Results for “Probabilistic models” — Page 20

DOE OSTI2020

A Dose Assessment Model for Radioactive Waste Exposed by Gully Erosion at West Valley - 20513

The Western New York Nuclear Service Center (WNYNSC), located approximately 48 km south of Buffalo, New York, is the site of a former nuclear fuel reprocessing and radioactive waste disposal facility. Spent nuclear fuel was processed there from 1966 to 1972, leaving behind radioactive and chemical wastes in two disposal areas and a waste tank farm. Site operations also resulted in releases of radioactivity to site soils, groundwater, and to surface waters draining the site. The New York State Energy Research and Development Authority (NYSERDA) and the U.S. Department of Energy (DOE) are collaborating in a process of decision making for decommissioning those facilities remaining at the WNYNSC following the completion of Phase 1 decommissioning. Neptune and Company, Inc. (Neptune) was contracted to develop a probabilistic performance assessment (PPA) computer model to assist the agencies in this process. The PPA Model includes a contaminant transport component focusing on the movement of contaminants within and among environmental media including groundwater and surface water transport, contaminant translocation by plants and animals, diffusion, and erosion. The model also includes evaluation of potential exposure and health effects for a Resident Farmer exposure scenario, where the Resident Farmer represents a critical group, described as that group of individuals reasonably expected to receive the greatest exposure to residual radioactivity for any applicable set of circumstances. The West Valley Site is located in the glaciated Allegheny Plateau region of western New York State. The waste reprocessing and disposal areas were constructed on a relatively fat area of plateau dissected by drainages of Buttermilk Creek, including Erdman Brook, Franks Creek, and Quarry Creek. An important aspect of the contaminant transport component of the PPA model is consideration of erosive processes such as slumping of the stream slopes and the advance of gullies from these streams. These erosion processes remove material from the plateau, growing the size of the creek valleys and making them wider and deeper. Of particular interest for the impact of erosion is radioactive waste contained in the Nuclear Regulatory Commission (NRC)-Licensed Disposal Area (NDA), the New York State-Licensed Disposal Area (SDA), and residual radiological inventory in the underground storage tanks at the Waste Tank Farm (WTF). The PPA Model is organized around geographically-defined facilities which were constructed upon the plateau, including the NDA, SDA, and WTF. Ongoing stream erosion processes will potentially transfer radioactive waste and residual inventory from these facilities to the ground surface on adjacent hillslope areas where erosion has breached the facility. Hence, it is important to evaluate the consequences of potential exposures to a Resident Farmer on the hillslopes below a breached facility. Two interrelated aspects of the dose assessment model related to hillslope exposure are discussed: 1) representation of the physical processes related to transport of radionuclides from facilities onto the hillslopes, and from the hillslopes into adjoining creeks where contaminated material migrates downstream with surface water and sediment, and; 2) adaptation of the activities associated with the Resident Farmer scenario to assess potential exposures to contamination in the hillslope areas. This discussion will cover the conceptual basis of the hillslope exposure and transport models, and also implementation in the PPA computer model. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2022

Occlusion-Perturbed Deep Learning for Probabilistic Solar Forecasting via Sky Images

Solar forecasting is shifting to the probabilistic paradigm due to the inherent uncertainty within the solar resource. Input uncertainty quantification is one of the widely used and best-performing ways to model solar uncertainty. However, compared to other sources of inputs, such as numerical weather prediction models, pure sky image-based probabilistic solar forecasting lags behind. In this research, an occlusion-perturbed convolutional neural network, named the PSolarNet, is developed. The PSolarNet provides very short-term deterministic forecasts, forecast scenarios, and probabilistic forecasts of the global horizontal irradiance from sky image sequences. Case studies based on 6 years of open-source data show that the developed PSolarNet is able to generate accurate 10-minute ahead deterministic forecasts with a 5.62% normalized root mean square error, realistic and diverse forecast scenarios with a 0.966 average correlation with the actual time series, and reliable and sharp probabilistic forecasts with a 2.77% normalized continuous ranked probability score.

