Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Decision Boundaries”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Assessing decision boundaries under uncertainty

In order to make design decisions, engineers may seek to identify regions of the design domain that are acceptable in a computationally efficient manner. A design is typically considered acceptable if its reliability with respect to parametric uncertainty exceeds the designer’s desired level of confidence. Despite major advancements in reliability estimation and in design classification via decision boundary estimation, the current literature still lacks a design classification strategy that incorporates parametric uncertainty and desired design confidence. To address this gap, this paper offers a novel interpretation of the acceptance region by defining the decision boundary as the hypersurface which isolates the designs that exceed a user-defined level of confidence given parametric uncertainty. This work addresses the construction of this novel decision boundary using computationally efficient algorithms that were developed for reliability analysis and decision boundary estimation. The approach proposed in this paper is verified on two physical examples from structural and thermal analysis using Support Vector Machines and Efficient Global Optimization-based contour estimation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Persistent Classification: Understanding Adversarial Attacks by Studying Decision Boundary Dynamics

ABSTRACT There are a number of hypotheses underlying the existence of adversarial examples for classification problems. These include the high‐dimensionality of the data, the high codimension in the ambient space of the data manifolds of interest, and that the structure of machine learning models may encourage classifiers to develop decision boundaries close to data points. This article proposes a new framework for studying adversarial examples that does not depend directly on the distance to the decision boundary. Similarly to the smoothed classifier literature, we define a (natural or adversarial) data point to be ( γ , σ)‐stable if the probability of the same classification is at least for points sampled in a Gaussian neighborhood of the point with a given standard deviation . We focus on studying the differences between persistence metrics along interpolants of natural and adversarial points. We show that adversarial examples have significantly lower persistence than natural examples for large neural networks in the context of the MNIST and ImageNet datasets. We connect this lack of persistence with decision boundary geometry by measuring angles of interpolants with respect to decision boundaries. Finally, we connect this approach with robustness by developing a manifold alignment gradient metric and demonstrating the increase in robustness that can be achieved when training with the addition of this metric.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

The transition from resistance to acceptance: Managing a marine invasive species in a changing world

Abstract Marine invasive species can transform coastal ecosystems, yet mitigating their effects can be difficult, and even impractical. Often, marine invasive species are managed at poorly matched spatial scales, and at the same time, rates of spread and establishment are increasing under climate change and can outpace resources available for population suppression. These circumstances challenge traditional conservation goals of maintaining a historic environmental state, especially for a species like the European green crab ( Carcinus maenas ), a formidable invader with few examples of successful long‐term removal programs. A management paradigm where decision alternatives include resisting or accepting a new ecological trajectory may be needed. We apply mathematical concepts from decision theory to develop a quantitative framework for navigating management decisions in this new resist‐accept paradigm. We develop a model of European green crab growth, removal and colonization, and we find optimal levels of removal effort that minimize both ecological change and removal cost. We establish a benchmark of colonization pressure at which green crab density becomes decoupled from a decision maker's actions, such that population control can no longer shape the invasion trajectory. For informing the decision boundary between resistance and acceptance, our results highlight that a decision maker's understanding of how removal cost scales with removal effort is more important than understanding the density‐impact relationship. We show that assuming stationary system dynamics can result in sub‐optimal levels of species removal effort, highlighting the importance of developing anticipatory management strategies by accounting for non‐stationary dynamics. Policy implications . For marine invasive species that can disperse across long distances and recolonize rapidly after removal, the focus of conservation policy should shift away from understanding how to resist change to understanding when to stop resisting change. Navigating this decision problem involves trade‐offs among competing objectives, highlighting the need for structured approaches to elicit objective weights that reflect the values of the decision maker. For natural resource managers facing possible ecosystem transformation, this decision framework can enable proactive and strategic decisions made under uncertainty in a changing world.

