Article
Engineering, Industrial
Fabrice Gamboa, Thierry Klein, Agnes Lagnoux, Leonardo Moreno
Summary: This paper introduces new sensitivity indices adapted to outputs valued in general metric spaces, encompassing classical ones such as Sobol indices and Cramer-von-Mises indices. Asymptotically Gaussian estimators of these indices based on U-statistics are provided, with surprising straightforward proof of asymptotic normality. The new procedure is illustrated on a toy model and two real-data examples.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Computer Science, Artificial Intelligence
Clement Benesse, Fabrice Gamboa, Jean-Michel Loubes, Thibaut Boissin
Summary: This study combines Global Sensitivity Analysis and Fairness, defining fairness using a special framework based on Global Sensitivity Analysis and demonstrating common indicators between the two fields. New Global Sensitivity Analysis indices and rates of convergence are also presented as fairness proxies.
Article
Mechanics
Filipe Meirelles Fonseca, Paulo Batista Goncalves
Summary: In the field of structural stability, there has been a renewed interest due to new applications involving smart and deployable structures, micro-and nanocomponents and mechanical metamaterials. This paper investigates the nonlinear behavior, bifurcations and instabilities of a hyperelastic von Mises truss exhibiting multistable behavior. Analyzing the influence of geometric and material nonlinearities, the study provides insights into the truss load carrying capacity and expands the understanding of nonlinear materials.
INTERNATIONAL JOURNAL OF NON-LINEAR MECHANICS
(2022)
Article
Statistics & Probability
Zdenek Hlavka, Daniel Hlubinka, Katerina Konasova
Summary: This paper studies functional two-sample tests based on empirical characteristic functionals, proposes a computationally feasible and powerful test statistic, and explores its small sample properties through simulation studies. The results show that, with the proper choice of probability measure, the proposed test can effectively detect shift and scale alternatives.
JOURNAL OF MULTIVARIATE ANALYSIS
(2022)
Article
Construction & Building Technology
Yuan-Zuo Wang, Guo-Qiang Li, Yan-Bo Wang, Yi-Fan Lyu
Summary: The study investigates the influence of the initial geometry of the specimen and the plastic hardening property of materials on the necking process, and proposes a modified power law function to characterize the von Mises stress-strain curve on circumferentially notched specimens without measuring the instantaneous geometry of the cross-section.
JOURNAL OF CONSTRUCTIONAL STEEL RESEARCH
(2021)
Article
Statistics & Probability
Eduardo Garcia-Portugues, Javier Alvarez-Liebana, Gonzalo Alvarez-Perez, Wenceslao Gonzalez-Manteiga
Summary: The article introduces a novel goodness-of-fit test for FLMFR, formulated using Cramer-von Mises norm and geometric arguments to compute a weighted quadratic norm. The test is calibrated through a resampling procedure and a new estimator is presented. Simulation studies and real data sets demonstrate the effectiveness of the test method.
SCANDINAVIAN JOURNAL OF STATISTICS
(2021)
Article
Computer Science, Interdisciplinary Applications
R. Azais, S. Ferrigno, M-J Martinez
Summary: Various goodness-of-fit tests have been developed for assessing assumptions in regression models. They include both 'directional' tests that detect departures from specific assumptions and 'global' tests that assess overall model fit. The focus of this study is on choosing the structural part of the regression function, with the development of nonparametric tests and an easy-to-use tool, cvmgof. The package is illustrated through a tutorial on real data and a simulation study is conducted to compare the three tests implemented.
JOURNAL OF STATISTICAL COMPUTATION AND SIMULATION
(2022)
Article
Statistics & Probability
Rasmus Erlemann, Richard Lockhart, Rihan Yao
Summary: This study focuses on two nonparametric tests for testing the hypothesis of identical distribution of a sequence of independent observations, and analyzes the situation where the distribution changes at a single change point. By using the Cramer-von Mises two-sample test, one test utilizes the maximum test statistic over all possible change points, while the other test averages over all possible change points, showing that the average statistic provides useful p-values more quickly, particularly in long sequences. Power analysis for contiguous alternatives reveals that the average statistic has larger limiting power than its level, while the maximal statistic does not. Asymptotic methods and bootstrapping are employed to construct the test distribution, and the tests' performance is verified through a Monte Carlo power study with various alternative distributions.