Bayesian model averaging↗

DOE OSTI2022

Occlusion-Perturbed Deep Learning for Probabilistic Solar Forecasting via Sky Images: Preprint

Solar forecasting is shifting to the probabilistic paradigm due to the inherent uncertainty within the solar resource. Input uncertainty quantification is one of the widely used and best-performing ways to model solar uncertainty. However, compared to other sources of inputs, such as numerical weather prediction models, pure sky image-based probabilistic solar forecasting lags behind. In this research, an occlusion-perturbed convolutional neural network, named the PSolarNet, is developed. The PSolarNet provides very short-term deterministic forecasts, forecast scenarios, and probabilistic forecasts of the global horizontal irradiance from sky image sequences. Case studies based on 6 years of open-source data show that the developed PSolarNet is able to generate accurate 10-minute ahead deterministic forecasts with a 5.62% normalized root mean square error, realistic and diverse forecast scenarios with a 0.966 average correlation with the actual time series, and reliable and sharp probabilistic forecasts with a 2.77% normalized continuous ranked probability score.

Bayesian model averaging↗

DOE OSTI2024

Light Water Reactor Sustainability Program: Use of Time Distributions to Predict Operator Procedure Performance in Dynamic Human Reliability Analysis

The Human Unimodel for Nuclear Technology to Enhance Reliability (HUNTER) framework affords software capable of conducting human reliability analysis (HRA) using a dynamic approach built around operating procedures (OPs) from nuclear power plants (NPPs). Previous HUNTER reports document the development of this software tool, the coupling of HUNTER to the simulator code, the collection of operator performance data by using simulators to calibrate HUNTER models, and linking HUNTER to probabilistic risk assessment (PRA) software. The present report largely addresses two topics. The first is a new function in HUNTER called the HUNTER Procedure Performance Predictor (P3). HUNTER P3 uses HUNTER’s built in Monte Carlo tools featuring human performance variability to identify potential error traps in procedures. The second topic is time distribution analysis to generate time inputs for dynamic HRA. The current analysis was performed to investigate time distributions for task primitives, which are the minimum task unit of analysis used in dynamic HRA modeling. Using the time distribution data, the elapsed time for human actions in an extended loss of AC power (ELAP) scenario is then investigated. Time data and prediction are essential for modeling procedure performance.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2020

The Influence of Aqueous Dispersivity on Transport of Radiological Material at LANL MDA G - 20450

Aqueous dispersivity is a phenomenon by which a solute moving in groundwater mixes with water that does not contain the solute and becomes diluted. Solutes take paths of varying lengths through pore networks or experience slower velocities due to interactions with pore walls. Aqueous dispersivity leads to breakthrough curves with earlier arrival than expected in purely advective systems and persistent 'tails,' or late-arriving, dilute concentrations of the solute. Dispersivity can be longitudinal (in the direction of flow) or transverse (orthogonal to flow). LANL has disposed of radioactive material at Material Disposal Area (MDA) G since the late 1950's. Most of the disposed materials at MDA G are low-level waste (LLW), but other types of radioactive wastes are disposed or stored at the site. For example, there is a collection of remote-handled transuranic waste (RH-TRU) waste in 33 shafts, which were intended to be stored retrievably. Aqueous-phase radionuclides included in models of contaminant transport at MDA G are subject to dispersive processes in porous media of the vadose zone and saturated zone. To model dispersivity at MDA G probabilistically, we created site-specific distributions of aqueous dispersivity. First, we performed a comprehensive review of peer-reviewed publications, reports from other performance assessments of MDA G, and documentation from sites that were considered similar based on expert opinion, such as Yucca Mountain. Data were included from a wide range of sources, so that distributions developed based on the data reflect the current state of knowledge in properties of the subsurface environment. Data were grouped based on value type (e.g., determined via experiment or used in a different model), rock type, zone type (saturated or unsaturated), and site, and were weighted in the distributions according to their relevance to the site and zone type. We combined Monte Carlo simulation and bootstrapping techniques to produce approximate distributions of mean aqueous dispersivity. Distributions of dispersivity for saturated and unsaturated zone transport are differentiated by data weight and the length scales over which transport occurs. We find that underestimating or disregarding transverse dispersivity may lead to overstating simulated well capture in the saturated zone at MDA G. Neglecting transverse dispersivity is a common assumption due in part to use of one-dimensional transport models that do not consider solute spreading orthogonal to flow. Also, we find that variability in longitudinal dispersivity is associated with substantial travel time alterations in the MDA G unsaturated zone and in simulated contaminant breakthrough times at a downstream observation plane in the MDA G saturated zone. Finally, at large scales (>1000 m), a linear trend between the dispersivity coefficient and length over which transport occurs may overestimate the dispersivity coefficient. Future modeling work with application to solute spreading could include interacting advective and diffusive zones in the highly fractured basalt below MDA G. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2024

