Search NASA⌕ Search

SEARCH · Search NASA

Results for “Bayesian method”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,621 records · Page 90

Image Informatics for Clinical and Preclinical Biomedical Analysis

Biomedical informatics refers to the study of the application of computationaland statistical algorithms, data structures, and methods to improve communication, understanding and management of biomedical information. Our objective through this chapter is to describe and demonstrate our research in the use of biomedical image databases - in both preclinical and clinical settings - to classify, predict, research, diagnose, and otherwise learn from the informational content encapsulated in historical image repositories. This will be accomplished by detailing our approach of describing image content in a Bayesian probabilistic framework to achieve learning from retrieved populations of similar images. We will use specific examples from two biomedical applications to describe anatomic segmentation, statistical featuregeneration and indexing, efficient retrieval architectures, and predictive results.

60 APPLIED LIFE SCIENCES↗

An iterative particle filter approach for coupled hydro-geophysical inversion of a controlled infiltration experiment

The modeling of unsaturated groundwater flow is affected by a high degree of uncertainty related to both measurement and model errors. Geophysical methods such as Electrical Resistivity Tomography (ERT) can provide useful indirect information on the hydrological processes occurring in the vadose zone. In this paper, we propose and test an iterated particle filter method to solve the coupled hydrogeophysical inverse problem. We focus on an infiltration test monitored by time-lapse ERT and modeled using Richards equation. The goal is to identify hydrological model parameters from ERT electrical potential measurements. Traditional uncoupled inversion relies on the solution of two sequential inverse problems, the first one applied to the ERT measurements, the second one to Richards equation. This approach does not ensure an accurate quantitative description of the physical state, typically violating mass balance. To avoid one of these two inversions and incorporate in the process more physical simulation constraints, we cast the problem within the framework of a SIR (Sequential Importance Resampling) data assimilation approach that uses a Richards equation solver to model the hydrological dynamics and a forward ERT simulator combined with Archie's law to serve as measurement model. ERT observations are then used to update the state of the system as well as to estimate the model parameters and their posterior distribution. The limitations of the traditional sequential Bayesian approach are investigated and an innovative iterative approach is proposed to estimate the model parameters with high accuracy. The numerical properties of the developed algorithm are verified on both homogeneous and heterogeneous synthetic test cases based on a real-world field experiment.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Conjunctive management of surface and groundwater resources under projected future climate change scenarios

Our study introduces a mixed integer linear fractional programming (MILFP) method to optimize conjunctive use of future surface water and groundwater resources under projected climate change scenarios. The conjunctive management model maximizes the ratio of groundwater usage to reservoir water usage. Future inflows to the reservoirs were estimated from the future runoffs projected through hydroclimate modeling considering the Variable Infiltration Capacity model, and 11 sets of downscaled Coupled Model Intercomparison Project phase 5 global climate model projections. Bayesian model averaging was adopted to quantify uncertainty in future runoff projections and reservoir inflow projections due to uncertain future climate projections. Optimized conjunctive management solutions were investigated for a water supply network in northern Louisiana which includes the Sparta aquifer. Runoff projections under climate change scenarios indicate that runoff will likely decrease in winter and increase in other seasons. Ultimately, results from the developed conjunctive management model with MILFP indicate that the future reservoir water, even at 2.5% low inflow cumulative probability level, could counterbalance groundwater pumping reduction to satisfy demands while improving the Sparta aquifer through conditional groundwater head constraint.

54 ENVIRONMENTAL SCIENCES↗

Phylogeny of Bacterial and Archaeal Genomes Using Conserved Genes: Supertrees and Supermatrices

Over 3000 microbial (bacterial and archaeal) genomes have been made publically available to date, providing an unprecedented opportunity to examine evolutionary genomic trends and offering valuable reference data for a variety of other studies such as metagenomics. The utility of these genome sequences is greatly enhanced when we have an understanding of how they are phylogenetically related to each other. Therefore, we here describe our efforts to reconstruct the phylogeny of all available bacterial and archaeal genomes. We identified 24, single-copy, ubiquitous genes suitable for this phylogenetic analysis. We used two approaches to combine the data for the 24 genes. First, we concatenated alignments of all genes into a single alignment from which a Maximum Likelihood (ML) tree was inferred using RAxML. Second, we used a relatively new approach to combining gene data, Bayesian Concordance Analysis (BCA), as implemented in the BUCKy software, in which the results of 24 single-gene phylogenetic analyses are used to generate a ‘‘primary concordance’’ tree. A comparison of the concatenated ML tree and the primary concordance (BUCKy) tree reveals that the two approaches give similar results, relative to a phylogenetic tree inferred from the 16S rRNA gene. After comparing the results and the methods used, we conclude that the current best approach for generating a single phylogenetic tree, suitable for use as a reference phylogeny for comparative analyses, is to perform a maximum likelihood analysis of a concatenated alignment of conserved, single-copy genes.

