Space-partition method for the variance-based sensitivity analysis: Optimal partition scheme and comparative study

Published on Nov 1, 2014in Reliability Engineering & System Safety4.039
· DOI :10.1016/j.ress.2014.06.013
Qingqing Zhai14
Estimated H-index: 14
(Beihang University),
Jun Yang13
Estimated H-index: 13
(Beihang University),
Yu Zhao10
Estimated H-index: 10
(Beihang University)
Abstract Variance-based sensitivity analysis has been widely studied and asserted itself among practitioners. Monte Carlo simulation methods are well developed in the calculation of variance-based sensitivity indices but they do not make full use of each model run. Recently, several works mentioned a scatter-plot partitioning method to estimate the variance-based sensitivity indices from given data, where a single bunch of samples is sufficient to estimate all the sensitivity indices. This paper focuses on the space-partition method in the estimation of variance-based sensitivity indices, and its convergence and other performances are investigated. Since the method heavily depends on the partition scheme, the influence of the partition scheme is discussed and the optimal partition scheme is proposed based on the minimized estimator׳s variance. A decomposition and integration procedure is proposed to improve the estimation quality for higher order sensitivity indices. The proposed space-partition method is compared with the more traditional method and test cases show that it outperforms the traditional one.
  • References (37)
  • Citations (17)
📖 Papers frequently viewed together
54 Citations
31 Citations
999 Citations
78% of Scinapse members use related papers. After signing in, all features are FREE.
#1Rui Peng (USTB: University of Science and Technology Beijing)H-Index: 18
#2Huadong Mo (CityU: City University of Hong Kong)H-Index: 7
Last. Gregory Levitin (Israel Electric Corporation)H-Index: 43
view all 4 authors...
Due to imperfect fault coverage, the reliability of redundant systems cannot be enhanced unlimitedly with the increase of redundancy. Thus it is essential to study the optimal structure of redundant systems. This paper considers a multi-state series-parallel system with two types of parallelization: redundancy and work sharing. Different from existing works which consider single-fault coverage, multi-fault coverage is considered in order to adapt to a wider range of fault tolerant mechanisms. Fo...
28 CitationsSource
#1Elmar Plischke (TUC: Clausthal University of Technology)H-Index: 10
#2Emanuele Borgonovo (Bocconi University)H-Index: 26
Last. Curtis Smith (INL: Idaho National Laboratory)H-Index: 13
view all 3 authors...
153 CitationsSource
#1Elmar Plischke (TUC: Clausthal University of Technology)H-Index: 10
We consider correlation ratios as estimators for first order sensitivity indices from given data. The computation is simplified by the introduction of the cumulative sum of the normalised reordered output. Ideas for the estimation using interpolation are also discussed.
19 CitationsSource
#1Sergey Oladyshkin (University of Stuttgart)H-Index: 9
#2Wolfgang Nowak (University of Stuttgart)H-Index: 30
We discuss the arbitrary polynomial chaos (aPC), which has been subject of research in a few recent theoretical papers. Like all polynomial chaos expansion techniques, aPC approximates the dependence of simulation model output on model parameters by expansion in an orthogonal polynomial basis. The aPC generalizes chaos expansion techniques towards arbitrary distributions with arbitrary probability measures, which can be either discrete, continuous, or discretized continuous and can be specified ...
139 CitationsSource
#1Eduardo Haro Sandoval (UP: Panamerican University)H-Index: 2
#2Floriane Anstett-Collin (University of Lorraine)H-Index: 3
Last. Michel BassetH-Index: 12
view all 3 authors...
Global sensitivity has mainly been analyzed in static models, though most physical systems can be described by differential equations. Very few approaches have been proposed for the sensitivity of dynamic models and the only ones are local. Nevertheless, it would be of great interest to consider the entire uncertainty range of parameters since they can vary within large intervals depending on their meaning. Other advantage of global analysis is that the sensitivity indices of a given parameter a...
