Bootstrap- and permutation-based inference for the Mann–Whitney effect for right-censored and tied data

The Mann–Whitney effect is an intuitive measure for discriminating two survival distributions. Here we analyse various inference techniques for this parameter in a two-sample survival setting with independent right-censoring, where the survival times are even allowed to be discretely distributed. This allows for ties in the data and requires the introduction of normalized versions of Kaplan–Meier estimators from which adequate point estimates are deduced. Asymptotically exact inference procedures based on standard normal, bootstrap, and permutation quantiles are developed and compared in simulations. Here, the asymptotically robust and—under exchangeable data—even finitely exact permutation procedure turned out to be the best. Finally, all procedures are illustrated using a real data set.     

Some recent work on multivariate Gaussian Markov random fields

Some recent work on conditional formulation of multivariate Gaussian Markov random fields is presented. The focus is on model constructions by compatible conditionals and coregionalization. Special attention is given to multivariate generalizations of univariate models. Beginning with univariate model constructions, a survey of key approaches to formulating multivariate extensions is presented. Two challenges in the formulation and implementation of multivariate models are highlighted: (1) entanglement of spatial and non-spatial components, and (2) enforcement for positivity condition. Managing the two challenges by decomposition, separation, and constrained parameterization is discussed. Also highlighted is the challenge of flexible modeling of (conditional) cross-spatial dependencies and, in particular, asymmetric cross-spatial dependencies. Interpretation of asymmetric cross-spatial dependencies is also discussed. A coregionalization framework which connects and unifies the various lines of model development is presented. The framework enables a systematic development of a broad range of models via linear and spatially varying coregionalization, respectively, with extensions to locally adaptive models. Formulation of multivariate models over variable-specific lattices is discussed. Selected models are illustrated with examples of Bayesian multivariate and spatiotemporal disease mapping. Potential applications of coregionalization models in imaging analysis, covariance modeling, dimension reduction, and latent variable analysis are briefly mentioned.     

Rejoinder on: Some recent work on multivariate Gaussian Markov random fields

I thank the discussants, Miguel A. Martinez-Beneito, Fedel Greco, Carlo Trivisano, Stephan R Sain, and Reinhard Furrer, for their insightful and stimulating commentary. The rejoinder is organized in five sections: (1) the M-based models, (2) posterior sensitivity to prior choices for \({\varvec{C}}\) and \({\varvec{\varSigma }}\), (3) stationary and non-stationary (M)GMRFs, (4) various approaches to model formulation and related applications, and (5) statistical computation.     

A thermovisco-hyperelastic constitutive model of HTPB propellant with damage at intermediate strain rates

The uniaxial compressive tests at different temperatures (223–298 K) and strain rates (\(0.40\mbox{--}63~\mbox{s}^{-1}\)) are reported to study the properties of hydroxyl-terminated polybutadiene (HTPB) propellant at intermediate strain rates, using a new INSTRON testing machine. The experimental results indicate that the compressive properties (mechanical properties and damage) of HTPB propellant are remarkably affected by temperature and strain rate and display significant nonlinear material behaviors at large strains under all the test conditions. Continuously decreasing temperature and increasing strain rate, the characteristics of stress-strain curves and damage for HTPB propellant are more complex and are significantly different from that at room temperature or at lower strain rates. A new constitutive model was developed to describe the compressive behaviors of HTPB propellant at room temperature and intermediate strain rates by simply coupling the effect of strain rate into the conventional hyperelastic model. Based on the compressive behaviors of HTPB propellant and the nonlinear viscoelastic constitutive theories, a new thermovisco-hyperelastic constitutive model with damage was proposed to predict the stress responses of the propellant at low temperatures and intermediate strain rates. In this new model, the damage is related to the viscoelastic properties of the propellant. Meanwhile, the effect of temperature on the hyperelastic properties, viscoelastic properties and damage are all considered by the macroscopical method. The constitutive parameters in the proposed constitutive models were identified by the genetic algorithm (GA)-based optimization method. By comparing the predicted and experimental results, it can be found that the developed constitutive models can correctly describe the uniaxial compressive behaviors of HTPB propellant at intermediate strain rates and different temperatures.     

Time-dependent behavior of layered arches with viscoelastic interlayers

This work studies the time-dependent behavior of a layered arch adhesively bonded by viscoelastic interlayers. The deformation of the viscoelastic interlayer is represented by the Maxwell–Wiechert model. The constitutive relation in an interlayer is simplified through the quasi-elastic approximation approach. The mechanical property of an arch layer is described by the exact two-dimensional (2-D) elasticity theory in polar coordinates. The stress and displacement components in an arch layer, which strictly satisfy the simply supported boundary conditions, have been analytically derived out. The stresses and displacements are efficiently obtained by means of the recursive matrix method for the arch with any number of layers. The comparison study shows that the 2-D finite element solution has good agreement with the present one, while the solution based on the one-dimensional (1-D) Euler–Bernoulli theory has considerable error, especially for thick arches. The influences of geometrical and material parameters on the time-dependent behavior of the layered arch are analyzed in detail.     