Keller, Abigail G. [Department of Environment Scie↗
DOE OSTI2025

Hyperplane decision trees as piecewise linear surrogate models for chemical process design

Recent trends in chemical engineering research point towards an increasing reliance on data-driven modeling approaches. Neural networks, for instance, have proven to be accurate when data is plentiful and high-dimensional, but in many cases, they require computationally-intensive training procedures. Here, in this work, we describe hyperplane decision trees (HT) as a highly expressive and low-compute machine learning model architecture. These models are locally linear and have linear decision boundaries, resulting in a piecewise linear model of the data. This property allows them to be converted into mixed-integer linear constraints which can be globally optimized. Our open-source PyTorch implementation of this method is a fast, flexible, and accessible way to build accurate piecewise linear models of data.

Decision trees↗
DOE OSTI2025

Robustness of Deep Learning Classification to Adversarial Input on GPUs: Asynchronous Parallel Accumulation Is a Source of Vulnerability

The ability of machine learning (ML) classification models to resist small, targeted input perturbations—known as adversarial attacks—is a key measure of their safety and reliability. We show that floating-point non associativity (FPNA) coupled with asynchronous parallel programming on GPUs is sufficient to result in misclassification, without any perturbation to the input. Additionally, we show that this misclassification is particularly significant for inputs close to the decision boundary and that standard adversarial robustness results may be overestimated up to 4.6 when not considering machine-level details. We first study a linear classifier, before focusing on standard Graph Neural Network (GNN) architectures and datasets used in robustness assessments. We develop a novel black-box attack using Bayesian optimization to discover external workloads that can change the instruction scheduling which bias the output of reductions on GPUs and reliably lead to misclassification. Motivated by these results, we present a new learnable permutation (LP) gradient-based approach to learning floating-point operation orderings that lead to misclassifications. The LP approach provides a worst-case estimate in a computationally efficient manner, avoiding the need to run identical experiments tens of thousands of times over a potentially large set of possible GPU states or architectures. Finally, using instrumentation-based testing, we investigate parallel reduction ordering across different GPU architectures under external background workloads, when utilizing multi-GPU virtualization, and when applying power capping. Our results demonstrate that parallel reduction ordering varies significantly across architectures under the first two conditions, substantially increasing the search space required to fully test the effects of this parallel scheduler-based vulnerability. These results and the methods developed here can help to include machine-level considerations into adversarial robustness assessments, which can make a difference in safety and mission critical applications.

Shanmugavelu, Sanjif [Maxeler Technologies, a Groq↗
DOE OSTI2025

A framework to evaluate machine learning crystal stability predictions

The rapid adoption of machine learning in various scientific domains calls for the development of best practices and community agreed-upon benchmarking tasks and metrics. We present Matbench Discovery as an example evaluation framework for machine learning energy models, here applied as pre-filters to first-principles computed data in a high-throughput search for stable inorganic crystals. We address the disconnect between (1) thermodynamic stability and formation energy and (2) retrospective and prospective benchmarking for materials discovery. Alongside this paper, we publish a Python package to aid with future model submissions and a growing online leaderboard with adaptive user-defined weighting of various performance metrics allowing researchers to prioritize the metrics they value most. To answer the question of which machine learning methodology performs best at materials discovery, our initial release includes random forests, graph neural networks, one-shot predictors, iterative Bayesian optimizers and universal interatomic potentials. We highlight a misalignment between commonly used regression metrics and more task-relevant classification metrics for materials discovery. Accurate regressors are susceptible to unexpectedly high false-positive rates if those accurate predictions lie close to the decision boundary at 0 eV per atom above the convex hull. The benchmark results demonstrate that universal interatomic potentials have advanced sufficiently to effectively and cheaply pre-screen thermodynamic stable hypothetical materials in future expansions of high-throughput materials databases.

