Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “evaluation datasets”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Understanding the compound flood risk along the coast of the contiguous United States

Abstract. Compound flooding is a type of flood event caused by multiple flood drivers. The associated risk has usually been assessed using statistics-based analyses or hydrodynamics-based numerical models. This study proposes a compound flood (CF) risk assessment (CFRA) framework for coastal regions in the contiguous United States (CONUS). In this framework, a large-scale river model is coupled with a global ocean reanalysis dataset to (a) evaluate the CF exposure related to the coastal backwater effects on river basins, and (b) generate spatially distributed data for analyzing the CF hazard using a bivariate statistical model of river discharge and storm surge. The two kinds of risk are also combined to achieve a holistic understanding of the continental-scale CF risk. The estimated CF risk shows remarkable inter- and intra-basin variabilities along the CONUS coast with more variabilities in the CF hazard over the US west and Gulf coastal basins. Different risk assessment methods present significantly different patterns in a few key regions such as the San Francisco Bay area, the lower Mississippi River, and Puget Sound. Our results highlight the need to weigh different CF risk measures and avoid using single statistics-based or hydrodynamics-based CFRAs. Uncertainty sources in these CFRAs include the use of gauge observations, which cannot account for the flow physics or resolve the spatial variability of risks, and underestimations of the flood extremes and the dependence of CF drivers in large-scale models, highlighting the importance of understanding the CF risks for developing a more robust CFRA.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Athena: High-Performance Sparse Tensor Contraction Sequence on Heterogeneous Memory

Sparse tensor contraction sequence has been widely employed in many fields, such as chemistry and physics. However, how to efficiently implement the sequence faces multiple challenges, such as redundant computations and memory operations, massive memory consumption, and inefficient utilization of hardware. To address the above challenges, we introduce Athena, a high-performance framework for SpTC sequences. Athena introduces new data structures, leverages emerging Optane-based heterogeneous memory (HM) architecture, and stage parallelism. In particular, Athena introduces shared hash table-represented sparse accumulator to eliminate unnecessary input processing and data migration; Athena uses a novel data-semantic guided dynamic migration solution to make the best use of the Optane-based HM for high performance; Athena also co-runs execution phases with different characteristics to enable high hardware utilization. Evaluating with 12 datasets, we show that Athena brings 327-7362× speedup over the state-of-the-art SpTC algorithm. With the dynamic data placement guided by data semantics, Athena brings performance improvement on Optane-based HM over a state-of-the-art software-based data management solution, a hardware-based data management solution, and PMM-only by 1.58×, 1.82×, and 2.34× respectively.

Liu, Jiawen↗
DOE OSTI2025

Burst pressure models and validations for thick-walled pipelines containing corrosion defects

Corrosion is one major threat to pipeline integrity. Over the past decades, many corrosion models have been developed for determining the remaining strength of corroded pipelines, including ASME B31.G, Modified B31.G, LPC, PCORRC and their modified models. All these corrosion models are applicable only to large diameter, thin-walled pipelines with a diameter to wall thickness ratio D/t ≥ 20. In practice, many pipelines have a small diameter and thick wall with a D/t ratio < 20, and thus an adequate corrosion model is needed for assessing remaining strength for corroded thick-walled pipelines. This paper briefly reviews the theoretical burst pressure models for defect-free thin and thick-walled pipelines and four representative corrosion assessment models for thin-walled corroded pipelines. On this basis, two modified corrosion models are proposed to thick-walled pipelines in terms of the average shear stress yield theory. To verify the proposed corrosion models, comprehensive validations are performed. Numerical validations include the elastic-plastic finite element analysis to determine burst pressure for pipelines without and with corrosion defects and the model evaluation using a large dataset of available FEA results of burst pressure for machined defects. Experimental validations include a set of burst pressure tests for defect-free thick-walled pipes with different thicknesses and the model evaluation using one large burst dataset for machined defects with flat bottoms and another large dataset for real corrosion defects with curved river bottom profiles. Both numerical and experimental validations show that the proposed corrosion models can more accurately predict the remaining strength for corroded thin and thick-walled pipelines.