An Approach to Dynamic Human Reliability Analysis and Its Data Collection Framework

Human reliability analysis (HRA) is a method for evaluating human errors in a variety of complex systems such as nuclear power plants, military systems, aircraft, and chemical plants. Most HRA methods currently used by regulatory institutes or utilities are called static HRA and are carried out by simple worksheets or simple calculators. To date, there are many unsolved or intrinsic challenges in static HRA. For example, existing static HRA does not realistically model and evaluate human actions as they would be performed at actual systems. There is no method with HRA to objectively estimate the time required for human actions despite being essential to HRA processes. In addition, many HRA methods still rely on a dataset generated prior to the 1980s, from unrelated industry experience or simply from expert judgment. Accordingly, this study attempted to research how to overcome the challenges of existing HRA via dynamic risk assessment (a.k.a., simulation-based or computation-based risk assessment) techniques. First, this study developed a dynamic HRA method, named as PRocedure-based Investigation Method of EMRALD Risk Assessment – HRA (PRIMERA-HRA). The PRIMERA-HRA mainly concentrates on providing HRA analysts with specific guidelines on how to reasonably model human actions, assign human reliability data and evaluate output of simulation within a dynamic probabilistic risk assessment tool, called as Event Modeling Risk Assessment using Linked Diagrams (EMRALD). Second, this study also developed a module for performance shaping factors (i.e., the key concept in HRA quantification) applicable to dynamic HRA, then implemented it based on PRIMERA-HRA within the EMRALD tool. Third, this study developed an HRA data collection framework to support dynamic HRA, called as Simplified Human Error Experimental Program (SHEEP). Originally, the SHEEP study aimed to support static HRA and its data collection, but recently extended the scope to the new technologies such as dynamic HRA or HRA for advanced reactors. SHEEP focuses on the use of data collected from simplified simulators to complement—but not replace—data collection studies using full-scope simulators and actual operators. To date, many experiments were conducted under the SHEEP framework. Multiple analyses, such as human performance analysis, human error analysis, task complexity analysis, learning effect analysis and time distribution analysis, were also carried out using the collected data. Then, based on the major insights, an approach to inferring full-scope data based on simplified simulator data was proposed. The PRIMERA-HRA and SHEEP research are expected to evaluate human actions more realistically than existing static HRA, provide an opportunity to collect more HRA data with reasonable cost and labor, then contribute to enhance the quality of HRA.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2024

An Approach to Dynamic Human Reliability Analysis and Its Data Collection Framework

