Model-based Procedure Research Articles

A Model-based Simulation Procedure for the Evolution of Tertiary Creep with Combined Damage Diffusion and Viscoplasticity

It appears from macro- and micro-experiments that the essential feature of the tertiary creep of certain materials is characterized by the evolution of localized damage that results in the formation and evolution of macrocracks. Higher-order continuum models, such as nonlocal integral or strain gradient models, have been developed by the research community over the last twenty years to predict the evolution of localized failure. However, the use of higher-order models yields higher-order governing differential equations with the ambiguity in the physics behind additional boundary conditions, and with the difficulty in large-scale model-based simulations. To simulate the evolution of localized failure in tertiary creep with the least computational cost, therefore, the relationship among the existing regularizing approaches of localization is investigated first. Based on the analysis, it is then proposed that a coupled viscoplasticity-damage model be combined with a damage diffusion law to simulate a complete creep process. As a result, the mesh-objective solutions could be obtained without invoking higher-order spatial terms in the strain-stress space. Example problems are considered to illustrate the potential of the proposed model-based numerical procedure in a large-scale simulation.

Model-based geostatistical interpolation of the annual number of ozone exceedance days in the Netherlands

This paper discusses two model-based geostatistical methods for spatial interpolation of the number of days that ground level ozone exceeds a threshold level. The first method assumes counts to approximately follow a Poisson distribution, while the second method assumes a log-Normal distribution. First, these methods were compared using an extensive data set covering the Netherlands, Belgium and Germany. Second, the focus was placed on only the Netherlands, where only a small data set was used. Bayesian techniques were used for parameter estimation and interpolation. Parameter estimates are comparable due to the log-link in both models. Incorporating data from adjacent countries improves parameter estimation. The Poisson model predicts more accurately (maximum kriging standard deviation of 2.16 compared to 2.69) but shows smoother surfaces than the log-Normal model. The log-Normal approach ensures a better representation of the observations and gives more realistic patterns (an RMSE of 2.26 compared to 2.44). Model-based geostatistical procedures are useful to interpolate limited data sets of counts of ozone exceedance days. Spatial risk estimates using existing prior information can be made relating health effects to environmental thresholds.

Comparison of grinding media—Cylpebs versus balls

Cylpebs are slightly tapered cylindrical grinding media with a ratio of length to diameter of unity. The manufactures have made conflicting claims regarding the milling performance of Cylpebs in comparison with balls. One major point of interest is which one grinds finer at the same operating conditions. The difficulty in comparison is due to the shape difference. The two grinding media have different surface area, bulk density and contact mechanisms in grinding action. Comparative tests were conducted using the two types of grinding media in a laboratory Bond ball mill at various conditions of equality such as media mass, size distribution, surface area and input specific energy. The laboratory results indicate that at the same specific energy input level the Cylpebs produce a product with slightly less oversize due to their greater surface area, but essentially the same sizing at the fine end as that produced with the balls. The reason may be that the advantage of greater surface area is balanced by the line contact and area contact grinding actions with the Cylpebs. A new ball mill scale-up procedure [Man, Y.T., 2001. Model-based procedure for scale-up of wet, overflow ball mills, Part I: outline of the methodology. Minerals Engineering 14 (10), 1237–1246] was employed to predict grinding performance of an industrial mill from the laboratory test results. The predicted full scale operation was compared with the plant survey data. Some problems in the original scale-up procedures were identified. The scale-up procedure was therefore modified to allow the predicted ball mill performance to match the observed one. The calibrated scale-up procedure was used to predict the Cylpebs performance in the full scale industrial mill using the laboratory tests results.