Pipeline↗
DOE OSTI2021

A review of imputation strategies for isobaric labeling-based shotgun proteomics

The throughput efficiency and increased depth of coverage provided by isobaric-labeled proteomics measurements have led to increased usage of these techniques. However, the structure of missing data is uniquely different than unlabeled studies. In this review, we compare the efficacy of nine imputation methods on a CPTAC proteomics iTRAQ dataset. Imputation methods were evaluated with regard to accuracy, variability, statistical hypothesis test inference and run time over datasets consisting of varying number of iTRAQ plexes and percentages of missing data. In general, expectation maximization and random forest imputation methods yielded the best performances, and constant-based methods performed poorly consistently across all dataset sizes and percentages of missing values. For datasets with small sample sizes and higher percentages of missing data, results indicate that statistical inference with no imputation may be preferable. Based on the findings in this review, there are core imputation methods that perform higher for isobaric-labeled proteomics data, but great care and consideration as to whether imputation should be used should be given for datasets comprised of a small number of samples, as well as to factors such as computational time and reproducibility of imputation values.

Bramer, Lisa M.↗
DOE OSTI2023

A Novel Spatial-Temporal Variational Quantum Circuit to Enable Deep Learning on NISQ Devices

Quantum computing presents a promising approach for machine learning with its capability for extremely parallel computation in high-dimension through superposition and entanglement. Despite its potential, existing quantum learning algorithms, such as Variational Quantum Circuits (VQCs), face challenges in handling more complex datasets, particularly those that are not linearly separable. What’s more, it encounters the deployability issue, making the learning models suffer a drastic accuracy drop after deploying them to the actual quantum devices. To overcome these limitations, this paper proposes a novel spatial-temporal design, namely “ST-VQC”, to integrate nonlinearity in quantum learning and improve the robustness of the learning model to noise. Specifically, ST-VQC can extract spatial features via a novel block-based encoding quantum sub-circuit coupled with a layer-wise computation quantum sub-circuit to enable temporal-wise deep learning. Additionally, a SWAP-Free physical circuit design is devised to improve robustness. These designs bring a number of hyperparameters. After a systematic analysis of the design space for each design component, an automated optimization framework is proposed to generate the ST-VQC quantum circuit. The proposed ST-VQC has been evaluated on two IBM quantum processors, ibm-cairo with 27 qubits and ibmq-lima with 7 qubits to assess its effectiveness. The results of the evaluation on the standard dataset for binary classification show that ST-VQC can achieve over 30% accuracy improvement compared with existing VQCs on actual quantum computers. Moreover, on a non-linear synthetic dataset, the STVQC outperforms a linear classifier by 27.9%, while the linear classifier using classical computing outperforms the existing VQC by 15.58%.

Li, Jinyang↗
DOE OSTI2021

A Hybrid System for Learning Classical Data in Quantum States

Deep neural network-powered artificial intelligence has rapidly changed our daily life with various applications. However, as one of the essential steps of deep neural networks, training a heavily-weighted network requires a tremendous amount of computing resources. Especially in the post Moore’s Law era, the limit of semiconductor fabrication technology has restricted the development of learning algorithms to cope with the increasing high intensity training data. Meanwhile, quantum computing has exhibited its significant potential in terms of speeding up the traditionally compute-intensive workloads. For example, Google illustrates quantum supremacy by completing a sampling calculation task in 200 seconds, which is otherwise impracticable on the world’s largest supercomputers. To this end, quantum-based learning becomes an area of interest, with the promising of a quantum speedup. In this paper, we propose GenQu, a hybrid and general-purpose quantum framework for learning classical data through quantum states. We evaluate GenQu with real datasets and conduct experiments on both simulations and real quantum computer IBM-Q. Our evaluation demonstrates that, comparing with classical solutions, the proposed models running on GenQu framework achieve similar accuracy with a much smaller number of qubits, while significantly reducing the parameter size by up to 95.86% and converging speedup by 66.67% faster.