59 BASIC BIOLOGICAL SCIENCES↗

A Bayesian Learning Approach to Wireless Outdoor Heatmap Construction using Deep Gaussian Process

We present a novel Bayesian learning approach to outdoor radio heatmap construction utilizing deep Gaussian process (GP). The proposed approach employs a two-layer hierarchy which consists of two cascaded Gaussian processes that are capable of modeling more complex input-output relations than standard single-layer Gaussian processes. Since deriving the exact model likelihood is challenging, a lower bound is optimized instead so that gradient descent-based methods can be performed to find out the optimal model parameters. Typically, inducing points are used in GPs to facilitate low-rank approximation of covariance (kernel) matrices for computation speedup. However, the inaccuracy induced by inducing points can accumulate when stacking multiple layers of GP which may hinder the performance of deep GP. Moreover, since inducing points need to be learned, having them at all layers of deep GP also incurs computational burden. To overcome the above challenges, in contrast to the canonical deep GP model, we use a modified architecture where a full standard GP resides in the first layer and inducing points are only introduced for the second layer. This modified architecture strikes a balance between model accuracy and training complexity. In the proposed model, the noise parameter of the first GP layer is also eliminated to improve the training efficiency as the noise parameter at the output of the second layer suffices to model the uncertainty in the output. The proposed approach is evaluated on real-world datasets, in the form of location-Received Signal Strength (RSS) pairs, collected from the Platform for Open Wireless Data-driven Experimental Research (POWDER) located at the campus of the University of Utah. Experiment results show that the proposed approach can achieve smaller prediction errors on various training and testing data configurations than DNN-based and GP-based methods.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Discrete-Direct Model Calibration and Uncertainty Propagation Method Confirmed on Multi-Parameter Plasticity Model Calibrated to Sparse Random Field Data

A discrete direct (DD) model calibration and uncertainty propagation approach is explained and demonstrated on a 4-parameter Johnson-Cook (J-C) strain-rate dependent material strength model for an aluminum alloy. The methodology's performance is characterized in many trials involving four random realizations of strain-rate dependent material-test data curves per trial, drawn from a large synthetic population. The J-C model is calibrated to particular combinations of the data curves to obtain calibration parameter sets which are then propagated to “Can Crush” structural model predictions to produce samples of predicted response variability. These are processed with appropriate sparse-sample uncertainty quantification (UQ) methods to estimate various statistics of response with an appropriate level of conservatism. This is tested on 16 output quantities (von Mises stresses and equivalent plastic strains) and it is shown that important statistics of the true variabilities of the 16 quantities are bounded with a high success rate that is reasonably predictable and controllable. The DD approach has several advantages over other calibration-UQ approaches like Bayesian inference for capturing and utilizing the information obtained from typically small numbers of replicate experiments in model calibration situations—especially when sparse replicate functional data are involved like force–displacement curves from material tests. The DD methodology is straightforward and efficient for calibration and propagation problems involving aleatory and epistemic uncertainties in calibration experiments, models, and procedures.

42 ENGINEERING↗

SU-E-T-49: Automatic Beam Angle Determination for Lung IMRT Planning Using a Beam Configuration Atlas