Comparison of grinding media—Cylpebs versus balls

Cylpebs are slightly tapered cylindrical grinding media with a ratio of length to diameter of unity. The manufactures have made conflicting claims regarding the milling performance of Cylpebs in comparison with balls. One major point of interest is which one grinds finer at the same operating conditions. The difficulty in comparison is due to the shape difference. The two grinding media have different surface area, bulk density and contact mechanisms in grinding action. Comparative tests were conducted using the two types of grinding media in a laboratory Bond ball mill at various conditions of equality such as media mass, size distribution, surface area and input specific energy. The laboratory results indicate that at the same specific energy input level the Cylpebs produce a product with slightly less oversize due to their greater surface area, but essentially the same sizing at the fine end as that produced with the balls. The reason may be that the advantage of greater surface area is balanced by the line contact and area contact grinding actions with the Cylpebs. A new ball mill scale-up procedure [Man, Y.T., 2001. Model-based procedure for scale-up of wet, overflow ball mills, Part I: outline of the methodology. Minerals Engineering 14 (10), 1237–1246] was employed to predict grinding performance of an industrial mill from the laboratory test results. The predicted full scale operation was compared with the plant survey data. Some problems in the original scale-up procedures were identified. The scale-up procedure was therefore modified to allow the predicted ball mill performance to match the observed one. The calibrated scale-up procedure was used to predict the Cylpebs performance in the full scale industrial mill using the laboratory tests results.

Missing Data in Homicide Research

This article is an introduction to the special issue of Homicide Studies on missing data. The first section is an overview of the status of missing data approaches in homicide research. It begins by describing the importance of missing data estimation in homicide. This is followed by a discussion of missing data mechanisms, complete case analysis, imputation and weighting, and model-based procedures. The second section is a brief description of each of the articles in this issue. The conclusion describes the myth associated with imputing missing data, the use of missing data approaches in public records, the Supreme Court case that found hot-deck imputation acceptable for the census, and guidelines for handling missing data published by the American Psychological Association. This section concludes by describing the kinds of research that need to be done.

Predicting auditory tone-in-noise detection performance: the effects of neural variability.

Collecting and analyzing psychophysical data is a fundamental mechanism for the study of auditory processing. However, because this approach relies on human listening experiments, it can be costly in terms of time and money spent gathering the data. The development of a theoretical, model-based procedure capable of accurately predicting psychophysical behavior could alleviate these issues by enabling researchers to rapidly evaluate hypotheses prior to conducting experiments. This approach may also provide additional insight into auditory processing by establishing a link between psychophysical behavior and physiology. Signal detection theory has previously been combined with an auditory model to generate theoretical predictions of psychophysical behavior. Commonly, the ideal processor outperforms human subjects. In order for this model-based technique to enhance the study of auditory processing, discrepancies must be eliminated or explained. In this paper, we investigate the possibility that neural variability, which results from the randomness inherent in auditory nerve fiber responses, may explain some of the previously observed discrepancies. In addition, we study the impact of combining information across nerve fibers and investigate several models of multiple-fiber signal processing. Our findings suggest that neural variability can account for much, but not all, of the discrepancy between theoretical and experimental data.

ESTIMATING POPULATION TRENDS WITH A LINEAR MODEL: TECHNICAL COMMENTS

Abstract Controversy has sometimes arisen over whether there is a need to accommodate the limitations of survey design in estimating population change from the count data collected in bird surveys. Analyses of surveys such as the North American Breeding Bird Survey (BBS) can be quite complex; it is natural to ask if the complexity is necessary, or whether the statisticians have run amok. Bart et al. (2003) propose a very simple analysis involving nothing more complicated than simple linear regression, and contrast their approach with model-based procedures. We review the assumptions implicit to their proposed method, and document that these assumptions are unlikely to be valid for surveys such as the BBS. One fundamental limitation of a purely design-based approach is the absence of controls for factors that influence detection of birds at survey sites. We show that failure to model observer effects in survey data leads to substantial bias in estimation of population trends from BBS data for the 20 specie...