Stein, Samuel A.↗
DOE OSTI2022

Evaluated Nuclear Structure Data File (ENSDF)

Contains recommended nuclear structure and decay data for all the known nuclides, which are obtained following a critical review of all available experimental data, supplemented with systematic trend studies and theoretical models. ENSDF data include (i) nuclear level properties, such as observation source, energy, half-life, decay modes, spin and parity; (ii) gamma ray energies, intensities, multipolarities, mixing ratios and conversion coefficients; (iii) nuclear radiation energy and intensity as well as radiation-specific data for different radiation types, such as such as gammas, alphas, betas and neutrons. Data for about 220 nuclides are fully evaluated each year; additionally, many datasets are partially updated each year. Nearly all the evaluation work is supported by the US Nuclear Data Program. The National Nuclear Data Center at Brookhaven National Laboratory is responsible for coordinating the evaluation effort as well as its web and journal dissemination.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2026

Particle Filter Based Inference Testing

The primary intent of PAR-FIT (Particle Filter based Inference Testing) is to provide hard inductive evidence that a machine learning model is capable and proven for an individual test input. By examining training data used to form the underlying model functional correlation, an estimate of the reliability that a model will make the correct prediction can be made. The Sequential Probability Ratio Test is used to derive a qualitative evaluation for reliability based on hypothesis testing. The PAR-FIT framework achieves this by implementing a particle filter and the sequential probability ratio test algorithms on the machine learning model training data to determine relevancy of new individual test samples to the training dataset. The kernel function evaluates the local proximity and density of training data used to derive a prediction outcome. Particles are used to probabilistically determine which training data to evaluate for proximity. For test samples that are within a close proximity to and surrounded by multiple training data points, the evaluated reliability of the prediction is high. For test samples that are anomalies not represented by the training dataset, in low density data clusters, or are far from existing data points, the evaluated reliability is low as insufficient training evidence exists to suggest the model is capable of making the correct prediction. Sequential Probability Ratio Test is further used to determine when a hypothesis on whether a signal can be rejected or accepted for use. The ratio test collects sequence information from the particle filter to test whether the signal is anomalous or normal via hypothesis testing of the underlying distributions.

Chen, Edward [Idaho National Laboratory (INL), Ida↗
DOE OSTI2025

Decentralized Distributed Proximal Policy Optimization (DD-PPO) for High Performance Computing Scheduling on Multi-User Systems

Resource allocation in High Performance Computing (HPC) environments presents a complex and multifaceted challenge for job scheduling algorithms. Beyond the efficient allocation of system resources, schedulers must account for and optimize multiple performance metrics, including job wait time and system throughput. Traditional heuristic-based scheduling algorithms increasingly struggle and lack the efficiency needed to meet the demands and address the complexity and scale of modern HPC systems. Consequently, recent research efforts have focused on leveraging advancements in Artificial Intelligence (AI) and Deep Learning (DL), particularly Reinforcement Learning (RL), to develop more adaptable and intelligent scheduling strategies. Previous RL-based scheduling approaches have explored a range of algorithms, from Deep Q-Networks (DQN) to Proximal Policy Optimization (PPO), and more recently, hybrid methods that integrate Graph Neural Networks (GNNs) with RL techniques. However, a common limitation across these methods is their reliance on relatively small datasets, with few methods being evaluated using large-scale, multi-million-job trace datasets representative of real-world HPC workloads. Moreover, existing RL schedulers face scalability issues due to centralized policy updates, which hinder training efficiency and performance when applied to large datasets. This study introduces a novel RL-based scheduler utilizing Decentralized Distributed Proximal Policy Optimization (DD-PPO) algorithm, which supports large-scale distributed training across multiple workers without requiring parameter synchronization at every step. By eliminating reliance on centralized updates to a shared policy, the DD-PPO scheduler enhances scalability, training efficiency, and sample utilization. Experimental validation using a large real-world dataset containing over 11.5 million job traces collected from petascale HPC systems over six years assesses the influence of dataset scale on training effectiveness and compares DD-PPO performance to traditional and advanced scheduling approaches. The experimental results demonstrate improved scheduling performance in comparison to both heuristic-based schedulers and existing RL-based scheduling algorithms.