Riebesell, Janosh↗
DOE OSTI2025

FuSED – Users Manual – (V.5.26)

The Fusion of Simulation, Experiment, and Data (FuSED) team provides a set of tools for solving inverse problems in structural dynamics (InverseSD) and thermal physics (InverseAria), a sensor placement optimization tool via Optimal Experimental Design (OED), and a decision boundary tool using SVMs (TRACE). These methods are used for designing experiments, model calibration, and verification/validation analysis of systems. This document provides a user’s guide.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Harnessing Machine Learning and Data Fusion for Accurate Undocumented Well Identification in Satellite Images

This study utilizes satellite data to detect undocumented oil and gas wells, which pose significant environmental concerns, including greenhouse gas emissions. Three key findings emerge from the study. Firstly, the problem of imbalanced data is addressed by recommending oversampling techniques like Rotation–GaussianBlur–Solarization data augmentation (RGS), the Synthetic Minority Over-Sampling Technique (SMOTE), or ADASYN (an extension of SMOTE) over undersampling techniques. The performance of borderline SMOTE is less effective than that of the rest of the oversampling techniques, as its performance relies heavily on the quality and distribution of data near the decision boundary. Secondly, incorporating pre-trained models trained on large-scale datasets enhances the models’ generalization ability, with models trained on one county’s dataset demonstrating high overall accuracy, recall, and F1 scores that can be extended to other areas. This transferability of models allows for wider application. Lastly, including persistent homology (PH) as an additional input improves performance for in-distribution testing but may affect the model’s generalization for out-of-distribution testing. A careful consideration of PH’s impact on overall performance and generalizability is recommended. Overall, this study provides a robust approach to identifying undocumented oil and gas wells, contributing to the acceleration of a net-zero economy and supporting environmental sustainability efforts.

SMOTE↗
DOE OSTI2026

SimLBR: Learning to Detect Fake Images by Learning to Detect Real Images

The rapid advancement of generative models has made the detection of AI-generated images a critical challenge for both research and society. Recent works have shown that most state-of-the-art fake image detection methods overfit to their training data and catastrophically fail when evaluated on curated hard test sets with strong distribution shifts. In this work, we argue that it is more principled to learn a tight decision boundary around the real image distribution and treat the fake category as a sink class. To this end, we propose SimLBR, a simple and efficient framework for fake image detection with Latent Blending Regularization (LBR). Our method significantly improves cross-generator generalization, achieving up to +24.85% accuracy and +69.62% recall on the challenging Chameleon benchmark. SimLBR is also highly efficient, training orders of magnitude faster than existing approaches. Furthermore, we emphasize the need for reliability-oriented evaluation in fake image detection, introducing risk-adjusted metrics and worst-case estimates to better assess model robustness. All the code and models are availabe at: https://github.com/mvrl/SimLBR

Dhakal, Aayush [Washington University, St. Louis]↗
DOE OSTI2026

Integrating science for water security governance

Hydrological extremes are intensifying globally, increasing the complexity of decisions required to ensure water security. Advances in hydrological science, modeling, and data systems have expanded the technical frontier of water research, yet uptake of scientific insights in policy and management decisions remains limited. This persistent science–policy gap is not primarily a failure of knowledge generation or robustness, but an institutional challenge shaped by how scientific and governance systems are organized, coordinated, and connected to support the effective use of scientific knowledge. These challenges are particularly pronounced in multi-level and transboundary water governance, where decisions span jurisdictions and require coordination across institutional and political boundaries. We synthesize research at the science–policy interface and evidence from water security initiatives to show how institutional arrangements, scientific tool development, and research practices enable or constrain the sustained use of scientific knowledge in water-security governance processes. Building on these insights, we develop ‘shared decision infrastructure’ as a framing to describe how scientific knowledge is embedded within the institutional, relational, and procedural arrangements that connect science to decision-making processes over time. We translate this framing into a practical intervention roadmap centered on institutional design, tool translation, sustained co-production, and outcome-oriented evaluation to support the integration of science into ongoing governance processes. By positioning science as shared decision infrastructure, the roadmap clarifies how researchers can design scientific efforts that support more coordinated, accountable, and adaptive water security decisions amid deepening uncertainty.