Development of an in vivo method for determining material properties of passive myocardium

Calculation of mechanical stresses and strains in the left ventricular (LV) myocardium by the finite element (FE) method relies on adequate knowledge of the material properties of myocardial tissue. In this paper, we present a model-based estimation procedure to characterize the stress-strain relationship in passive LV myocardium. A 3D FE model of the LV myocardium was used, which included morphological fiber and sheet structure and a nonlinear orthotropic constitutive law with different stiffness in the fiber, sheet, and sheet-normal directions. The estimation method was based on measured wall strains. We analyzed the method's ability to estimate the material parameters by generating a set of synthetic strain data by simulating the LV inflation phase with known material parameters. In this way we were able to verify the correctness of the solution and to analyze the effects of measurement and model error on the solution accuracy and stability. A sensitivity analysis was performed to investigate the observability of the material parameters and to determine which parameters to estimate. The results showed a high degree of coupling between the parameters governing the stiffness in each direction. Thus, only one parameter in each of the three directions was estimated. For the tested magnitudes of added noise and introduced model errors, the resulting estimated stress-strain characteristics in the fiber and sheet directions converged with good accuracy to the known relationship. The sheet-normal stress-strain relationship had a higher degree of uncertainty as more noise was added and model error was introduced.

Model-based deconvolution for dominant, thin-bed seismic reflections

Predictive deconvolution is widely treated as a universally applicable tool for multiple removal and wavelet compression. The fundamental assumption of random reflectivity is seriously compromised in geological situations where the reflection sequence comprises a small number of dominant horizons. This situation is not uncommon in coal environments. Where the primary seismic objective is high quality imaging of particular target horizons, an improved result can be achieved if the deconvolution is designed according to assumptions more relevant to the geological situation. We outline a simple example of this approach, aimed at imaging a production coal seam, of thickness 5-10 m, at a mine in the Bowen Basin, Australia. Using horizon time picks from a preliminary volume, the full reflection package associated with the seam is extracted and deterministically filtered to obtain an estimate of the intrinsic wavelet. A Wiener spiking filter, designed on the extracted wavelet, is then used to deconvolve the seam package. In comparison to the predictive deconvolution approach, this model-based procedure provides improved resolution of the top and base coal interfaces. In addition, derived amplitude and frequency attributes are more robust in terms of known geology. Variants of this simple model-based procedure should have relevance in a range of dominant-horizon situations where predictive deconvolution is invalid.

How to Deal with Missing Categorical Data: Test of a Simple Bayesian Method

The authors analyze the efficiency of six missing data techniques for categorical item nonresponse under the assumption that data are missing at random or missing completely at random. By efficiency, the authors mean a procedure that produces an unbiased estimate of true sample properties that is also easy to implement. The investigated techniques include listwise deletion, mode substitution, random imputation, two regression imputations, and a Bayesian model-based procedure. The authors analyze efficiency under six experimental conditions for a survey-based data set. They find that listwise deletion is efficient for the data analyzed. If data loss due to listwise deletion is an issue, the analysis points to the Bayesian method. Regression imputation is also efficient, but the result is conditioned on the specific data structure and may not hold in general. Additional problems arise when using regression imputation, making it less appropriate.

How to Deal with Missing Categorical Data: Test of a Simple Bayesian Method

The authors analyze the efficiency of six missing data techniques for categorical item nonresponse under the assumption that data are missing at random or missing completely at random. By efficiency, the authors mean a procedure that produces an unbiased estimate of true sample properties that is also easy to implement. The investigated techniques include listwise deletion, mode substitution, random imputation, two regression imputations, and a Bayesian model-based procedure. The authors analyze efficiency under six experimental conditions for a survey-based data set. They find that listwise deletion is efficient for the data analyzed. If data loss due to listwise deletion is an issue, the analysis points to the Bayesian method. Regression imputation is also efficient, but the result is conditioned on the specific data structure and may not hold in general. Additional problems arise when using regression imputation, making it less appropriate.

A failed model-based attempt to implement an evidence-based nursing guideline for fall prevention.