Purpose: To present a technique to automatically determine beam angle configurations for lung IMRT planning based on the patient-specific anatomy and tumor geometry. Methods: The relationship between individual patient anatomy and proper beam configurations was learned from high quality clinical plans in three steps. First, a beam configuration atlas was obtained by classifying 60 lung IMRT plans into 6 beam configuration clusters based on a dissimilarity measure defined between different beam configurations. A beam configuration template was extracted from each cluster to form an atlas. Second, a beam efficiency index map (EI map) was constructed to characterize the geometry of the tumor relative to the lungs, the body and other OARs along each candidate beam direction. Finally, the EI maps of the clinical cases and the cluster assignments of their beam configurations were paired to train a Bayesian classification model. This technique was validated by leave-one-out cross validation with 16 cases randomly selected from the original dataset. An IMRT plan (autobeam plan) for each test case was generated using the beam configuration template according to the cluster assignment given by the model and was compared with the corresponding clinical plan. Results: The dosimetric parameters (mean±S.D. in percentage of prescription dose) in the auto-beam plans and in the clinical plans, respectively, and the p-values by a paired ttest (in parenthesis) are: lung Dmean: 16.3±9.3, 18.6±7.4 (0.48), esophagus Dmean: 28.4±18, 30.7±19.3 (0.02), Heart Dmean: 21.5±17.5,21.1±17.2 (0.76), Spinal Cord D2%: 48±23, 51.2±21.8 (0.01), PTV dose homogeneity (D2%–D99%): 22±27.4, 20.4±12.8 (0.10).The dose reductions by the autobeam plans in esophagus Dmean and cord D02 are statistically significant but the differences (<4%) may not be clinically significant. The other dosimetric parameters are not statistically different. Conclusion: Plans generated by the automatic beam angle determination method can achieve dosimetric quality equivalent to that of clinical plans. Partially supported by NIH/NCI under grant #R21CA161389 and a master research grant by Varian Medical System.

60 APPLIED LIFE SCIENCES↗

Power generation forecasting for solar plants based on Dynamic Bayesian networks by fusing multi-source information

A Dynamic Bayesian network (DBN) model for solar power generation forecasting in solar plants is proposed in this paper. The key idea is to fuse sensor data, operational indicators, meteorological data, lagged output power information, and model errors for more accurate short-term (e.g., hours) and mid-term (e.g., days to weeks) power generation forecasting. The proposed DBN augments automated data-driven structure learning with expert knowledge encoding using continuous and categorical data given constraints to represent causal relationships within a solar inverter system. Additionally, an error compensation mechanism is proposed to capture temporal fluctuation. The effectiveness of the DBN on solar power generation forecasting was evaluated by rolling window analysis with one-year testing data collected from a local solar plant. The proposed DBN is compared with four state-of-art methods including support-vector regression (SVR), k-nearest neighbors (kNN), artificial neural network (ANN), and long short-term memory (LSTM) models. The result show that the proposed DBN achieves better accuracy in general, and it is not as data-hungry as some neural network-based models. The proposed DBN is also shown to have robust and consistent forecasting power with different forecasting horizons. The accuracy is 92% - 95% from one hour to one week ahead forecasting.

14 SOLAR ENERGY↗

Surface and subsurface oceanographic features drive forage fish distributions and aggregations: Implications for prey availability to top predators in the US Northeast Shelf ecosystem

Abstract Forage fishes are a critical food web link in marine ecosystems, aggregating in a hierarchical patch structure over multiple spatial and temporal scales. Surface‐level forage fish aggregations (FFAs) represent a concentrated source of prey available to surface‐ and shallow‐foraging marine predators. Existing survey and analysis methods are often imperfect for studying forage fishes at scales appropriate to foraging predators, making it difficult to quantify predator–prey interactions. In many cases, general distributions of forage fish species are known; however, these may not represent surface‐level prey availability to predators. Likewise, we lack an understanding of the oceanographic drivers of spatial patterns of prey aggregation and availability or forage fish community patterns. Specifically, we applied Bayesian joint species distribution models to bottom trawl survey data to assess species‐ and community‐level forage fish distribution patterns across the US Northeast Continental Shelf (NES) ecosystem. Aerial digital surveys gathered data on surface FFAs at two project sites within the NES, which we used in a spatially explicit hierarchical Bayesian model to estimate the abundance and size of surface FFAs. We used these models to examine the oceanographic drivers of forage fish distributions and aggregations. Our results suggest that, in the NES, regions of high community species richness are spatially consistent with regions of high surface FFA abundance. Bathymetric depth drove both patterns, while subsurface features, such as mixed layer depth, primarily influenced aggregation behavior and surface features, such as sea surface temperature, sub‐mesoscale eddies, and fronts influenced forage fish diversity. In combination, these models help quantify the availability of forage fishes to marine predators and represent a novel application of spatial models to aerial digital survey data.