SCANDINAVIAN JOURNAL OF STATISTICS
(2022)
Article
Engineering, Aerospace
Alejandro Cano, Alejandro Pastor, Diego Escobar, Joaquin Miguez, Manuel Sanjurjo-Rivo
Summary: The reliability of uncertainty characterization is crucial for Space Situational Awareness (SSA) services, and the most relevant source of uncertainty is the inherent uncertainty of dynamic models. This study proposes a methodology to infer the variances of uncertain parameters based on the observed distribution of the Mahalanobis distance between predicted and estimated orbits. The methodology is validated in simulated scenarios that emulate the complexity of operational applications.
ADVANCES IN SPACE RESEARCH
(2023)
Article
Materials Science, Multidisciplinary
Dawei Ji, Xianming Hu, Zuopeng Zhao, Xu Jia, Xuteng Hu, Yingdong Song
Summary: Creep tests were conducted on notched plate specimens of nickel-based superalloy GH4169 with varying stress concentration coefficients. The study found a correlation between the stress concentration coefficient and the creep behavior, and four constitutive models were utilized to predict the creep life of the specimens. The results showed that the predicted creep life using the minimum average Von Mises equivalent stress of notched specimens fell within a certain dispersion band for the different constitutive models.
Article
Statistics & Probability
Fabrice Gamboa, Pierre Gremaud, Thierry Klein, Agnes Lagnoux
Summary: In this paper, we propose a new statistical estimation framework for calculating various global sensitivity analysis indices. We demonstrate the effectiveness and numerical efficiency of our method and establish the consistency of the resulting estimators.
Article
Construction & Building Technology
Chengbin Wu, Xingzhou Zhu, Wei Si
Summary: This study presents a groundbreaking reliability model for the evaluation of asphalt pavement performance. The study identifies four key uncertainty factors that affect pavement performance and explores the complex dynamics of pavement structure strength. It also analyzes five random variables that reflect the inherent uncertainties during its lifecycle.
CASE STUDIES IN CONSTRUCTION MATERIALS
(2023)
Article
Engineering, Multidisciplinary
Diab W. Abueidda, Ziliang Kang, Seid Koric, Kai A. James, Iwona M. Jasiuk
Summary: This article introduces a computational design framework for obtaining three-dimensional periodic elastoplastic architected materials with enhanced performance, using a nonlinear finite element model and path-dependent adjoint sensitivity formulation for optimization. The optimization problem is parametrized using the solid isotropic material penalization method to produce materials with enhanced performance.
INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING
(2021)
Article
Engineering, Marine
Sadaoki Matsui
Summary: This study presents a practical method for estimating the extreme value distribution of von Mises stress for ship structural strength evaluations. The proposed method is based on an asymptotic approximation and can be easily calculated. Through structural analysis and parametric analysis, it is verified that the method has high accuracy for structural strength evaluation.
JOURNAL OF MARINE SCIENCE AND TECHNOLOGY
(2023)
Article
Statistics & Probability
Daniel Gaigall, Julian Gerstenberg
Summary: The Cramer-von-Mises distance is used to measure the distribution of the excess over a confidence level. The study explores the asymptotics of related statistics and finds that the obtained limit distributions differ from the classical ones. To address this, quantiles of the new limit distributions are provided and new bootstrap techniques for approximation purposes are introduced and justified. The results lead to new one-sample goodness-of-fit tests and a new confidence interval for the fitting error of the distribution of the excess over a confidence level. Simulation studies evaluate the performance of the tests and the coverage probabilities of the confidence interval in finite sample cases. An application of the Cramer-von-Mises tests in determining an appropriate confidence level for the fitting approach is also discussed.
JOURNAL OF NONPARAMETRIC STATISTICS
(2023)