Human reliability analysis (HRA) is a method for evaluating human errors in a variety of complex systems such as nuclear power plants, military systems, aircraft, and chemical plants. Most HRA methods currently used by regulatory institutes or utilities are called static HRA and are carried out by simple worksheets or simple calculators. To date, there are many unsolved or intrinsic challenges in static HRA. For example, existing static HRA does not realistically model and evaluate human actions as they would be performed at actual systems. There is no method with HRA to objectively estimate the time required for human actions despite being essential to HRA processes. In addition, many HRA methods still rely on a dataset generated prior to the 1980s, from unrelated industry experience or simply from expert judgment. Accordingly, this study attempted to research how to overcome the challenges of existing HRA via dynamic risk assessment (a.k.a., simulation-based or computation-based risk assessment) techniques. First, this study developed a dynamic HRA method, named as PRocedure-based Investigation Method of EMRALD Risk Assessment – HRA (PRIMERA-HRA). The PRIMERA-HRA mainly concentrates on providing HRA analysts with specific guidelines on how to reasonably model human actions, assign human reliability data and evaluate output of simulation within a dynamic probabilistic risk assessment tool, called as Event Modeling Risk Assessment using Linked Diagrams (EMRALD). Second, this study also developed a module for performance shaping factors (i.e., the key concept in HRA quantification) applicable to dynamic HRA, then implemented it based on PRIMERA-HRA within the EMRALD tool. Third, this study developed an HRA data collection framework to support dynamic HRA, called as Simplified Human Error Experimental Program (SHEEP). Originally, the SHEEP study aimed to support static HRA and its data collection, but recently extended the scope to the new technologies such as dynamic HRA or HRA for advanced reactors. SHEEP focuses on the use of data collected from simplified simulators to complement—but not replace—data collection studies using full-scope simulators and actual operators. To date, many experiments were conducted under the SHEEP framework. Multiple analyses, such as human performance analysis, human error analysis, task complexity analysis, learning effect analysis and time distribution analysis, were also carried out using the collected data. Then, based on the major insights, an approach to inferring full-scope data based on simplified simulator data was proposed. The PRIMERA-HRA and SHEEP research are expected to evaluate human actions more realistically than existing static HRA, provide an opportunity to collect more HRA data with reasonable cost and labor, then contribute to enhance the quality of HRA.

99 - GENERAL AND MISCELLANEOUS↗

DOE OSTI2024

Short-Term Probabilistic Solar Forecasting via Reinforcement Learning over ECMWF

In this paper, we present an innovative reinforcement learning approach for short-term solar forecasting, leveraging data from the European Centre for Medium-Range Weather Forecasts (ECMWF). The methodology begins with the application of the System Advisor Model (SAM) to transform various ECMWF numerical weather prediction members into predictive photovoltaic power generation. To enhance the precision of deterministic forecasting, we introduce a dynamic model selection algorithm based on Q-learning. This algorithm dynamically identifies and utilizes the most accurate ensemble member for forecasting purposes. Furthermore, we employ a support vector regression surrogate model with a Gaussian distribution to generate probabilistic forecasts, providing a holistic view of solar energy generation uncertainty. To expedite the training process and make it more practical for real-world applications, we integrate a rolling update workflow. This innovative workflow reduces the training period from months to a mere 19 days, making our method highly efficient. Numerical results of the case study show that in comparison to benchmark models, the proposed method improves the deterministic and probabilistic solar forecasting accuracy by up to 40.84% and 48.42%, respectively.

ensemble forecasting↗

DOE OSTI2023

Efficient Probabilistic Prediction and Uncertainty Quantification of Tropical Cyclone–Driven Storm Tides and Inundation

Abstract This study proposes and assesses a methodology to obtain high-quality probabilistic predictions and uncertainty information of near-landfall tropical cyclone–driven (TC-driven) storm tide and inundation with limited time and resources. Forecasts of TC track, intensity, and size are perturbed according to quasi-random Korobov sequences of historical forecast errors with assumed Gaussian and uniform statistical distributions. These perturbations are run in an ensemble of hydrodynamic storm tide model simulations. The resulting set of maximum water surface elevations are dimensionality reduced using Karhunen–Loève expansions and then used as a training set to develop a polynomial chaos (PC) surrogate model from which global sensitivities and probabilistic predictions can be extracted. The maximum water surface elevation is extrapolated over dry points incorporating energy head loss with distance to properly train the surrogate for predicting inundation. We find that the surrogate constructed with third-order PCs using elastic net penalized regression with leave-one-out cross validation provides the most robust fit across training and test sets. Probabilistic predictions of maximum water surface elevation and inundation area by the surrogate model at 48-h lead time for three past U.S. landfalling hurricanes (Irma in 2017, Florence in 2018, and Laura in 2020) are found to be reliable when compared to best track hindcast simulation results, even when trained with as few as 19 samples. The maximum water surface elevation is most sensitive to perpendicular track-offset errors for all three storms. Laura is also highly sensitive to storm size and has the least reliable prediction. Significance Statement The purpose of this study is to develop and evaluate a methodology that can be used to provide high-quality probabilistic predictions of hurricane-induced storm tide and inundation with limited time and resources. This is important for emergency management purposes during or after the landfall of hurricanes. Our results show that sampling forecast errors using quasi-random sequences combined with machine learning techniques that fit polynomial functions to the data are well suited to this task. The polynomial functions also have the benefit of producing exact sensitivity indices of storm tide and inundation to the forecasted hurricane properties such as path, intensity, and size, which can be used for uncertainty estimation. The code implementing the presented methodology is publicly available on GitHub.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Modeling airborne pathogen transport and transmission risks of SARS-CoV-2