54 ENVIRONMENTAL SCIENCES↗

Ripening of Rh Nanoparticle Catalysts in Reverse Water–Gas Shift via a Data-Driven Model Combining Physics, Theory, and Experiment

Degradation via sintering is an ongoing challenge that impedes the broad commercial success of supported metallic nanoparticle catalysts. To mitigate degradation via informed catalyst design and process operations, here we aim to disambiguate the underlying mechanisms of sintering by combining theory and experiment in a quantitative framework. While mechanistic sintering models exist, they only model a single sintering pathway, even though multiple sintering mechanisms can occur simultaneously or dominate at different stages of the process. Data-driven machine learning models have emerged as a means to represent complex processes through data regression. However, machine learning models have very large data needs and lack mechanistic insights due to their black-box encoding. To develop an interpretive model of catalyst degradation via sintering, we constructed a hybrid model combining mechanistic “physics-based” models and data-driven methods to obtain both reliable predictions and mechanistic insights regarding experimentally observed sintering phenomena. Focusing on nanoparticle sintering in the Rh–TiO 2 catalyst for the reverse water–gas shift (RWGS) reaction, the hybrid model couples a mechanistic term for Ostwald ripening with energy values calculated via density functional theory (DFT) with a parametric, data-driven discrepancy function term for unmodeled mechanisms. The hybrid model is trained using Bayesian inference with data collected from small-angle X-ray scattering (SAXS) in situ experiments wherein average nanoparticle diameter versus time was measured at three relevant operating temperatures. The calibrated hybrid model results show that an Ostwald ripening-only model parameterized with fixed DFT energies does not fully capture the time and temperature dependence of the SAXS-observed sintering kinetics, and that an additional functional contribution, or DFT energy calibration, is required to reconcile simulation and experiment. Analysis of the hybrid-model error confirms that the hybrid model outperforms both the purely mechanistic and purely data-driven alternatives in terms of expected predictive accuracy for time-evolving average particle sizes. Furthermore, the results support the hypothesis that the Ostwald ripening mechanism is less important for explaining the sintering phenomena as operating temperature increases under an assumed fixed DFT parameterization. This could be explained in one of two ways: either latent, unmodeled sintering mechanisms dominate at higher temperatures, or the DFT uncertainty increases with temperature. The proposed modeling approach directly links theory to experiments and simulations via a statistical hybrid modeling framework and can be extended to other catalytic systems to improve predictive models and mechanistic understanding.

Bayesian hybrid modeling↗

Multilevel Hierarchical Decomposition of Finite Element White Noise with Application to Multilevel Markov Chain Monte Carlo

In this work we develop a new hierarchical multilevel approach to generate Gaussian random field realizations in an algorithmically scalable manner that is well suited to incorporating into multilevel Markov chain Monte Carlo (MCMC) algorithms. This approach builds off of other partial differential equation (PDE) approaches for generating Gaussian random field realizations; in particular, a single field realization may be formed by solving a reaction-diffusion PDE with a spatial white noise source function as the right-hand side. While these approaches have been explored to accelerate forward uncertainty quantification tasks, e.g., multilevel Monte Carlo, the previous constructions are not directly applicable to multilevel MCMC frameworks which build fine-scale random fields in a hierarchical fashion from coarse-scale random fields. Our new hierarchical multilevel method relies on a hierarchical decomposition of the white noise source function in $L^2$ which allows us to form Gaussian random field realizations across multiple levels of discretization in a way that fits into multilevel MCMC algorithmic frameworks. After presenting our main theoretical results and numerical scaling results to showcase the utility of this new hierarchical PDE method for generating Gaussian random field realizations, this method is tested on a four-level MCMC algorithm to explore its feasibility.

algebraic multigrid↗

Landmark-Warped Emulators for Models with Misaligned Functional Response

Many computer models output functional data, and in some cases, these functional data have similar, but misaligned, shape characteristics. In this paper, we introduce a general approach for building emulators for computer models that output misaligned functional data when key values in the functional response (landmarks) can be easily identified. This approach has two main parts: modeling the aligned (using the landmarks) functional data, and modeling the functions that map the misaligned data to the aligned space (warping functions). As the warping functions are required to be monotonic, we give special attention to modeling monotonic functional response data. We discuss how our approach can be easily applied for a variety of typical emulators, such as Gaussian processes, Bayesian multivariate adaptive regression splines, and Bayesian additive regression trees, and how sensitivity analysis can be performed. We demonstrate our approach by building emulators for two applications: (1) a high-energy-density physics computer model used to simulate inertial confinement fusion ignition experiments, where model outputs are highly misaligned, and (2) a multiphysics continuum hydrocode used to simulate high-velocity impact experiments, where model outputs are only slightly misaligned. In case (1) traditional methods cannot be applied, while in (2) they can be applied, but the proposed method performs significantly better.