AI↗
DOE OSTI2021

Ionizing radiation effects in SONOS-based neuromorphic inference accelerators

Here, we evaluate the sensitivity of neuromorphic inference accelerators based on Silicon-Oxide-Nitride-Oxide-Silicon (SONOS) charge trap memory arrays to total ionizing dose (TID) effects. Data retention statistics were collected for 16 Mbit of 40 nm SONOS digital memory exposed to ionizing radiation from a Co-60 source, showing good retention of the bits up to the maximum dose of 500 krad(Si). Using this data, we formulate a rate-equation-based model for the TID response of trapped charge carriers in the ONO stack, and predict the effect of TID on intermediate device states between ‘program’ and ‘erase’. This model is then used to simulate arrays of low-power, analog SONOS devices that store 8-bit neural network weights and support in situ matrix-vector multiplication. We evaluate the accuracy of the irradiated SONOS-based inference accelerator on two image recognition tasks – CIFAR-10 and the challenging ImageNet dataset – using state-of-the-art convolutional neural networks, such as ResNet-50. We find that across the datasets and neural networks evaluated, the accelerator tolerates a maximum TID between 10 krad(Si) and 100 krad(Si), with deeper networks being more susceptible to accuracy losses due to TID.

43 PARTICLE ACCELERATORS↗
DOE OSTI2021

Optimizing Error-Bounded Lossy Compression for Scientific Data by Dynamic Spline Interpolation

Today's scientific simulations are producing vast volumes of data that cannot be stored and transferred efficiently because of limited storage capacity, parallel I/O bandwidth, and network bandwidth. The situation is getting worse over time because of the ever-increasing gap between relatively slow data transfer speed and fast-growing computation power in modern supercomputers. Error-bounded lossy compression is becoming one of the most critical techniques for resolving the big scientific data issue, in that it can significantly reduce the scientific data volume while guaranteeing that the reconstructed data is valid for users because of its compression-error-bounding feature. In this paper, we present a novel error-bounded lossy compressor based on a state-of-the-art prediction-based compression framework. Our solution exhibits substantially better compression quality than all of the existing error-bounded lossy compressors, with comparable compression speed. Specifically, our contribution is threefold. (1) We provide an in-depth analysis of why the best-existing prediction-based lossy compressor can only minimally improve the compression quality. (2) We propose a dynamic spline interpolation approach with a series of optimization strategies that can significantly improve the data prediction accuracy, substantially improving the compression quality in turn. (3) We perform a thorough evaluation using six real-world scientific simulation datasets across different science domains to evaluate our solution vs. all other related works. Experiments show that the compression ratio of our solution is higher than that of the second-best lossy compressor by 20%similar to 460% with the same error bound in most of the cases.

Zhao, Kai↗
DOE OSTI2023

Data-driven key performance indicators and datasets for building energy flexibility: A review and perspectives

Energy flexibility, through short-term demand-side management (DSM) and energy storage technologies, is now seen as a major key to balancing the fluctuating supply in different energy grids with the energy demand of buildings. This is especially important when considering the intermittent nature of ever-growing renewable energy production, as well as the increasing dynamics of electricity demand in buildings. This paper provides a holistic review of (1) data-driven energy flexibility key performance indicators (KPIs) for buildings in the operational phase and (2) open datasets that can be used for testing energy flexibility KPIs. The review identifies a total of 48 data-driven energy flexibility KPIs from 87 recent and relevant publications. These KPIs were categorized and analyzed according to their type, complexity, scope, key stakeholders, data requirement, baseline requirement, resolution, and popularity. Moreover, 330 building datasets were collected and evaluated. Of those, 16 were deemed adequate to feature building performing demand response or building-to-grid (B2G) services. The DSM strategy, building scope, grid type, control strategy, needed data features, and usability of these selected 16 datasets were analyzed. This review reveals future opportunities to address limitations in the existing literature: (1) developing new data-driven methodologies to specifically evaluate different energy flexibility strategies and B2G services of existing buildings; (2) developing baseline-free KPIs that could be calculated from easily accessible building sensors and meter data; (3) devoting non-engineering efforts to promote building energy flexibility, standardizing data-driven energy flexibility quantification and verification processes; and (4) curating and analyzing datasets with proper description for energy flexibility assessm.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

