DATAWorks Sessions Archive

Loss of control in flight has been a leading cause of accidents and incidents in commercial aviation worldwide. The Commercial Aviation Safety Team (CAST) requested studies on virtual day-visual meteorological conditions displays, such as synthetic vision, in order to combat loss of control. Over the last four years NASA has conducted a series of experiments evaluating the efficacy of synthetic vision displays for increased spatial awareness. Commercial pilots with various levels of experience from both domestic and international airlines were used as subjects. This presentation describes the synthetic vision research and how pilot subjects affected experiment design and statistical analyses.

Many software reliability models characterize the number of faults detected during the testing process as a function of testing time, which is performed over multiple stages. Typically, the later stages are progressively more expensive because of the increased number of personnel and equipment required to support testing as the system nears completion. Such transitions from one stage of testing to the next change in the operational environment. One statistical approach to combine software reliability growth models in a manner capable of characterizing multi-stage testing is the concept of a change-point process, where the intensity of a process experiences a distinct change at one or more discrete times during testing. Thus, change-point processes can be used to model change in the failure rate of software due to changes in the testing strategy and environment, integration testing, and resource allocation as it proceeds through multiple stages of testing. This presentation generalizes change-point models to the heterogeneous case, where fault detection before and after a change-point can be characterized by distinct nonhomogeneous Poisson processes (NHPP). Experimental results suggest that heterogeneous change-point models better characterize some failure data sets, which can improve the applicability of software reliability models to large-scale software systems that are tested over multiple stages.

Within NASA’s Air Traffic Management Technology Demonstration # 1 (ATD-1), Interval Management (IM) is a flight deck tool that enables pilots to achieve or maintain a precise in-trail spacing behind a target aircraft. Previous research has shown that violations of aircraft spacing requirements can occur between an IM aircraft and its surrounding non-IM aircraft when it is following a target on a separate route. This talk focuses on the experimental design and analysis of a computer experiment which models the airspace configuration of interest in order to determine airspace/aircraft conditions leading to spacing violations during IM operation. We refer to multi-layered nested continuous factors as those that are continuous and ordered in their selection; they can only be selected sequentially with a level selected for one factor affecting the range of possible values for each subsequently nested factor. While each factor is nested within another factor, the exact nesting relationships have no closed form solution. In this talk, we describe our process of engineering an appropriate space-filling design for this situation. Using this space-filling design and Gaussian process modeling, we found that aircraft delay assignments and wind profiles significantly impact the likelihood of spacing violations and the interruption of IM operations.

Many modern processes generate complex data records not readily analyzed by traditional techniques. For example, a single observation from a process might be a radar signal consisting of n pairs of bivariate data described via some functional relation between reflection and direction. Methods are examined here for detecting changes in such sequences from some known or estimated nominal state. Additionally, estimates of the degree of change (scale, location, frequency, etc.) are desirable and discussed. The proposed methods are designed to take advantage of all available data in a sequence. This can become unwieldy for long sequences of large-sized observations, so dimension reduction techniques are needed. In order for these methods to be as widely applicable as possible, we make limited distributional assumptions and so we propose new nonparametric and Bayesian tools to implement these estimators.

Sample size drives the resources and supports the conclusions of operational test. Power analysis is a common statistical methodology used in planning efforts to justify the number of samples. Power analysis is sensitive to extreme performance (e.g. 0.1% correct responses or 99.999% correct responses) relative to a threshold value, extremes in response variable variability, numbers of factors and levels, system complexity, and a myriad of other design- and system-specific criteria. This discussion will describe considerations (correlation/aliasing, operational significance, thresholds, etc.) and relationships (design, difference to detect, noise, etc.) associated with power. The contribution of power to design selection or adequacy must often be tempered when significant uncertainty or test resources constraints exist. In these situations, other measures of merit and alternative analytical approaches become at least as important as power in the development of designs that achieve the desired technical adequacy. In conclusion, one must understand what power is, what factors influence the calculation, and when to leverage alternative measures of merit.

Frangible Joints are linear pyrotechnic devices used to separate launch vehicle and spacecraft stages and fairings. Advantages of these systems include low mass, low dynamic shock, and low debris. However the primary disadvantage for human space flight applications is the design’s use of a single explosive cord to effect function, rendering the device zero fault tolerant. Commercial company proposals to utilize frangible joints in human space flight applications spurred a NASA Engineering and Safety Center (NESC) assessment of the reliability of frangible joints. Empirical test and LS-DYNA based finite element analysis was used to understand and assess the design and function, and a deterministic system Design of Experiments (dsDOE) study was conducted to assess the sensitivity of function to frangible joint design variables and predict the device’s design reliability. The collaboration between statistical engineering experts and LS-DYNA analysis experts enabled a comprehensive understanding of these devices.