Comparing interconversion methods between linear viscoelastic material functions

A variety of methods applicable to the interconversion of static (creep) and dynamic (relaxation) functions, with regard to appropriate experimental data of various polymers is investigated and compared. The effectiveness of the selected methods was verified by a series of creep experimental data of various polymeric structures. While most of the employed methods are well established in the literature, some further modifications have been introduced for an improvement of the conversion procedure. Furthermore, a new approach is also employed, which is based on the stretched-exponential function, usually applied to represent both relaxation and retardation functions. It is seen that the examined methods produce a similar result, concerning the creep compliance function, having as a beginning storage and loss modulus experimental data. The same observation applies to the retardation spectra, pointing the fact that discrete spectra deviates significantly from the continuous spectra. As a result, it is shown that the creep compliance function, or the relaxation modulus function, can be predicted using experimental dynamic data (relaxation or creep, respectively), as well as anyone of the examined interconversion methods, with an accuracy close to 5%. The use of approximate or exact relations in the whole procedure was proved not to have a significant effect on the final result (referring mostly to the retardation spectra).     

Ratcheting in a nonlinear viscoelastic adhesive

Uniaxial time-dependent creep and cycled stress behavior of a standard and toughened film adhesive were studied experimentally. Both adhesives exhibited progressive accumulation of strain from an applied cycled stress. Creep tests were fit to a viscoelastic power law model at three different applied stresses which showed nonlinear response in both adhesives. A third order nonlinear power law model with a permanent strain component was used to describe the creep behavior of both adhesives and to predict creep recovery and the accumulation of strain due to cycled stress. Permanent strain was observed at high stress but only up to 3% of the maximum strain. Creep recovery was under predicted by the nonlinear model, while cycled stress showed less than 3% difference for the first cycle but then over predicted the response above 1000 cycles by 4–14% at high stress. The results demonstrate the complex response observed with structural adhesives, and the need for further analytical advancements to describe their behavior.     

Spectrophotometric analysis of disintegration mechanisms (abrasion and crushing) of agglomerates during the disc granulation of dolomite

The analysis of mechanisms which affect the formation of agglomerates and determine the granulation process in a broad sense encounters difficulties related to the many ways of the formation of granules. The aim of the study was to perform a qualitative and quantitative analysis of granulation mechanisms with special reference to agglomerates’ disintegration in the disc granulation process. This paper contains an analysis of disintegration mechanisms (abrasion and crushing) of agglomerates during the disc granulation of dolomite. The analysis of the mechanisms taking place during the process concerns the granulation stage after wetting. During the research, each time after the wetting stage, the size fraction 10–12.5 mm was substituted with an analogous fraction wetted by means of an aqueous solution of a coloring agent and the process was continued. After the specified time of granulation, the obtained product was sieved through laboratory sieves and then the content of the provided coloring agent in different size fractions was analyzed by means of a spectrophotometer. Measuring the absorbance of the analyzed samples and granulometric composition of the bed, the level and cause of the migration of material of the tested fraction into other size classes were determined, and at the same time, the occurring granulation mechanisms were analyzed. The proposed model and measurement method consisting of determining the absorbance of the tested granulometric fraction enable the qualitative and quantitative analysis of granulation mechanisms are encountered during the carrying-out of the process after wetting the bed.     

Probabilistic calibration of discrete element simulations using the sequential quasi-Monte Carlo filter

The calibration of discrete element method (DEM) simulations is typically accomplished in a trial-and-error manner. It generally lacks objectivity and is filled with uncertainties. To deal with these issues, the sequential quasi-Monte Carlo (SQMC) filter is employed as a novel approach to calibrating the DEM models of granular materials. Within the sequential Bayesian framework, the posterior probability density functions (PDFs) of micromechanical parameters, conditioned to the experimentally obtained stress–strain behavior of granular soils, are approximated by independent model trajectories. In this work, two different contact laws are employed in DEM simulations and a granular soil specimen is modeled as polydisperse packing using various numbers of spherical grains. Knowing the evolution of physical states of the material, the proposed probabilistic calibration method can recursively update the posterior PDFs in a five-dimensional parameter space based on the Bayes’ rule. Both the identified parameters and posterior PDFs are analyzed to understand the effect of grain configuration and loading conditions. Numerical predictions using parameter sets with the highest posterior probabilities agree well with the experimental results. The advantage of the SQMC filter lies in the estimation of posterior PDFs, from which the robustness of the selected contact laws, the uncertainties of the micromechanical parameters and their interactions are all analyzed. The micro–macro correlations, which are byproducts of the probabilistic calibration, are extracted to provide insights into the multiscale mechanics of dense granular materials.     

Numerical simulation of 2D granular flow entrainment using DEM

To understand the entrainment process in granular flow, numerical experiments have been conducted using a Discrete Element Method model. A flow channel of 8 m long with \(15^\circ \) slope is setup with monitoring points located in an erodible bed. Particles, ranging from 3 to 4 mm in diameters, are used in the simulations. In the simulations, translational, rotational and average velocities, total volume, shear stresses are calculated in the measurement circles. The sizes of the measurement circles have been varied to see their effects on the results. It is found the minimum size of the measurement circles should include 20–30 particles. An new analytical model has been developed to calculate entrainment in granular flow. Results of the numerical experiment are compared with analytical model. Shear stresses at the interface between flowing particles in motion and the immobile particles in the channel bed, change of depth of erosion and entrainment rate are used to verify the analytical model. It is found that the calculated shear stresses in the PFC model agree well with the shear stresses calculated using Mohr–Coulomb frictional relationship in the analytical model. The calculated depth of erosion using the new analytical model is also compared with that from dynamic and static entrainment model. The results indicates that the analytical model is able to capture the mechanism of erosion and it can be used in granular flow analysis.