An integrated modeling approach has been developed to better understand the relative impacts of different expiratory and environmental factors on airborne pathogen transport and transmission, motivated by the recent COVID-19 pandemic. Computational fluid dynamics (CFD) modeling was used to simulate spatial-temporal aerosol concentrations and quantified risks of exposure as a function of separation distance, exposure duration, environmental conditions (e.g., airflow/ventilation), and face coverings. The CFD results were combined with infectivity models to determine probability of infection, which is a function of the spatial-temporal aerosol concentrations, viral load, infectivity rate, viral viability, lung-deposition probability, and inhalation rate. Uncertainty distributions were determined for these parameters from the literature. Probabilistic analyses were performed to determine cumulative distributions of infection probabilities and to determine the most important parameters impacting transmission. This modeling approach has relevance to both pathogen and pollutant dispersion from expelled aerosol plumes.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2025

Probabilistic Deliverability Assessment of Distributed Energy Resources via Scenario-Based AC Optimal Power Flow

As electric grids decarbonize and distributed energy resources (DERs) become increasingly prevalent, interconnection assessments must evolve to reflect operational variability and control flexibility. This paper highlights key modeling limitations observed in practice and reviews approaches for modeling uncertainty. It then introduces a Probabilistic Deliverability Assessment (PDA) framework designed to complement and extend existing procedures. The framework integrates scenario-based AC optimal power flow (AC OPF), corrective dispatch, and optional multi-temporal constraints. Together, these form a structured methodology for quantifying DER utilization, deliverability, and reliability under uncertainty in load, generation, and topology. Outputs include interpretable metrics with confidence intervals that inform siting decisions and evaluate compliance with reliability thresholds across sampled operating conditions. A case study on Puerto Rico’s publicly available bulk power system model demonstrates the framework’s application using minimal input data, consistent with current interconnection practice. Across staged fossil generation retirements, the PDA identifies high-value DER sites and regions requiring additional reactive power support. Results are presented through mean dispatch signals, reliability metrics, and geospatial visualizations, demonstrating how the framework provides transparent, data-driven siting recommendations. The framework’s modular design supports incremental adoption within existing workflows, encouraging broader use of AC OPF in interconnection and planning contexts.

14 SOLAR ENERGY↗

DOE OSTI2022

Assessment of doses in contaminated urban areas: modelling exercise based on Fukushima data

State-of-the-art dose assessment models were applied to estimate doses to the population in urban areas contaminated by the Fukushima Daiichi Nuclear Power Plant accident. Assessment results were compared among five models, and comparisons of model predictions with actual measurements were also made. Assessments were performed using both probabilistic and deterministic approaches. Predicted dose distributions for indoor and outdoor workers from a probabilistic approach were in good agreement with the actual measurements. In addition, when the models were applied to assess the doses to the representative person, based on a concept recommended by the International Commission on Radiological Protection and in the International Atomic Energy Agency Safety Standards, it was evident that doses to the representative person obtained with a deterministic approach were always higher than those obtained with a probabilistic approach using the same model.

dose assessment↗

DOE OSTI2024

Building Intelligence with Layered Defense Using Security-Constrained Optimization and Security Risk Detection (BUILD-SOS): A Probabilistic Approach

In this project, we employ a layered protection strategy incorporating advanced optimization and detection techniques using a probabilistic approach. The probabilistic approach is not only applied when detecting cyber attacks, but also incorporated in control strategies, which greatly increases the attacking difficulties. Hackers need to understand both probabilistic detection algorithms and uncertainty modeling methods in control in order to execute any effective attacks. The end-to-end solutions enable us to provide Building Intelligence with Layered Defense using Security-Constrained Optimization and Security Risk Detection (BUILD-SOS).