36 CitationsSource
This paper presents a polynomial dimensional decomposition (PDD) method for global sensitivity analysis of stochastic systems subject to independent random input following arbitrary probability distributions. The method involves Fourier-polynomial expansions of lower-variate component functions of a stochastic response by measure-consistent orthonormal polynomial bases, analytical formulae for calculating the global sensitivity indices in terms of the expansion coefficients, and dimension-reduct...
39 CitationsSource
#2William CastaingsH-Index: 1
Last. Stephano Tarantola (Institute for the Protection and Security of the Citizen)H-Index: 1
view all 3 authors...
Moment independent methods for the sensitivity analysis of model output are attracting growing attention among both academics and practitioners. However, the lack of benchmarks against which to compare numerical strategies forces one to rely on ad hoc experiments in estimating the sensitivity measures. This article introduces a methodology that allows one to obtain moment independent sensitivity measures analytically. We illustrate the procedure by implementing four test cases with different...
94 CitationsSource
#1Chonggang Xu (NCSU: North Carolina State University)H-Index: 26
#2George Z. Gertner (UIUC: University of Illinois at Urbana–Champaign)H-Index: 28
Fourier Amplitude Sensitivity Test (FAST) is one of the most popular uncertainty and sensitivity analysis techniques. It uses a periodic sampling approach and a Fourier transformation to decompose the variance of a model output into partial variances contributed by different model parameters. Until now, the FAST analysis is mainly confined to the estimation of partial variances contributed by the main effects of model parameters, but does not allow for those contributed by specific interactions ...
33 CitationsSource
#1Géraud Blatman (International Facility Management Association)H-Index: 1
#2Bruno Sudret (International Facility Management Association)H-Index: 28
Global sensitivity analysis aims at quantifying the relative importance of uncertain input variables onto the response of a mathematical model of a physical system. ANOVA-based indices such as the Sobol’ indices are well-known in this context. These indices are usually computed by direct Monte Carlo or quasi-Monte Carlo simulation, which may reveal hardly applicable for computationally demanding industrial models. In the present paper, sparse polynomial chaos (PC) expansions are introduced in ...
186 CitationsSource
#1Zhi-Sheng Ye (NUS: National University of Singapore)H-Index: 21
#2Zhizhong Li (THU: Tsinghua University)H-Index: 19
Last. Min Xie (NUS: National University of Singapore)H-Index: 64
view all 3 authors...
Adaptive genetic algorithms (GAs) have been shown to be able to improve GA performance in reliability-related optimization studies. However, there are different ways to implement adaptive GAs, some of which are even in conflict with each other. In this study, a simple parameter-adjusting method using mean and variance of each generation is introduced. This method is used to compare two of such conflicting adaptive GA methods: GAs with increasing mutation rate and decreasing crossover rate and GA...
52 CitationsSource
Cited By17
#1Ruggiero Lovreglio (Massey University)H-Index: 11
#2Enrico Ronchi (Lund University)H-Index: 20
Last. Michael J. Kinsey (ARUP Laboratories)H-Index: 7
view all 3 authors...
Pedestrian evacuation models are often used to assess life safety in the performance-based design process within fire safety engineering. Within this paper, a summary of data collected via an international online survey regarding the models and users’ experiences and needs is presented. This survey consisted of 22 questions focusing on: the assessment of the pedestrian evacuation model user community; their stated importance of model features to select a model; usage/awareness of models; knowled...
3 CitationsSource
#1Ruggiero Lovreglio (Massey University)H-Index: 11
#2Erica D. Kuligowski (NIST: National Institute of Standards and Technology)H-Index: 22
Last. Ken StrahanH-Index: 1
view all 4 authors...
Abstract The occurrence of wildfire threats has increased in the last few decades creating serious challenges for thousands of communities around the world. Understanding the physical and social dynamics imposed by wildfires is fundamental to assessing and reducing the ensuing risk to different communities. Although, several studies investigating household wildfire risk perception and decision-making are available in the literature, modelling solutions to predict household behaviour in wildfire ...
4 CitationsSource
#1Ruggiero Lovreglio (Massey University)H-Index: 11
#2Michael Spearpoint (Cant.: University of Canterbury)H-Index: 14
Last. Mathilde GiraultH-Index: 1
view all 3 authors...