M whitney, Kristen [NASA Goddard Space Flight Cent↗
DOE OSTI2026

Using Visual Systems Mapping to Improve Transparency and Comparability of Life Cycle Assessment Baseline Scenarios

Visual systems mapping is a systems engineering approach used to represent complex processes and interactions. This study evaluates its application for documenting assumptions in life cycle assessment (LCA) baseline scenarios. In LCA, the baseline or reference case represents the business as usual system against which changes in impacts (e.g., emissions) are assessed. These baseline assumptions are particularly influential in biomass LCAs, yet they often vary across studies due to regional context, system boundaries, and simplifying assumptions that are not consistently or transparently documented. As a result, key feedbacks, omitted processes, and boundary choices may remain unclear, limiting comparability across studies and weakening their usefulness for decision-making. This study examines whether visual systems mapping can improve the transparency and comparability of biomass LCA baseline scenarios. A case study of five published biomass-related LCAs were reviewed, and their baseline scenarios were translated into visual system maps to identify included processes, omitted components, and underlying assumptions. The analysis demonstrates that visual systems mapping can make baseline assumptions more explicit, highlight excluded dynamics, and improve documentation of system boundaries. Based on these findings, the study recommends the use of visual systems mapping alongside open data repositories and reproducible workflows to support greater transparency, reproducibility, and comparability in LCAs. These improvements can strengthen the role of LCAs in informing decisions related to sustainable biomass systems.

Davis, Maggie [ORNL] (ORCID:0000000181319328)↗
DOE OSTI2026

ORCHID: Orchestrated Retrieval-Augmented Classification of High-Risk Property with Intelligent Decision-Making

High-Risk Property (HRP) classification is critical at U.S. Department of Energy (DOE) sites, where inventories include sensitive and often dual-use equipment. Compliance must track evolving rules designated by various export control policies to make transparent and auditable decisions. Traditional expert-only workflows are time-consuming, backlog-prone, and struggle to keep pace with shifting regulatory boundaries. We propose ORCHID, a modular agentic framework for HRP classification that pairs retrieval-augmented generation (RAG) with human oversight to produce policy based outputs that can be audited. Small cooperating agents—retrieval, description refiner, classifier, validator, and feedback logger—coordinate via agent-to-agent messaging and invoke tools through the Model Context Protocol (MCP) for model-agnostic on-premise operation. The interface follows an "Item to Evidence to Decision" loop with step-by-step reasoning, on-policy citations, and append-only audit bundles (run-cards, prompts, evidence). In preliminary tests on real HRP cases, ORCHID improves accuracy and traceability over a non-agentic baseline while deferring uncertain items to Subject Matter Experts (SMEs). The demonstration shows single item submission, grounded citations, SME feedback capture, and exportable audit artifacts—illustrating a practical path to trustworthy LLM assistance in sensitive DOE compliance workflows.

Das, Sanjay [ORNL] (ORCID:0009000542591915)↗
DOE OSTI2026

Influence of the as-built microstructure on the recrystallization of an additively manufactured Inconel939 Ni-based superalloy

This study investigates the influence of the as-built microstructure on the recrystallization (RX) behavior and mechanical properties of the Ni-based superalloy Inconel 939 produced by laser powder bed fusion (PBF-LB/M). Two distinct as-built microstructures were obtained by varying the hatch distance (h d ): a columnar, strongly textured condition (h d =50, termed h d 50) and an equiaxed, weakly textured condition (h d =70, termed h d 70)). Both were subjected to nine solution treatments combining three temperatures (1100, 1150, and 1200 °C) and three holding times (1, 4, and 8 h). Comprehensive microstructural characterization was conducted to assess grain morphology, texture, grain boundary character, dislocation density, and precipitate distribution. Recrystallization was found to be significantly slower than in cast counterparts, requiring higher temperatures and longer times for completion. The initial microstructure plays a decisive role: full RX was achieved only in hd70 specimens after treatment at 1200 °C for 8 h, whereas hd50 samples exhibited delayed and incomplete RX under identical conditions. This behavior is attributed to the finer grain size and higher fraction of high-angle grain boundaries in hd70, which promote recrystallization. Mechanical testing revealed that hd70 samples subjected to a 1200 °C/8 h treatment followed by standard double ageing show higher yield and tensile strengths across the investigated temperature range than both printed and cast Inconel939 processed under conventional conditions, albeit with slightly reduced ductility. The enhanced mechanical performance is attributed to the larger grain size, which limits grain boundary sliding. These results demonstrate the critical importance of controlling the as-built microstructure and tailoring post-processing strategies to optimize high-temperature performance of PBF-LB/M Inconel939.

