Search

Most research in design of experiments focuses on appropriate designs for a system with just one type of response, rather than multiple responses. In a decision‐making process, relying on only one objective can lead to oversimplified, suboptimal choices that ignore important considerations. Consequently, the problem of constructing a design for an experiment when multiple types of responses are of interest often does not have a single definitive answer, particularly when the response variables have different distributions. Each of these response distributions imposes different requirements on the experimental design. Computer‐generated optimal designs are popular design choices for less standard scenarios where classical designs are not ideal. This work presents a new approach to experimental designs for dual‐response systems. The normal and binomial distributions are considered as potential responses. Using the D‐criterion for the linear model and the Bayesian D‐criterion for the logistic regression model, a weighted criterion is implemented in a coordinate‐exchange algorithm. Designs are evaluated and compared across different weights. The sensitivity of the designs to the priors supplied for the Bayesian D‐criterion is also explored. [ABSTRACT FROM AUTHOR]

The average prediction variance for an I‐optimal design for a specified normal theory linear model decreases nonlinearly with respect to sample size. In this paper, we develop a prediction equation to explain the relationship between average prediction variance and sample size. We investigate methods for determining what sample size is efficient for a given experiment using the average prediction variance (APV) versus sample size curves. The sample size determination is studied assuming a variety of cost structures for the trials in each experiment. For example, in practice, the length of time before an experiment is complete may be considered an implicit cost of experimentation. We provide results for designs and models based on two to five factors. We also present a potential application of the methods using a military system experiment. [ABSTRACT FROM AUTHOR]

When experimental resources are significantly constrained, resolution V fractional factorial designs are often prohibitively large for experiments with 6 or more factors. Resolution IV designs may also be cost prohibitive, as additional experimentation may be required to de-alias active 2-factor interactions (2FI). This paper introduces 20-run no-confounding screening designs for 6 to 12 factors as alternatives to resolution IV designs. No-confounding designs have orthogonal main effects, and since no 2FI is completely confounded with another main effects or 2FI, the experimental results can be analyzed without follow-on experimentation. The paper concludes with the results of a Monte Carlo simulation used to assess the model-fitting accuracy of the recommended designs. [ABSTRACT FROM AUTHOR]

This paper discusses an application of confidence intervals to the threshold decision value used in logistic regres]sion and discusses its effect on changing the quantification of false positive and false negative errors. In doing this a grey area, in which observations are not classified as success (1) or failure (0), but rather 'uncertain' is developed. The size of this grey area is related to the level of confidence chosen to create the interval around the threshold as well as the quality of logistic regression model fit. This method shows that potential errors may be mitigated. Monte Carlo simulation and an experimental design approach are used to study the relationship between a number of responses relating to classification of observations and the following factors: threshold level, confidence level, noise in the data, and number of observations collected. Copyright © 2015 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

In this paper, we study the relationship between the fitted parameters in a Gaussian process (kriging) model and the complexity of the resulting response surface. This study is done for models with one response and two input variables. An analytical calculation of surface roughness is used as a measure of the complexity of the response surface fit by the Gaussian process model. Our findings indicate that the size of the fitted model parameters as measured across different fits and data sets do not give indication as to the complexity of the surface. We do, however, show that the magnitude of each of the parameters in a single fitted model gives indication about the amount of variability in the direction of that fitted parameter. Copyright © 2015 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Classical D-optimal design is used to create experimental designs for situations in which an underlying system model is known or assumed known. The D-optimal strategy can also be used to add additional experimental runs to an existing design. This paper demonstrates a study of variable choices related to sequential D-optimal design and how those choices influence the D-efficiency of the resulting complete design. The variables studied are total sample size, initial experimental design size, step size, whether or not to include center points in the initial design, and complexity of initial model assumption. The results indicate that increasing total sample size improves the D-efficiency of the design, less effort should be placed in the initial design, especially when the true underlying system model isn't known, and it is better to start off with assuming a simpler model form, rather than a complex model, assuming that the experimenter can reach the true model form during the sequential experiments. Copyright © 2013 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

As unmanned aerial vehicles (UAVs) become more prevalent on the battlefield, ground forces will increasingly have to rely on them for intelligence, surveillance and reconnaissance, as well as target marking and overwatch operations. This paper presents the use of the Situational Awareness for Surveillance and Interdiction Operations simulation analysis tool in conjunction with the design and analysis of experiments to study aspects of UAVs’ surveillance characteristics in conjunction with ground-based interdiction teams to aid in increasing the number of targets cleared from the area of interest. Different teaming strategies and coordination measures between searching and interdicting assets are studied in order to understand the effectiveness of the interdictor possessing an organic tracker UAV. The objective of this research is to quantify the benefit or penalty of an additional UAV asset that is organic to a quick reaction force in the context of the overall surveillance and interdiction operation. [ABSTRACT FROM PUBLISHER]

This article explores a probabilistic formulation for exhaustive search of a bounded area by a single searcher for a single static target. The searcher maintains an aggregate belief of the target's presence or absence in the search area, concluding with a positive or negative search decision on crossing of decision thresholds. The measure of search performance is defined as the expected time until a search decision is made as well as the probability of the search decision being correct. The searcher gathers observations using an imperfect detector, that is, one with false positive and negative errors, and integrates them in an iterative Bayesian manner. Analytic expressions for the Bayesian update recursion of the aggregate belief are given, with theoretical results describing the role of positive and negative detections, as well as sensitivity results for the effect of the detection errors on the aggregate belief evolution. Statistical studies via design of simulation experiments provide insights into the significant search parameters, including imperfect sensor characteristics, initial belief value, search decision threshold values, and the available prior probability information. Regression analysis yields statistical models to provide prescriptive guidance on the search performance as a function of these search parameters.Copyright © 2014 Wiley Periodicals, Inc. Naval Research Logistics 61: 164-178, 2014 [ABSTRACT FROM AUTHOR]

Today's military operating environments are more operationally diverse and technically challenging. Fielding relevant weapons systems to meet the demands of this environment is increasingly difficult, prompting policy shifts that mandate a focus on systems capable of combating a wide threat range. The capabilities-based test and evaluation construct is the Department of the Navy's effort to concentrate on integrated system design with the objective of satisfying a particular operational response (capability) under a robust range of operating conditions. One aspect of capabilities-based test and evaluation is the increased employment of advanced mathematical and statistical techniques in the test and evaluation (T&E) process. This case study illustrates advantages of incorporating these invaluable techniques, like design of experiments and modeling and simulation, within the T&E process. We found through statistical analysis that the application of design of experiment concepts to the System Under Test throughout three primary phases of T&E quantifiably improved the accomplishment of the selected response variable of interest. Copyright © 2012 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

This study aims to understand the effectiveness of typical admissions criteria in identifying students who will complete the Physics Ph.D. Multivariate statistical analysis of roughly one in eight physics Ph.D. students from 2000 to 2010 indicates that the traditional admissions metrics of undergraduate grade point average (GPA) and the Graduate Record Examination (GRE) Quantitative, Verbal, and Physics Subject Tests do not predict completion as effectively admissions committees presume. Significant associations with completion were found for undergraduate GPA in all models and for GRE Quantitative in two of four studied models; GRE Physics and GRE Verbal were not significant in any model. It is notable that completion changed by less than 10% for U.S. physics major test takers scoring in the 10th versus 90th percentile on the Quantitative test. Aside from these limitations in predicting Ph.D. completion overall, overreliance on GRE scores in admissions processes also selects against underrepresented groups.