An evidence-based nursing guideline had been locally developed in 1993 to reduce fall incidence rates, creating a 30% reduction. Implementation had failed though. Between 1999 and 2001 the guideline was updated. A multifaceted intervention was chosen based on a model for implementing change. The study was performed in 2 wards. All recommendations of Grol's 5-step implementation model were followed. The aim was a reduction of 30% in fall incidence within a year. Data on falls were extracted from nursing records and Incidence Report Forms (IRFs). In a pilot study an average of 9 falls per 1000 patients per day had been recorded in the department of internal medicine and 16 in the neurology ward. Given the desired reduction of 30%, the target averages were 6 and 11 falls respectively. During the intervention year the average incidences were 8 and 13 falls (95% CI: 6-11 and 10-15). There was a changeable pattern over time without any declining trend. The percentage filled in IRFs varied strongly, with an average of 52% in the department of internal medicine and 60% in the neurology department. There has been no durable decrease in monthly falls despite the use of a model-based procedure for implementing change. Neither did we observe any improvement in filling in IRFs. It can be questioned if the nurses themselves did experience patient falls to be troublesome enough. Investigating this is difficult though. Although the most successful strategy still appears to be changing attitudes of nurses in order to increase fall prevention, there is no clear strategy on how to create this successfully.

Clustering for Sparsely Sampled Functional Data

We develop a flexible model-based procedure for clustering functional data. The technique can be applied to all types of curve data but is particularly useful when individuals are observed at a sparse set of time points. In addition to producing final cluster assignments, the procedure generates predictions and confidence intervals for missing portions of curves. Our approach also provides many useful tools for evaluating the resulting models. Clustering can be assessed visually via low-dimensional representations of the curves, and the regions of greatest separation between clusters can be determined using a discriminant function. Finally, we extend the model to handle multiple functional and finite-dimensional covariates and show how it can be applied to standard finite-dimensional clustering problems involving missing data.

Recovering and profiling the true segmentation structure in markets: an empirical investigation

Although a variety of approaches for inferring market segments exist, little, if any, effort has been devoted to comparing the relative validity of these approaches. This study conducts two extensive simulation experiments in a scanner data setting to empirically compare and validate alternative mixture model-based procedures for segmenting households using choice behaviors and household characteristics. Compared to existing two-stage approaches, a new method known as the joint approach produced 23–27% less error in estimates of characteristics and 30–38% less error in estimates of choice model parameters. Contrary to conventional wisdom, the joint approach, which simultaneously uses household choice and characteristic data, is shown to be superior even when one is interested in recovering only characteristic-based segments.

Estimating Vaccine Efficacy From Secondary Attack Rates

Epidemiologists have used secondary attack rates (SARs) to estimate the protective effects of vaccination since the 1930s. SARs can also be used to estimate the effect of vaccination on reducing infectiousness in breakthrough cases. The conventional SAR approach has been to pool the denominators and numerators across transmission units, then to use a confidence interval for a simple relative risk. We demonstrate appropriate model-based methods to estimate vaccine efficacy (VE) from SARs using generalized estimating equations taking correlation within transmission units into account. The model-based procedures require transformation of the parameter estimates to the SAR scale to obtain vaccine efficacy estimates. Appropriate confidence intervals are then based on the bootstrap, with resampling done by transmission unit. We show that the usual confidence intervals are too narrow. We estimated the effect of pertussis vaccination on person-to-person transmission. The results show that pertussis vaccination reduces the ability of a breakthrough clinical case to produce other clinical cases. The methods can be used in evaluating VE for susceptibility and infectiousness from SARs in other infectious diseases.

Spatial and non-spatial model-based protection procedures for the release of business microdata

In this paper we discuss methodology for the safe release of business microdata. In particular we extend the model-based protection procedure of Franconi and Stander (2002, The Statistician 51: 1–11) by allowing the model to take account of the spatial structure underlying the geographical information in the microdata. We discuss the use of the Gibbs sampler for performing the computations required by this spatial approach. We provide an empirical comparison of these non-spatial and spatial disclosure limitation methods based on the Italian sample from the Community Innovation Survey. We quantify the level of protection achieved for the released microdata and the error induced when various inferences are performed. We find that although the spatial method often induces higher inferential errors, it almost always provides more protection. Moreover the aggregated areas from the spatial procedure can be somewhat more spatially smooth, and hence possibly more meaningful, than those from the non-spatial approach. We discuss possible applications of these model-based protection procedures to more spatially extensive data sets.