Evaluation of Data Catalog Software for Hanford Site Environmental Datasets

Environmental information and data underpin achievement of the U.S. Department of Energy (DOE) Office of Environmental Management (EM) mission at the Hanford Site. The Hanford Environmental Data Management (HEDM) Program is the DOE Richland Operations Office (RL) approach to develop and implement a formal program for managing environmental data and the associated records, materials, and systems at the Hanford Site. The current project, contract, organization, and contractor-specific efforts at managing environmental data sets are insufficient to provide orderly, long-term, site-wide access. A vital element to be created within the HEDM program plan is a catalog of data sources, called the Hanford Environmental Information and Data Index (HEIDI), that will enable long-term access and retrievability for the multiple independent sources of data that might otherwise be difficult to discover. This report compares leading open source and commercial data catalog platforms using criteria to assess the functionality needed to develop the HEIDI catalog of Hanford data sources that connects and exchanges data with established Hanford Local Area Network (HLAN) enterprise information technology systems. Proprietary platforms evaluated included ArcGIS Enterprise Sites, Junar, OpenDataSoft, and Socrata, and non-proprietary platforms included Energy Data eXchange (EDX), Comprehensive Knowledge Archive Network (CKAN), and DKAN (a Drupal-based open data portal based on CKAN). Capabilities supporting data discoverability, retrieval, and archival, as well as metadata standard requirements and integration into the HLAN were rated as either failing to meet requirements (F), meeting requirements (M), or exceeding requirements by delivering additional desired features (E). The lowest rating for any capability area was assigned as the overall rating for the platform. These findings enable DOE-RL and the contractors implementing the HEDM plan to focus on candidate tools likely to meet the requirements for implementing HEIDI. All of the platforms receiving an overall rating of ‘F’ were unable to be deployed on Hanford infrastructure or within dedicated cloud resources. A propriety software-as-a-service (SaaS) model of delivering a data catalog (e.g., found in software such as Junar and OpenDataSoft) favors consistency across customers at the expense of customization and configurable roles that are needed for Hanford work. Hosting data on a shared commercial platform places limits on dataset size (maximum of 240 Mb for OpenDataSoft), a significant limitation for HEIDI implementation. EDX, a government data catalog based on CKAN, received the ‘F’ rating due to an inability to incorporate authentication from HLAN into the system. Among platforms rated ‘M’ or ‘E’, only the Socrata platform had a SaaS delivery model. In contrast to other SaaS platforms, Socrata provided custom roles and gateways that allow local datasets to be incorporated into an online catalog. Socrata also complies with the Federal Risk and Authorization Management Program, a significant benefit for cloud-based management of Hanford data. The other platforms rated ‘M’ or ‘E’, ArcGIS Enterprise Sites, CKAN, and DKAN, provide fully self-hosted options, allowing for greater control and flexibility with the HEIDI catalog. These widely used tools have supportive communities of practice, extensive customization options, and demonstrated deployments that provide evidence that they can meet requirements, often deliver additional desired features, and work well with federal government systems. Completely customized alternatives built on a collection of applications were not evaluated because achieving similar performance to CKAN or DKAN requires substantial resources, especially in the absence of the active communities that have grown to support these tools. ArcGIS Enterprise Sites, Socrata, CKAN, and DKAN were evaluated as strong candidates for successful implementation with HEIDI.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Evaluation of normalization strategies for mass spectrometry-based multi-omics datasets

Introduction Data normalization is crucial for multi-omics integration, reducing systematic errors and maximizing the likelihood of discovering true biological variation. Most studies assess normalization for a single omics type or use datasets from separate experiments. Few address time-course data, where normalization might bias temporal differentiation. In this study, we compared common normalization methods and a machine learning approach, Systematical Error Removal using Random Forest (SERRF), using multi-omics datasets generated from the same experiment—even from the same cell lysate. Objectives To develop a straightforward process to assess normalization effects and identify the most robust methods across multi-omics datasets. Methods We analyzed metabolomics, lipidomics, and proteomics datasets from primary human cardiomyocytes and motor neurons exposed to acetylcholine-active compounds over time. Normalization effectiveness was evaluated based on improvement in QC features consistency and observing the change in treatment and time-related variance. Results Probabilistic Quotient Normalization (PQN) and Locally Estimated Scatterplot Smoothing (LOESS) QC were identified as optimal for metabolomics and lipidomics, while PQN, Median, and LOESS normalization excelled for proteomics. These methods consistently enhanced QC feature consistency in metabolomics and lipidomics, and preserved time-related variance or treatment-related variance in proteomics, demonstrating their effectiveness and robustness. SERRF normalization, applied only to metabolomics in this study, outperformed other methods in some datasets but inadvertently masked treatment-related variance in others. Conclusion Our evaluation identified PQN and LoessQC as the top methods for metabolomics and lipidomics, and PQN, Median, and Loess normalization for proteomics, in multi-omics integration in a temporal study.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Quantum-assisted associative adversarial network: applying quantum annealing in deep learning