Abstract Simulating human behaviour in fire is often one of the main challenges in designing complex buildings, structures or sites for the life safety of occupants. In fact, evacuation simulations represent a fundamental input to assess fire safety performance using a risk analysis approach. The variability in evacuee behaviours (e.g. pre-evacuation delays and uncongested walking speed) can be probabilistically simulated in egress models using distribution functions. The application of probabil...
1 CitationsSource
#1Di Wu (Ministry of Education)H-Index: 1
#2Xiang-dong Liu (USTB: University of Science and Technology Beijing)H-Index: 2
Last. Gang Li (Ministry of Education)H-Index: 15
view all 5 authors...
Block withholding attack is an action where a miner who has found a legal block chooses not to submit it but rather directly abandons it. This attack makes the mining pool lose all bitcoin rewards contained within the block. In this paper, we construct a generalized model where two participants can choose to either cooperate with each other or employ a block withholding attack in the mining pool. To make the model more realistic, we consider both the cost of partial proof of work and the cost of...
4 CitationsSource
#1Wanying Yun (NPU: Northwestern Polytechnical University)H-Index: 9
#2Zhenzhou Lu (NPU: Northwestern Polytechnical University)H-Index: 21
Last. Xian JiangH-Index: 8
view all 3 authors...
Abstract To efficiently execute the variance-based global sensitivity analysis, the law of total variance in the successive intervals without overlapping is proved at first, on which an efficient space-partition sampling-based approach is subsequently proposed in this paper. Through partitioning the sample points of output into different subsets according to different inputs, the proposed approach can efficiently evaluate all the main effects concurrently by one group of sample points. In additi...
8 CitationsSource
#1Wanying Yun (NPU: Northwestern Polytechnical University)H-Index: 9
#2Zhenzhou Lu (NPU: Northwestern Polytechnical University)H-Index: 21
Last. Xian JiangH-Index: 8
view all 3 authors...
The importance sampling method is an extensively used numerical simulation method in reliability analysis. In this paper, a modification to the importance sampling method (ISM) is proposed, and the modified ISM divides the sample set of input variables into different subsets based on the contributive weight of the importance sample defined in this paper and the maximum super-sphere denoted by β-sphere in the safe domain defined by the truncated ISM. By this proposed modification, only samples wi...
14 CitationsSource
#1Sinan Xiao (NPU: Northwestern Polytechnical University)H-Index: 9
#2Zhenzhou Lu (NPU: Northwestern Polytechnical University)H-Index: 7
Last. Pan Wang (NPU: Northwestern Polytechnical University)H-Index: 9
view all 3 authors...
In this paper, a new kind of multivariate global sensitivity index based on energy distance is proposed. The covariance decomposition based index has been widely used for multivariate global sensitivity analysis. However, it just considers the variance of multivariate model output and ignores the correlation between different outputs. The proposed index considers the whole probability distribution of dynamic output based on characteristic function and contains more information of uncertainty tha...
17 CitationsSource
#1Sinan Xiao (NPU: Northwestern Polytechnical University)H-Index: 9
#2Zhenzhou Lu (NPU: Northwestern Polytechnical University)H-Index: 7
Abstract In structural reliability analysis, sensitivity analysis can be used to measure how the input variable influences the failure of structure. In this work, a new reliability sensitivity analysis method is proposed. In the proposed method, the model output is separated into two classes (failure domain and safe domain). The basic idea is that if the failure-conditional probability density function of input variable is significantly different from its unconditional probability density functi...
20 CitationsSource
#1A. Kaveh (IUST: Iran University of Science and Technology)H-Index: 1
In this chapter a single-solution metaheuristic optimizer, namely, global sensitivity analysis-based (GSAB) algorithm [1], is presented that uses a basic set of mathematical techniques, namely, global sensitivity analysis. Sensitivity analysis (SA) studies the sensitivity of the model output with respect to its input parameters (Rahman [2]). This analysis is generally categorized as local SA and global SA techniques. While local SA studies the sensitivity of the model output about variations aro...
1 CitationsSource
4 CitationsSource