97 MATHEMATICS AND COMPUTING↗

OGLE-2012-BLG-0724Lb: a Saturn-mass planet around an M dwarf

We report the discovery of a planet by the microlensing method, OGLE-2012-BLG-0724Lb. Although the duration of the planetary signal for this event was one of the shortest seen for a planetary event, the anomaly was well covered thanks to high-cadence observations taken by the survey groups OGLE and MOA. By analyzing the light curve, this planetary system is found to have a mass ratio q=(1.58±0.15)×10{sup −3}. By conducting a Bayesian analysis, we estimate that the host star is an M dwarf with a mass of M{sub L}=0.29{sub −0.16}{sup +0.33} M{sub ⊙} located at D{sub L}=6.7{sub −1.2}{sup +1.1} kpc away from the Earth and the companion’s mass is m{sub P}=0.47{sub −0.26}{sup +0.54} M{sub Jup}. The projected planet–host separation is a{sub ⊥}=1.6{sub −0.3}{sup +0.4} AU. Because the lens–source relative proper motion is relatively high, future high-resolution images would detect the lens host star and determine the lens properties uniquely. This system is likely a Saturn-mass exoplanet around an M dwarf, and such systems are commonly detected by gravitational microlensing. This adds another example of a possible pileup of sub-Jupiters (0.2<m{sub P}/M{sub Jup}<1) in contrast to a lack of Jupiters (∼1--2 M{sub Jup}) around M dwarfs, supporting the prediction by core accretion models that Jupiter-mass or more massive planets are unlikely to form around M dwarfs.

79 ASTRONOMY AND ASTROPHYSICS↗

Experimental and Sampling Design for the INL-2 Sample Collection Operational Test

This report describes the experimental and sampling design developed to assess sampling approaches and methods for detecting contamination in a building and clearing the building for use after decontamination. An Idaho National Laboratory (INL) building will be contaminated with BG (Bacillus globigii, renamed Bacillus atrophaeus), a simulant for Bacillus anthracis (BA). The contamination, sampling, decontamination, and re-sampling will occur per the experimental and sampling design. This INL-2 Sample Collection Operational Test is being planned by the Validated Sampling Plan Working Group (VSPWG). The primary objectives are: 1) Evaluate judgmental and probabilistic sampling for characterization as well as probabilistic and combined (judgment and probabilistic) sampling approaches for clearance, 2) Conduct these evaluations for gradient contamination (from low or moderate down to absent or undetectable) for different initial concentrations of the contaminant, 3) Explore judgment composite sampling approaches to reduce sample numbers, 4) Collect baseline data to serve as an indication of the actual levels of contamination in the tests. A combined judgmental and random (CJR) approach uses Bayesian methodology to combine judgmental and probabilistic samples to make clearance statements of the form "X% confidence that at least Y% of an area does not contain detectable contamination” (X%/Y% clearance statements). The INL-2 experimental design has five test events, which 1) vary the floor of the INL building on which the contaminant will be released, 2) provide for varying the amount of contaminant released to obtain desired concentration gradients, and 3) investigate overt as well as covert release of contaminants. Desirable contaminant gradients would have moderate to low concentrations of contaminant in rooms near the release point, with concentrations down to zero in other rooms. Such gradients would provide a range of contamination levels to challenge the sampling, sample extraction, and analytical methods to be used in the INL-2 study. For each of the five test events, the specified floor of the INL building will be contaminated with BG using a point-release device located in the room specified in the experimental design. Then quality control (QC), reference material coupon (RMC), judgmental, and probabilistic samples will be collected according to the sampling plan for each test event. Judgmental samples will be selected based on professional judgment and prior information. Probabilistic samples were selected with a random aspect and in sufficient numbers to provide desired confidence for detecting contamination or clearing uncontaminated (or decontaminated) areas. Following sample collection for a given test event, the INL building will be decontaminated. For possibly contaminated areas, the numbers of probabilistic samples were chosen to provide 95% confidence of detecting contaminated areas of specified sizes. For rooms that may be uncontaminated following a contamination event, or for whole floors after decontamination, the numbers of judgmental and probabilistic samples were chosen using the CJR approach. The numbers of samples were chosen to support making X%/Y% clearance statements with X = 95% or 99% and Y = 96% or 97%. The experimental and sampling design also provides for making X%/Y% clearance statements using only probabilistic samples. For each test event, the numbers of characterization and clearance samples were selected within limits based on operational considerations while still maintaining high confidence for detection and clearance aspects. The sampling design for all five test events contains 2085 samples, with 1142 after contamination and 943 after decontamination. These numbers include QC, RMC, judgmental, and probabilistic samples. The experimental and sampling design specified in this report provides a good statistical foundation for achieving the objectives of the INL-2 study.