Abstract Generative models have the capacity to model and generate new examples from a dataset and have an increasingly diverse set of applications driven by commercial and academic interest. In this work, we present an algorithm for learning a latent variable generative model via generative adversarial learning where the canonical uniform noise input is replaced by samples from a graphical model. This graphical model is learned by a Boltzmann machine which learns low-dimensional feature representation of data extracted by the discriminator. A quantum processor can be used to sample from the model to train the Boltzmann machine. This novel hybrid quantum-classical algorithm joins a growing family of algorithms that use a quantum processor sampling subroutine in deep learning, and provides a scalable framework to test the advantages of quantum-assisted learning. For the latent space model, fully connected, symmetric bipartite and Chimera graph topologies are compared on a reduced stochastically binarized MNIST dataset, for both classical and quantum sampling methods. The quantum-assisted associative adversarial network successfully learns a generative model of the MNIST dataset for all topologies. Evaluated using the Fréchet inception distance and inception score, the quantum and classical versions of the algorithm are found to have equivalent performance for learning an implicit generative model of the MNIST dataset. Classical sampling is used to demonstrate the algorithm on the LSUN bedrooms dataset, indicating scalability to larger and color datasets. Though the quantum processor used here is a quantum annealer, the algorithm is general enough such that any quantum processor, such as gate model quantum computers, may be substituted as a sampler.

Wilson, Max (ORCID:0000000207983391)↗
DOE OSTI2026

Evaluation of daily gridded climate products using in situ FLUXNET data and tree growth modeling

Gridded climate data products have facilitated research in climate and ecology by providing meteorological data continuously across large spatial scales. However, the sensitivity of scientific outcomes to dataset choice remains poorly understood, and evaluation using station-based records can favor datasets built heavily on weather stations. Here, we evaluate seven high-resolution daily gridded datasets covering the contiguous United States using independent meteorology from the FLUXNET2015 dataset, with a focus on the implications of dataset choice for process-based tree growth modeling. We find that gridded products tend to capture temperature accurately while consistently overestimating the magnitude and frequency of precipitation and its extremes. Moreover, datasets vary in how they define a ‘day,’ which significantly affects temporal alignment with FLUXNET2015 observations. Despite differences among the datasets, the interannual variability in tree ring simulations is insensitive to dataset choice, likely because daily-scale biases are averaged out through accumulated growth across several months. However, inaccuracies in temperature and precipitation can significantly bias modeled xylem cell production, with systematically higher annual precipitation in the gridded datasets leading to greater xylem production compared to simulations using in situ data. Our results suggest that model applications, especially those that integrate to time scales longer than one day, are likely insensitive to climate dataset choice, but applications that are sensitive to daily climate variations or to absolute climate values need to carefully consider biases in gridded climate products.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

A Publicly Available, Annotated Dataset for Naturalistic Driving Study and Computer Vision Algorithm Development

Oak Ridge National Laboratory developed and implemented a data collection effort to create a dataset for use in evaluating and testing algorithms for analyzing driver behavior under controlled settings for support of the Federal Highway Administration’s Exploratory Advanced Research Program. This collection is called the ORNL Naturalistic Driving Study Sample (ONDSS). The dataset is designed to emulate aspects of the Second Strategic Highway Research Project (SHRP2), which contained a massive naturalistic driving study (NDS) with over 3000 drivers between 2010 and 2013 using their personal vehicles, with over 4300 person-years of data collected [HANKEY].

42 ENGINEERING↗