Several optimization models are described for allocating resources to different testing activities in a system’s reliability growth program. These models assume availability of an underlying reliability growth model for the system, and capture the tradeoffs associated with focusing testing resources at various levels (e.g., system, subsystem, component) and/or how to divide resources within a given level. In order to demonstrate insights generated by solving the model, we apply the optimization models to an example series-parallel system in which reliability growth is assumed to follow the Crow/AMSAA reliability growth model. We then demonstrate how the optimization models can be extended to incorporate uncertainty in Crow/AMSAA parameters.

The problem of constructing a design for an experiment when multiple responses are of interest does not have a clear answer, particularly when the response variables are of different types. Planning an experiment for an air-to-air missile simulation, for example, might have the following responses simultaneously: hit or miss the target (a binary response) and the time to acquire the target (a continuous response). With limited time and resources and only one experiment possible, the question of selecting an appropriate design to model both responses is important. In this presentation, we discuss a method for creating designs when two responses, each with a different distribution (normal, binomial, or Poisson), are of interest. We demonstrate the proposed method using various weighting schemes for the two models to show how the designs change as the weighting scheme changes. In addition, we explore the effect of the specified priors for the nonlinear models on these designs.

Spectrograms (i.e., squared magnitude of short-time Fourier transform) are commonly used as features to classify audio signals in the same way that social media companies (e.g., Google, Facebook, Yahoo) use images to classify or automatically tag people in photos. However, a serious problem arises when using spectrograms to classify acoustic signals, in that the user must choose the input parameters (hyperparameters), and such choices can have a drastic effect on the accuracy of the resulting classifier. Further, considering all possible combinations of the hyperparameters is a computationally intractable problem. In this study, we simplify the problem making it computationally tractable, explore the utility of response surface methods for sampling the hyperparameter space, and find that response surface methods are a computationally efficient means of identifying the hyperparameter combinations that are likely to give the best classification results.

One of the major pillars of experimental design is sequential learning. The experimental design should not be viewed as a “one-shot” effort, but rather as a series of experiments where each stage builds upon information learned from the previous study. It is within this realm of sequential learning that experimentation soundly supports the application of the scientific method. This presentation illustrates the value of sequential experimentation and also the connection between the scientific method and experimentation through a discussion of a multi-stage project supported by NASA’s Engineering Safety Center (NESC) where the objective was to assess the safety of composite overwrapped pressure vessels (COPVs). The analytical team was tasked with devising a test plan to model stress rupture failure risk in carbon fiber strands that encase the COPVs with the goal of understanding the reliability of the strands at use conditions for the expected mission life. This presentation highlights the recommended experimental design for the strand tests and then discusses the benefits that resulted from the suggested sequential testing protocol.

Frangible Joints are linear pyrotechnic devices used to separate launch vehicle and spacecraft stages and fairings. Advantages of these systems include low mass, low dynamic shock, and low debris. However the primary disadvantage for human space flight applications is the design’s use of a single explosive cord to effect function, rendering the device zero fault tolerant. Commercial company proposals to utilize frangible joints in human space flight applications spurred a NASA Engineering and Safety Center (NESC) assessment of the reliability of frangible joints. Empirical test and LS-DYNA based finite element analysis was used to understand and assess the design and function, and a deterministic system Design of Experiments (dsDOE) study was conducted to assess the sensitivity of function to frangible joint design variables and predict the device’s design reliability. The collaboration between statistical engineering experts and LS-DYNA analysis experts enabled a comprehensive understanding of these devices.

Uncertainty appears in many aspects of systems design including stochastic design parameters, simulation inputs, and forcing functions. Uncertainty Quantification (UQ) has emerged as the science of quantitative characterization and reduction of uncertainties in both simulation and test results. UQ is a multidisciplinary field with a broad base of methods including sensitivity analysis, statistical calibration, uncertainty propagation, and inverse analysis. Because of their ability to bring greater degrees of confidence to decisions, uncertainty quantification methods are playing a greater role in test, evaluation, and modeling and simulation in defense and aerospace. The value of UQ comes with better understanding of risk from assessing the uncertainty in test and modeling and simulation results. The presentation will provide an overview of UQ and then discuss the use of some advanced statistical methods, including DOEs and emulation for multiple simulation solvers and statistical calibration, for efficiently quantifying uncertainties. These statistical methods effectively link test, evaluation and modeling and simulation by coordinating the valuation of uncertainties, simplifying verification and validation activities.

