<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Research in Progress (RIP)</title>
    <link>https://rip.trb.org/</link>
    <atom:link href="https://rip.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Research in Progress (RIP)</title>
      <url>https://rip.trb.org/Images/PageHeader-wTitle-RIP.jpg</url>
      <link>https://rip.trb.org/</link>
    </image>
    <item>
      <title>Enhancing Heavy Vehicle Crash Prevention in North Dakota through Machine Learning and Weather Data Integration</title>
      <link>https://rip.trb.org/View/2683255</link>
      <description><![CDATA[Heavy vehicle crashes continue to be a persistent safety concern across the Midwest, with several states reporting disproportionately high rates of incidents involving large trucks. According to the National Safety Council, in 2023, North Dakota recorded 18% of its fatal crashes involving large trucks, placing it among the highest in the nation. Neighboring states, such as Nebraska (16%) and Iowa, also face elevated risks. Illinois reported 7,509 truck accidents in 2022, ranking among the top five states nationwide. In North Dakota, the risks are especially pronounced during the winter months. In 2023, 64% of heavy vehicle crashes occurred between October and March, with 81% of these crashes taking place in rural areas. These figures highlight how weather conditions and geography amplify the risk associated with large-truck travel in the region. Further, crashes in rural areas in challenging weather conditions poses immense issues for first responders and their ability to provide timely medical care to crash victims.   

Traditional safety strategies have struggled to account for the dynamic, real-time factors that contribute to crash risk. Static approaches often fall short when adverse weather, road conditions, and traffic volume interact in unpredictable ways. This gap highlights the urgent need for predictive, data-driven solutions.  

This proposal aims to investigate the application of machine learning (ML) models, combined with weather and crash data, to predict high-risk scenarios before accidents occur, to support planning for safety and emergency response needs. By leveraging predictive analytics, North Dakota could enhance resource allocation, deploy preventive interventions, and reduce the frequency and severity of heavy vehicle crashes. The high incidence of winter crashes and the limitations of conventional methods make North Dakota an ideal proving ground for an innovative, ML-driven approach to roadway safety.  

The study will utilize historical crash records for heavy vehicles in North Dakota, including crash type, severity, date, and time, combined with corresponding weather data such as temperature, precipitation, snowfall, and visibility. Feature engineering will create representations of temporal and weather conditions relevant to crash severity. Machine learning models, including Random Forest, XGBoost, and Neural Networks, will be trained to predict crash severity. To ensure interpretability, SHAP (SHapley Additive exPlanations) will be applied to quantify the contribution of each feature to individual predictions and overall model behavior. This analysis will reveal which weather or temporal factors most strongly influence severe crashes, both globally across the dataset and locally for specific incidents. High-risk periods and conditions identified by the model, along with explanations provided via SHAP, will be visualized both temporally and geographically, offering actionable insights to support targeted preventive measures and inform DOT decision-making.  ]]></description>
      <pubDate>Tue, 24 Mar 2026 14:09:40 GMT</pubDate>
      <guid>https://rip.trb.org/View/2683255</guid>
    </item>
    <item>
      <title>Smart AI-Technology Employment for Crash Data Analysis</title>
      <link>https://rip.trb.org/View/2658653</link>
      <description><![CDATA[In 2024, motor vehicle traffic crashes in the United States resulted in 39,345 fatalities and approximately 2.44 million injuries. Drug-involved driving has emerged as a significant and growing contributor to these crashes, yet unlike alcohol impairment, drug involvement remains difficult to detect due to the diversity of substances and their varying effects on driver behavior. Research indicates that more than 20% of drug-related crashes are erroneously recorded, compromising data accuracy and undermining effective policy responses. While most traffic safety studies rely on statistical approaches and quantitative data to identify crash severity contributors, these methods often fail to capture the nuanced circumstances surrounding drug-involved incidents. Manual review of crash narratives is inefficient and error-prone, and traditional keyword-based searches frequently miss contextual information, producing false positives and negatives. Although previous studies have demonstrated that crash narratives can reveal contributing factors difficult to obtain through conventional quantitative analysis, these efforts were limited to classic text mining techniques that consider only individual word frequencies without capturing language semantics. This research leverages advances in Artificial Intelligence and Natural Language Processing to systematically analyze crash narrative data, a historically underutilized resource to extract new insights about drug-involved vehicle crashes. The study aims to provide foundational evidence for addressing drug-involved crashes and improving traffic safety management while advancing methodological approaches for narrative-based crash analysis. Key findings and methodological innovations will be integrated into educational and outreach initiatives.]]></description>
      <pubDate>Mon, 26 Jan 2026 00:27:52 GMT</pubDate>
      <guid>https://rip.trb.org/View/2658653</guid>
    </item>
    <item>
      <title>Improving Crash Data for Active Transportation Users</title>
      <link>https://rip.trb.org/View/2655767</link>
      <description><![CDATA[In recent years, the United States has experienced sharp, inexplicable increases in the number of pedestrian fatalities. In response to this disturbing trend, state highway safety offices (SHSOs) and state departments of transportation (DOTs) have been conducting safety analyses to better understand the problem and develop remediation plans.  

