gayton N. Reliability-based Phys 1927;6(1–4):164–89.
sensitivity estimators of rare event probability in the presence of distribution [44] Chevreuil M, Lebrun R, Nouy A, Rai P. A least-squares method for sparse low rank
parameter uncertainty. Reliab Eng Syst Saf 2018;178:164–78. approximation of multivariate functions. SIAM/ASA J Uncertain Quantif
[16] Papaioannou I, Breitung K, Straub D. Reliability sensitivity analysis with monte 2015;3(1):897–921.
carlo methods. ICOSSAR 2013. New York City: University of N.Y.C.; 2013. [45] Konakli K, Sudret B. Polynomial meta-models with canonical low-rank approx-
[17] Sobol’ I. Sensitivity estimates for nonlinear mathematical models. Math Model imations: numerical insights and comparison to sparse polynomial chaos expan-
Comp Exp 1993;1:407–14. sions. J Comput Phys 2016;321:1144–69.
[18] Borgonovo E. A new uncertainty importance measure. Reliab Eng Syst Saf [46] Sudret B. Global sensitivity analysis using polynomial chaos expansions. Reliab Eng
2007;92(6):771–84. Syst Saf 2008;93(7):964–79.
[19] Homma T, Saltelli A. Importance measures in global sensitivity analysis of non- [47] Konakli K, Sudret B. Global sensitivity analysis using low-rank tensor approxima-
linear models. Reliab Eng Syst Saf 1996;52(1):1–17. tions. Reliab Eng Syst Saf 2016;156(Supplement C):64–83. https://doi.org/10.
[20] Jansen MJ. Analysis of variance designs for model output. Comput Phys Commun 1016/j.ress.2016.07.012.
1999;117(1):35–43. [48] Marelli S, Sudret B. An active-learning algorithm that combines sparse polynomial
[21] Saltelli A, Annoni P, Azzini I, Campolongo F, Ratto M, Tarantola S. Variance based chaos expansions and bootstrap for structural reliability analysis. Struct Saf
sensitivity analysis of model output. design and estimator for the total sensitivity 2018;75:67–74.
index. Comput Phys Commun 2010;181:259–70. [49] Papaioannou I, Ehre M, Straub D. Efficient PCE representations for reliability
[22] Li L, Lu Z, Feng J, Wang B. Moment-independent importance measure of basic analysis in high dimensions. In: Song J, editor. Proceedings of the 19th working
variable and its state dependent parameter solution. Struct Saf 2012;38:40–7. conference of the IFIP Working Group 7.5 on Reliability and Optimization of
[23] Papaioannou I, Breitung K, Straub D. Reliability sensitivity estimation with se- Structural Systems. ETH Zürich; 2018.
quential importance sampling. Struct Saf 2018;75:24–34. [50] Schöbi R, Sudret B. Global sensitivity analysis in the context of imprecise prob-
[24] Iooss B, Lemaître P. A review on global sensitivity analysis methods. In: Meloni C, abilities (p-boxes) using sparse polynomial chaos expansions. Reliab Eng Syst Saf
Dellino G, editors. Uncertainty management in Simulation-Optimization of 2019;187:129–41. Sensitivity Analysis of Model Output
Complex Systems: Algorithms and Applications. Springer; 2015. [51] Lee SH, Kwak BM. Response surface augmented moment method for efficient re-
[25] Saltelli A, Chan K, Scott E. Sensitivity analysis. John Wiley & Sons, Inc.; 2000. liability analysis. Struct Saf 2006;28(3):261–72.
[26] Wu Y-T. Computational methods for efficient structural reliability and reliability [52] Ehre M, Papaioannou I, Straub D. Efficient estimation of variance-based reliability
sensitivity analysis. AIAA J 1994;32(8):1717–23. sensitivites in the presence of multi-uncertainty. In: Song J, editor. Proceedings of
[27] Song S, Lu Z, Qiao H. Subset simulation for structural reliability sensitivity analysis. the 19th working conference of the IFIP Working Group 7.5 on Reliability and
Reliab Eng Syst Saf 2009;94(2):658–65. Optimization of Structural Systems. ETH Zürich; 2018.
[28] Dubourg V, Sudret B. Meta-model-based importance sampling for reliability sensi- [53] Depina I, Papaioannou I, Straub D, Eiksund G. Coupling the cross-entropy with the
tivity analysis. Struct Saf 2014;49:27–36. Special Issue In Honor of Professor Wilson line sampling method for risk-based design optimization. Struct Multidiscip Optim
H. Tang 2017;55(5):1589–612.
[29] Au S. Reliability-based design sensitivity by efficient simulation. Comput Struct [54] Jiang S-H, Papaioannou I, Straub D. Bayesian updating of slope reliability in spa-
2005;83(14):1048–61. tially variable soils with in-situ measurements. Eng Geol 2018;239:310–20.
[30] Hohenbichler M, Rackwitz R. Sensitivity and importance measures in structural [55] Phoon K-K, Kulhawy FH. Characterization of geotechnical variability. Can Geotech
reliability. Civil Eng Syst 1986;3(4):203–9. J 1999;36(4):612–24.
[31] Kim T, Song J. Generalized reliability importance measure (GRIM) using gaussian [56] Phoon K-K, Kulhawy FH. Evaluation of geotechnical property variability. Can
mixture. Reliab Eng Syst Saf 2018;173(C):105–15. Geotech J 1999;36(4):625–39.
[32] Wang P, Lu Z, Tang Z. An application of the kriging method in global sensitivity [57] Rackwitz R. Reviewing probabilistic soils modelling. Comput Geotech
analysis with parameter uncertainty. Appl Math Model 2013;37(9):6543–55. 2000;26(3):199–223.
[33] Morio J. Influence of input PDF parameters of a model on a failure probability [58] Cao Z, Wang Y, Li D. Quantification of prior knowledge in geotechnical site char-
estimation. Simul Modell Pract Theory 2011;19(10):2244–55. acterization. Eng Geol 2016;203:107–16.
[34] Marrel A, Iooss B, Da Veiga S, Ribatet M. Global sensitivity analysis of stochastic [59] Papaioannou I, Ehre M, Straub D. Pls-based adaptation for efficient pce re-
computer models with joint metamodels. Stat Comput 2012;22(3):833–47. presentation in high dimensions. J Comput Phys 2019;387:186–204. https://doi.
[35] Maume-Deschamps V, Niang I. Estimation of quantile oriented sensitivity indices. org/10.1016/j.jcp.2019.02.046. http://www.sciencedirect.com/science/article/
Stat Probab Lett 2018;134:122–7. pii/S0021999119301639
[36] Lemaitre P, Sergienko E, Arnaud A, Bousquet N, Gamboa F, Iooss B. Density [60] Oakley JE. Decision-Theoretic sensitivity analysis for complex computer models.
modification-based reliability sensitivity analysis. J Stat Comput Simul Technometrics 2009;51(2):121–9.
2015;85(6):1200–23.
15