Inconel939↗
DOE OSTI2026

Quantifying atmospheric carbon removal at pulp and paper mills: a life cycle assessment across system boundaries

The pulp and paper industry is a promising yet underexplored platform for large-scale carbon dioxide removal (CDR) due to its use of biogenic feedstocks and production of concentrated CO 2 emissions from point sources. This study presents the first comprehensive life cycle assessment (LCA) of retrofitting an amine-based carbon capture and storage (CCS) system into a representative virgin kraft pulp and paper mill in the Southeastern U.S. We evaluate carbon removal across five system configurations, applying both static and dynamic LCA methods under multiple functional units: CO 2 captured, biomass input, and paper output. Results show that CCS retrofits can convert a conventional mill from a net emitter into a net carbon sink, with total removal efficiencies from 17% to 92% (metric tonnes of CO 2 removed per metric tonne of CO 2 available for removal under selected boundary conditions). When carbon removal is normalized to the quantity of biogenic CO 2 captured—a narrow, gate-to-gate system boundary that considers only CCS facility emissions—removal efficiencies reached as high as 92%. The use of such narrow boundaries aligns with precedents in traditional LCA methodology, where gate-to-gate assessments are commonly applied to isolate process-level performance and allocate emissions accordingly, providing a consistent basis for comparison across technologies. Under broader cradle-to-grave boundaries—which begin tracking carbon at the point of its physical removal from the atmosphere via photosynthesis in the forest, and extend to include upstream forest operations, mill-wide emissions, and downstream product decomposition—efficiencies declined, ranging from 17% to 46% under static assumptions and dropping to 12% when accounting for dynamic biogenic carbon fluxes over time. These results underscore how system boundary definitions influence reported outcomes, while also illustrating the complementary roles of narrow and broad perspectives for different decision-making contexts.

09 BIOMASS FUELS↗
DOE OSTI2025

Tactical Analysis for Calculating Contextual Risk at Boundaries: Summary of Laboratory Directed Research & Development Effort

The Tactical Analysis for Calculating Contextual Risk at Boundaries (TACCRAB) tool is an innovative digital twin (DT) platform and automated risk algorithm designed to transform operational decision-making in structured screening environments, with an initial focus on Southern Border Land Ports of Entry (POEs). The invention provides integration points for advanced artificial intelligence, predictive modeling, and real-time data analysis to produce a comprehensive risk management tool that enables proactive, data-informed security strategies. The core inventive features of TACCRAB center on its unique risk algorithm, which dynamically calculates contextual risk by synthesizing historical data, near real-time streaming data from the checkpoints themselves, and AI-generated predictions. Unlike traditional risk assessment methods, TACCRAB utilizes a DT to provide comprehensive operational insights, allowing stakeholders to visualize, simulate, and optimize checkpoint configurations with unprecedented speed and contextual awareness. TACCRAB's key innovation lies in its ability to combine multiple complex inputs - including technology detection probabilities, resource availability, screening pathway characteristics, and threat actor behavioral patterns - into a unified risk calculation and update these inputs based on changing operational and environmental conditions. By leveraging a DT that continuously updates and learns from linked data, TACCRAB can suggest adaptive mitigation strategies that minimize risk while maintaining operational efficiency. Particularly novel is the platform's approach to decision support, which goes beyond static risk assessment. The DT provides dynamic metrics such as wait times, resource allocation effectiveness, and potential emerging threat scenarios, enabling users to view sophisticated, relevant what-if simulations and optimize checkpoint operations in near real-time. The system's architecture allows for generalized application across different screening environments, such as secure facilities, ports of entry, and soft targets, making it a versatile tool for security and operational management. The invention distinguishes itself through its comprehensive integration of predictive modeling, AI-driven pattern discovery, and user-friendly interface design. By combining these elements, TACCRAB transforms complex risk data into actionable insights, supporting decision-makers at various organizational levels - from booth agents making split-second screening decisions to checkpoint managers optimizing the day's resource allocation to strategic planners managing long-term investments.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗
DOE OSTI2024