Crash data is the primary source of information used for safety analysis. This critical data source, however, has many limitations, including inconsistencies in reporting, inaccurate or incomplete coding of crashes, and underreporting, especially for active transportation/non-motorized users (herein after referred to as active transportation users). Also, crash typing (used to describe events and movements prior to a crash) can lack details for pedestrian- and bicycle-involved crashes, and in some cases must be constructed using multiple variables.

Improving pedestrian and bicyclist injury and fatality data, adopting consistent typing methods at the national and local levels, improving data storage, sharing and accessibility, and integrating police and hospital crash data would help practitioners understand risk factors and potential countermeasures. It is important to understand the reasons for crash data limitations, related implications, and measures that can be taken to improve the completeness, consistency, and accuracy of crash data for active transportation users.

Research is needed to improve the current state of the practice for collecting injury and fatality data for active transportation users.

For the purposes of this project, active transportation users include pedestrians, bicyclists/pedalcyclists (both manual and electric), and persons on personal conveyances (i.e., e-scooter users, mobility device users [e.g., manual and motorized wheelchairs], skateboard riders, users of similar personal conveyances).

OBJECTIVE: The objective of this research is to continue and complete the work begun under NCHRP Project 07-35, “Improving Crash Data for Active Transportation Users” to develop a guide to improve the completeness, consistency, and accuracy of crash data for active transportation users. ]]></description>
      <pubDate>Mon, 19 Jan 2026 16:41:20 GMT</pubDate>
      <guid>https://rip.trb.org/View/2655767</guid>
    </item>
    <item>
      <title>Investigating Transit Characteristics and Road Safety Outcomes</title>
      <link>https://rip.trb.org/View/2643028</link>
      <description><![CDATA[Public transit systems influence roadway safety through changes in travel behavior, congestion levels, and modal distribution, yet the specific mechanisms linking transit characteristics to safety outcomes are not well quantified. While prior studies suggest that higher transit use is associated with improved safety, agencies lack clear guidance on which system features contribute most to these effects. This project addresses that gap through large-scale data integration and predictive modeling.

The research will combine crash data, transit system attributes, roadway network characteristics, and demographic indicators from national and regional data sources. Machine learning models will be developed to predict crash rates as a function of transit network size, service intensity, demand, and multimodal shares. Explainable modeling techniques will be used to identify the most influential predictors of safety outcomes at both metropolitan and town levels. Results will provide actionable evidence to support data-driven transit planning and roadway safety strategies.]]></description>
      <pubDate>Thu, 18 Dec 2025 14:54:04 GMT</pubDate>
      <guid>https://rip.trb.org/View/2643028</guid>
    </item>
    <item>
      <title>Development of a Network-Level Data-Driven High Friction Surface Treatment Location Selection Approach Leveraging Remote Sensing Technologies</title>
      <link>https://rip.trb.org/View/2643023</link>
      <description><![CDATA[High Friction Surface Treatment (HFST) is an effective countermeasure for reducing crashes at horizontal curves, yet current site selection practices rely heavily on historical crash data and manual field inspections. These approaches limit agencies’ ability to proactively identify high-risk locations and efficiently allocate limited safety resources. This project addresses these limitations by developing a scalable, data-driven framework for HFST site prioritization at the network level.