59 BASIC BIOLOGICAL SCIENCES↗

Machine Learning-Guided Optimization of SABRE Hyperpolarization for α-Ketoglutarate in Acetone–Water

Signal amplification by reversible exchange (SABRE) is a hyperpolarization method that polarizes target nuclei of metabolites quickly and efficiently. Recent SABRE advances, including Ace-SABRE, yield biocompatible, aqueous solutions of hyperpolarized markers for metabolic monitoring. Building on recent advancements, expanding the substrate scope of Ace-SABRE is desirable. However, SABRE polarization is sensitive to many different parameters; therefore, traditional optimization approaches are experimentally time-consuming. In this proof-of-concept application of machine learning (ML), Bayesian optimization (BO) is used for four important input parameters to model the complex SABRE dynamics while saving experimental time. The presented ML model also provides chemical insights that enable predictions of sample compositions for increased polarization levels. In this paper, we transition from an original average free polarization of p = ∼0.90% to a maximum observed free polarization of p = ∼6.6% for 1- 13 C alpha-ketoglutarate (AKG) with 13 C at natural abundance, utilizing both direct outputs as well as chemical insights revealed by the ML model.

Catalysts↗

Assessing the Quality of Bioforensic Signatures

We present a mathematical framework for assessing the quality of signature systems in terms of fidelity, cost, risk, and utility—a method we refer to as Signature Quality Metrics (SQM). We demonstrate the SQM approach by assessing the quality of a signature system designed to predict the culture medium used to grow a microorganism. The system consists of four chemical assays designed to identify various ingredients that could be used to produce the culture medium. The analytical measurements resulting from any combination of these four assays can be used in a Bayesian network to predict the probabilities that the microorganism was grown using one of eleven culture media. We evaluated fifteen combinations of the signature system by removing one or more of the assays from the Bayes network. We demonstrated that SQM can be used to distinguish between the various combinations in terms of attributes of interest. The approach assisted in clearly identifying assays that were least informative, largely in part because they only could discriminate between very few culture media, and in particular, culture media that are rarely used. There are limitations associated with the data that were used to train and test the signature system. Consequently, our intent is not to draw formal conclusions regarding this particular bioforensic system, but rather to illustrate an analytical approach that could be useful in comparing one signature system to another.

Bayesian networks↗

Towards a New Supply Chain Cybersecurity Risk Analysis Technique

Supply chain cyber-attacks, such as the SolarWinds Orion attack, are occurring with greater frequency. These attacks compromise a digital device before it is sent to customers, bypassing traditional security controls to remain persistent and undetected in operational environments. While supply chain attacks are prevalent, methods for analyzing the risk of these attacks are currently unavailable. This paper proposes new supply chain cyber-attack difficulty and risk metrics to evaluate the relative risk of an attack throughout the supply chain lifecycle. Difficulty metrics for each stakeholder in a digital device’s supply chain (e.g., hardware manufacturing, firmware development, software development, storage, and distribution entities) are calculated using scores from cybersecurity maturity questionnaires in a Bayesian Network leaky Noisy-MAX model. These difficulty metrics are then used to calculate an overall supply chain cyber-attack risk. Vulnerability and recoverability metrics are also proposed to evaluate the relative stakeholder influence in the attack risk. These proposed relative risk metrics enable continuous supply chain monitoring, provide decision-makers with information necessary for improved supplier selection, and help drive improvements in the cybersecurity posture of the stakeholders in their supply chain.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Statistical tools for prognostics and health management of complex systems

Prognostics and Health Management (PHM) is increasingly important for understanding and managing today's complex systems. These systems are typically mission- or safety-critical, expensive to replace, and operate in environments where reliability and cost-effectiveness are a priority. We present background on PHM and a suite of applicable statistical tools and methods. Our primary focus is on predicting future states of the system (e.g., the probability of being operational at a future time, or the expected remaining system life) using heterogeneous data from a variety of sources. We discuss component reliability models incorporating physical understanding, condition measurements from sensors, and environmental covariates; system reliability models that allow prediction of system failure time distributions from component failure models; and the use of Bayesian techniques to incorporate expert judgments into component and system models.

42 ENGINEERING↗