Nearest-Neighbor Variance Estimation (NNVE)

Robust covariance estimation commonly proceeds by downweighting outliers. In this article we measure the “outlyingness” of a data point by the standardized distance between the point and its Kth nearest neighbor. The appropriate weights for robust estimation are found by a model-based mixture modeling approach that follows from considering the data cloud as a realization of a high-dimensional point process. To correct a potential bias when there are no outliers, we introduce a boundary correction procedure that artificially adds in extra outlying points; the resulting methodology is called nearest-neighbor variance estimation (NNVE). The strength of NNVE is its robustness against a large proportion of noise points and against deviation from normality of the signal. A consistency result for the method is established. Under some reasonable assumptions, it is shown that the covariance estimate is bounded and that each point has only bounded influence on the final estimates. NNVE outperformed the popular minimum volume ellipsoid (MVE) estimator in simulation studies, with a big improvement when the proportion of outliers was very large (> 50%). In our simulation study, when the proportion of outliers was ≥ 50%, the mean squared error of the NNVE estimator of variance was at least 100 times smaller than that of the MVE estimator. The proposed estimator also outperformed MVE in cases where the underlying data distribution was not normal. Good performance of NNVE in several real examples is demonstrated. A potential drawback of NNVE is that data points condensed in moderate-sized clusters would be classified as signal. Even though we do not support approaches discarding moderate-sized clusters as outliers without checking, this feature of NNVE could be problematic, particularly when only the main data cloud is of interest. A simple diagnostic tool built on existing model-based clustering procedure is proposed. This procedure allows us to check whether there is more than one separate data cloud in the data after cleaning. It also supplies the central locations of the separated moderate-sized clusters, which allows further investigation. Finally, because NNVE reduces the problem of finding the robustness weights to a one-dimensional problem, it may be useful in high-dimensional problems, such as those encountered in data mining.

About model-based time series procedures : some remarks to TRAMO/SEATS and CENSUS X-12-ARIMA

About model-based time series procedures : some remarks to TRAMO/SEATS and CENSUS X-12-ARIMA

MBPCA application for fault detection in NMOS fabrication

This paper describes the application of model-based principal component analysis (MBPCA) to the identification and isolation of faults in NMOS manufacture. In MBPCA, multivariate statistics are applied to the analysis of the portion of the data variance that is unexplained by models based on material and energy balances carried out on the unit operations used in manufacture. It is demonstrated that the failure detection and isolation performance achievable using the model-based procedure exceeds that of commonly used univariate SPC or conventional PCA approaches.

Model-based procedure for scale-up of wet, overflow ball mills part II: worked example

A new ball mill scale-up procedure is developed which uses laboratory data to predict the performance of full-scale ball mill circuits. This procedure contains two laboratory tests. These laboratory tests give the data for the determination of the parameters of a ball mill model. A set of scale-up criteria then scales-up these parameters. The procedure uses the scaled-up parameters to simulate the steady state performance of the full-scale mill circuit. At the end of the simulation, the scale-up procedure gives the size distribution, the volumetric flowrate and the mass flowrate of all the streams in the circuit, and the mill power draw.A worked example shows how the new ball mill scale-up procedure is executed. This worked example uses laboratory data to predict the performance of a full-scale re-grind mill circuit. This circuit consists of a ball mill in closed circuit with hydrocyclones. The full-scale ball mill has a diameter (inside liners) of 1.85m. The scale-up procedure shows that the full-scale circuit produces a product (hydrocyclone overflow) that has an 80% passing size of 80 μm. The circuit has a recirculating load of 173%. The calculated power draw of the full-scale mill is 92kW.