Exploring the role of judgement and shared situation awareness when working with AI recommender systems

Abstract AI-advised Decision Making is a form of human-autonomy teaming in which an AI recommender system suggests a solution to a human operator, who is responsible for the final decision. This work seeks to examine the importance of judgement and shared situation awareness between humans and automated agents when interacting together in the form of a recommender systems. We propose manipulating both human judgement and shared situation awareness by providing the human decision maker with relevant information that the automated agent (AI), in the form of a recommender system, uses to generate possible courses of action. This paper presents the results of a two-phase between-subjects study in which participants and a recommender system jointly make a high-stakes decision. We varied the amount of relevant information the participant had, the assessment technique of the proposed solution, and the reliability of the recommender system. Findings indicate that this technique of supporting the human’s judgement and establishing a shared situation awareness is effective in (1) boosting the human decision maker’s situation awareness and task performance, (2) calibrating their trust in AI teammates, and (3) reducing overreliance on an AI partner. Additionally, participants were able to pinpoint the limitations and boundaries of the AI partner’s capabilities. They were able to discern situations where the AI’s recommendations could be trusted versus instances when they should not rely on the AI’s advice. This work proposes and validates a way to provide model-agnostic transparency into recommender systems that can support the human decision maker and lead to improved team performance.

Srivastava, Divya↗
DOE OSTI2024

From Text to Maps: LLM-Driven Extraction and Geotagging of Epidemiological Data

Epidemiological datasets are essential for public health analysis and decision-making, yet they remain scarce and often difficult to compile due to inconsistent data formats, language barriers, and evolving political boundaries. Traditional methods of creating such datasets involve extensive manual effort and are prone to errors in accurate location extraction. To address these challenges, we propose utilizing large language models (LLMs) to automate the extraction and geotagging of epidemiological data from textual documents. Our approach significantly reduces the manual effort required, limiting human intervention to validating a subset of records against text snippets and verifying the geotagging reasoning, as opposed to reviewing multiple entire documents manually to extract, clean, and geotag. Additionally, the LLMs identify information often overlooked by human annotators, further enhancing the dataset’s completeness. Our findings demonstrate that LLMs can be effectively used to semi-automate the extraction and geotagging of epidemiological data, offering several key advantages: (1) comprehensive information extraction with minimal risk of missing critical details; (2) minimal human intervention; (3) higher-resolution data with more precise geotagging; and (4) significantly reduced resource demands compared to traditional methods.

Harrod, Karly↗
DOE OSTI2024

Long duration battery sizing, siting, and operation under wildfire risk using progressive hedging

Battery sizing and siting problems are computationally challenging due to the need to make long-term planning decisions that are cognizant of short-term operational decisions. This paper considers sizing, siting, and operating batteries in a power grid to maximize their benefits, including price arbitrage and load shed mitigation, during both normal operations and periods with high wildfire ignition risk. Here we formulate a multi-scenario optimization problem for long duration battery storage while considering the possibility of load shedding during Public Safety Power Shutoff (PSPS) events that de-energize lines to mitigate severe wildfire ignition risk. To enable a computationally scalable solution of this problem with many scenarios of wildfire risk and power injection variability, we develop a customized temporal decomposition method based on a progressive hedging framework. Extending traditional progressive hedging techniques, we consider coupling in both placement variables across all scenarios and state-of-charge variables at temporal boundaries. This enforces consistency across scenarios while enabling parallel computations despite both spatial and temporal coupling. The proposed decomposition facilitates efficient and scalable modeling of a full year of hourly operational decisions to inform the sizing and siting of batteries. With this decomposition, we model a year of hourly operational decisions to inform optimal battery placement for a 240-bus WECC model in under 70 min of wall-clock time.

25 ENERGY STORAGE↗