Since its publication, Statistical Methods for Reliability Data by W. Q. Meeker and L. A. Escobar has been recognized as a foundational resource in analyzing failure time to and survival data. Along with the text, the authors provided an S-Plus software package, called SPLIDA, to help readers utilize the methods presented in the text. Today, R is the most popular statistical computing language in the world, largely supplanting S-Plus. The SMRD package is the result of a multi-year effort to completely rebuild SPLIDA, to take advantage of the improved graphics and workflow capabilities available in R. This presentation introduces the SMRD package, outlines the improvements and shows how the package works seamlessly with the rmarkdown and shiny packages to dramatically speed up your workflow. The presentation concludes with a discussion on what improvements still need to be made prior to publishing the package on the CRAN.

Jones and Nachtsheim (2011) introduced a class of three-level screening designs called definitive screening designs (DSDs). The structure of these designs results in the statistical independence of main effects and two-factor interactions; the absence of complete confounding among two-factor interactions; and the ability to estimate all quadratic effects. Because quadratic effects can be estimated, DSDs can allow for the screening and optimization of a system to be performed in one step, but only when the number of terms found to be active during the screening phase of analysis is less than about half the number or runs required by the DSD (Errore, et al., 2016). Otherwise, estimation of second-order models requires augmentation of the DSD. In this paper we explore the construction of series of augmented designs, moving from the starting DSD to designs capable of estimating the full second-order model. We use power calculations, model-robustness criteria, and model-discrimination criteria to determine the number of runs by which to augment in order to identify the active second-order effects with high probability.

The process for testing military systems which are largely software intensive involves techniques and procedures often different from those for hardware-based systems. Much of the testing can be performed in laboratories at many of the acquisition stages, up to operational testing. Testing software systems is not different from testing hardware-based systems in that testing earlier and more intensively benefits the acquisition program in the long run. Automated testing of software systems enables more frequent and more extensive testing, allowing for earlier discovery of errors and faults in the code. Automated testing is beneficial for unit, integrated, functional and performance testing, but there are costs associated with automation tool license fees, specialized manpower, and the time to prepare and maintain the automation scripts. This presentation discusses some of the features unique to automated software testing and offers a framework organizations can implement to make the business case for, to organize for, and to execute and benefit from automating the right aspects of their testing needs. Automation has many benefits in saving time and money, but is most valuable in freeing test resources to perform higher value tasks.

This presentation will provide a brief overview of sensitivity testing, and emphasize applications to several products and system of importance to the Defense as well as private industry, including Insensitive Energetics, Ballistic testing of protective armor, testing of munition fuzes and Microelectromechanical Systems (MEMS) components, and safety testing of high-pressure test ammunition, and packaging for high-value materials.

“Machine learning is quickly gaining importance in being able to infer meaning from large, high-dimensional datasets. It has even demonstrated performance meeting or exceeding human capabilities in conducting a particular set of tasks such as speech recognition and image recognition. Employing these machine learning capabilities can lead to increased efficiency in data collection, processing, and analysis. Presenters will provide an overview of common examples of supervised and unsupervised learning tasks and algorithms as an introduction to those without experience in machine learning. Presenters will also provide motivation for machine learning tasks and algorithms in a variety of test and evaluation settings. For example, in both developmental and operational test, restrictions on instrumentation, number of sorties, and the amount of time allocated to analyze collected data make data analysis challenging. When instrumentation is unavailable or fails, a common back-up data source is an over-the-shoulder video recording or recordings of aircraft intercom and radio transmissions, which traditionally are tedious to analyze. Machine learning based image and speech recognition algorithms can assist in extracting information quickly from hours of video and audio recordings. Additionally, unsupervised learning techniques may be used to aid in the identification of influences of logged or uncontrollable factors in many test and evaluation settings. Presenters will provide a potential example for the application of unsupervised learning techniques to test and evaluation.”

Answers to real world questions are often based on the use of judiciously chosen mathematical/statistical/physical models. In particular, assessment of failure probabilities of physical systems rely heavily on such models. Since no model describes the real world exactly, sensitivity analyses are conducted to examine influences of (small) perturbations of an assumed model. In this talk we present a structured approach, using an “Assumptions Lattice” and corresponding “Uncertainty Pyramid”, for transparently conveying the influence of various assumptions on analysis conclusions. We illustrate this process in the context of a simple multicomponent system.

One of the most powerful features of Bayesian analyses is the ability to combine multiple sources of information in a principled way to perform inference. For example, this feature can be particularly valuable in assessing the reliability of systems where testing is limited for some reason (e.g., expense, treaty). At their most basic, Bayesian methods for reliability develop informative prior distributions using expert judgment or similar systems. Appropriate models allow the incorporation of many other sources of information, including historical data, information from similar systems, and computer models. I will introduce the approach and then consider examples from defense acquisition and lifecycle extension, focusing on the strengths and weaknesses of the Bayesian analyses.