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2022

Adversarial autoencoder ensemble for fast and probabilistic reconstructions of few-shot photon correlation functions for solid-state quantum emitters

Second-order photon correlation measurements [g (2) (τ) functions] are widely used to classify single-photon emission purity in quantum emitters or to measure the multiexciton quantum yield of emitters that can simultaneously host multiple excitations – such as quantum dots – by evaluating the value of g (2) (τ = 0). Accumulating enough photons to accurately calculate this value is time consuming and could be accelerated by fitting of few-shot photon correlations. Here, we develop an uncertainty-aware, deep adversarial autoencoder ensemble (AAE) that reconstructs noise-free g (2) (τ) functions from noise-dominated, few-shot inputs. The model is trained with simulated g (2) (τ) functions that are facilely generated by Poisson sampling time bins. The AAE reconstructions are performed orders-of-magnitude faster, with reconstruction errors and estimates of g (2) (τ = 0) that are lower in variance and similar in accuracy compared to Maximum likelihood estimation and Levenberg-Marquardt least-squares fitting approaches, for simulated and experimentally measured few-shot g (2) (τ) functions (~100 two-photon events) of InP/ZnS/ZnSe and CdS/CdSe/CdS quantum dots. The deep-ensemble model comprises eight individual autoencoders, allowing for probabilistic reconstructions of noise-free g (2) (τ) functions, and we show that the predicted variance scales inversely with number of shots, with comparable uncertainties to computationally intensive Markov chain Monte Carlo sampling. Furthermore, this work demonstrates the advantage of machine learning models to perform uncertainty-aware, fast, and accurate reconstructions of simple Poisson-distributed photon correlation functions, allowing for on-the-fly reconstructions and accelerated materials characterization of solid-state quantum emitters.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2022

GDSA Framework Development and Process Model Integration FY2022

The Spent Fuel and Waste Science and Technology (SFWST) Campaign of the U.S. Department of Energy (DOE) Office of Nuclear Energy (NE), Office of Spent Fuel & Waste Disposition (SFWD) is conducting research and development (R&D) on geologic disposal of spent nuclear fuel (SNF) and high-level nuclear waste (HLW). A high priority for SFWST disposal R&D is disposal system modeling (Sassani et al. 2021). The SFWST Geologic Disposal Safety Assessment (GDSA) work package is charged with developing a disposal system modeling and analysis capability for evaluating generic disposal system performance for nuclear waste in geologic media. This report describes fiscal year (FY) 2022 advances of the Geologic Disposal Safety Assessment (GDSA) performance assessment (PA) development groups of the SFWST Campaign. The common mission of these groups is to develop a geologic disposal system modeling capability for nuclear waste that can be used to assess probabilistically the performance of generic disposal options and generic sites. The modeling capability under development is called GDSA Framework (pa.sandia.gov). GDSA Framework is a coordinated set of codes and databases designed for probabilistically simulating the release and transport of disposed radionuclides from a repository to the biosphere for post-closure performance assessment. Primary components of GDSA Framework include PFLOTRAN to simulate the major features, events, and processes (FEPs) over time, Dakota to propagate uncertainty and analyze sensitivities, meshing codes to define the domain, and various other software for rendering properties, processing data, and visualizing results.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2024

Evidential Deep Learning: Enhancing Predictive Uncertainty Estimation for Earth System Science Applications

Abstract Robust quantification of predictive uncertainty is a critical addition needed for machine learning applied to weather and climate problems to improve the understanding of what is driving prediction sensitivity. Ensembles of machine learning models provide predictive uncertainty estimates in a conceptually simple way but require multiple models for training and prediction, increasing computational cost and latency. Parametric deep learning can estimate uncertainty with one model by predicting the parameters of a probability distribution but does not account for epistemic uncertainty. Evidential deep learning, a technique that extends parametric deep learning to higher-order distributions, can account for both aleatoric and epistemic uncertainties with one model. This study compares the uncertainty derived from evidential neural networks to that obtained from ensembles. Through applications of the classification of winter precipitation type and regression of surface-layer fluxes, we show evidential deep learning models attaining predictive accuracy rivaling standard methods while robustly quantifying both sources of uncertainty. We evaluate the uncertainty in terms of how well the predictions are calibrated and how well the uncertainty correlates with prediction error. Analyses of uncertainty in the context of the inputs reveal sensitivities to underlying meteorological processes, facilitating interpretation of the models. The conceptual simplicity, interpretability, and computational efficiency of evidential neural networks make them highly extensible, offering a promising approach for reliable and practical uncertainty quantification in Earth system science modeling. To encourage broader adoption of evidential deep learning, we have developed a new Python package, Machine Integration and Learning for Earth Systems (MILES) group Generalized Uncertainty for Earth System Science (GUESS) (MILES-GUESS) ( https://github.com/ai2es/miles-guess ), that enables users to train and evaluate both evidential and ensemble deep learning. Significance Statement This study demonstrates a new technique, evidential deep learning, for robust and computationally efficient uncertainty quantification in modeling the Earth system. The method integrates probabilistic principles into deep neural networks, enabling the estimation of both aleatoric uncertainty from noisy data and epistemic uncertainty from model limitations using a single model. Our analyses reveal how decomposing these uncertainties provides valuable insights into reliability, accuracy, and model shortcomings. We show that the approach can rival standard methods in classification and regression tasks within atmospheric science while offering practical advantages such as computational efficiency. With further advances, evidential networks have the potential to enhance risk assessment and decision-making across meteorology by improving uncertainty quantification, a longstanding challenge. This work establishes a strong foundation and motivation for the broader adoption of evidential learning, where properly quantifying uncertainties is critical yet lacking.

Schreck, John S.↗

DOE OSTI2026

Modeling and Calibration of Supplier Selection Problem in Freight Agent-Based Simulations

Freight transportation modeling often struggles with data limitations, especially in accurately representing complex supplier selection processes and their impact on network flows. This research addresses this critical gap by developing a large-scale, calibrated agent-based model for supplier selection, complemented by a probabilistic heuristic for international shipments. Our approach integrates trade relationships between industry sectors, transportation costs, and a supplier-rating model adapted from existing literature. The model’s core objective is to minimize the discrepancy between modeled and observed commodity flows while ensuring a close match to regional shipping distance distributions. Implemented and tested across four major U.S. metropolitan areas—Atlanta, Chicago, Dallas–Fort Worth, and Los Angeles—the model demonstrates high fidelity in replicating observed freight patterns. Key findings reveal consistent alignment with national shipping distance trends and highlight significant spatial variations in commodity trade assignments and demand across the study regions. This behaviorally informed and transport-sensitive framework is designed to approximate real-world decision making, providing a robust tool for policymakers and planners to evaluate targeted interventions, assess infrastructure investments, and enhance supply chain resilience in the face of disruptions.

Ismael, Abdelrahman (ORCID:0000000303712110)↗

DOE OSTI2022

Insights into Cation Ordering of Double Perovskite Oxides from Machine Learning and Causal Relations

This work investigates origins of cation ordering in double perovskites using first-principles theory computations combined with machine learning (ML) and causal relations. We have considered various oxidation states of A, A', B, and B' from the family of transition metal ions to construct a diverse compositional space. A conventional framework employing traditional ML classification algorithms such as Random Forest (RF) coupled with appropriate features including geometry-driven and key structural modes leads to accurate prediction (~98%) of A-site cation ordering. We have evaluated the accuracy of ML models by employing analyses of decision paths, assignments of probabilistic confidence bound, and finally a direct non-Gaussian acyclic structural equation model to investigate causality. Our study suggests that structural modes are crucial for classifying layered, columnar, and rock-salt ordering. The charge difference between A and A' is the most important feature for predicting clear layered ordering, which in turn depends on the B and B' charge separation. We have also designed mathematical relationships with these features to derive energy differences to form clear layered ordering. Here, the trilinear coupling between tilt, in-phase rotation, and A-site antiferroelectric displacement in the Landau free-energy expansion becomes the necessary condition behind formation of A-site cation ordering.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