The research will create an automated data-processing pipeline that extracts roadway geometry and surface characteristics from mobile LiDAR and video log imagery, including curve radius, superelevation, signage, and surface condition. These features will be integrated with pavement condition and crash data to identify high-risk and constructible HFST locations. The approach will be validated through a case study using Massachusetts Department of Transportation (MassDOT) roadway and crash data. Results will provide transportation agencies with a transferable methodology for proactive HFST deployment, improving safety outcomes and supporting more efficient infrastructure management.]]></description>
      <pubDate>Thu, 18 Dec 2025 14:36:51 GMT</pubDate>
      <guid>https://rip.trb.org/View/2643023</guid>
    </item>
    <item>
      <title>Using Linked Data to Explore the Accuracy of Crash Reported Injuries of Minors</title>
      <link>https://rip.trb.org/View/2640191</link>
      <description><![CDATA[Police crash reports often provide the first record of injury severity for minors involved in motor vehicle crashes, yet these reports may not always match clinical assessments. Differences between the reported level of injury and the medically confirmed level can influence emergency response decisions and limit the usefulness of crash databases for safety analysis. This project will link crash data from the Connecticut Crash Data Repository with hospital discharge and Emergency Medical Services (EMS) datasets to compare police reported injury codes with medically derived measures. The analysis will document where inconsistencies occur and examine how factors such as crash location, agency type, passenger protection, and driver behavior relate to reporting accuracy.

The linked dataset will cover crashes involving minors from 2015 through 2024 and will support regression based evaluations of injury classification accuracy across multiple contexts. By identifying sources of error, the study will help improve data quality and support better training and data collection procedures for law enforcement and partner agencies. The resulting insights will strengthen statewide injury surveillance systems and guide the development of safety strategies for children and adolescents.]]></description>
      <pubDate>Thu, 11 Dec 2025 13:50:56 GMT</pubDate>
      <guid>https://rip.trb.org/View/2640191</guid>
    </item>
    <item>
      <title>Calibration and Implementation of Highway Safety Manual Bicyclist and Pedestrian Intersection Crash Prediction Models</title>
      <link>https://rip.trb.org/View/2635921</link>
      <description><![CDATA[The forthcoming 2nd Edition of the AASHTO Highway Safety Manual (HSM2) will introduce dedicated crash prediction models (CPMs) for pedestrian and bicyclist crashes at intersections, midblock crossings, and roadway segments.  The goal of this research is to calibrate the HSM2 pedestrian and bicyclist intersection CPMs using Virginia-specific data. The research outcomes will enhance the accuracy of nonmotorized crash predictions and support Virginia Department of Transportation's (VDOT’s) broader goals of data-driven planning, design decision-making, and funding prioritization for safety improvements. To achieve this goal, the research will (1) assemble a comprehensive dataset for selected representative intersections in Virginia, including crash history, exposure data, and roadway and roadside design features required by the HSM2 CPMs, (2) develop appropriate methods for estimating pedestrian and bicyclist exposure at intersections, considering available data sources, (3) develop a robust calibration methodology that accounts for the variability of contextual settings, exposure ranges, facility types and jurisdictions, etc., and (4) design a practical tool and accompanying guidance to help VDOT implement and maintain the calibrated pedestrian and bicyclist CPMs.]]></description>
      <pubDate>Thu, 04 Dec 2025 08:52:41 GMT</pubDate>
      <guid>https://rip.trb.org/View/2635921</guid>
    </item>
    <item>
      <title>Regional Disparities in Work Zone Crashes: Understanding Factors and Predictive Modeling for Targeted Safety Measures
</title>
      <link>https://rip.trb.org/View/2627354</link>
      <description><![CDATA[Roadway work zones play a vital role in maintaining and improving infrastructure, yet they often expose workers and drivers to dangerous situations, leading to concerning frequencies of occupational and traffic accidents in the United States. With over 700 fatalities and thousands of injuries annually attributed to work zone crashes, efforts to enhance safety have been hindered by the complexity and variability of contributing factors. The escalating fatalities, coupled with growing infrastructure demands in U.S. Department of Transportation Region 7—encompassing Missouri, Iowa, Nebraska, and Kansas—underscore the imperative to address underlying causes and improve work zone safety. This study aims to address this persistent issue by analyzing work zone crash data in Region 7 and comparing it with other regions to identify influential factors. By leveraging recurrent neural networks (RNNs) to develop region-specific predictive models, the research seeks to forecast crash occurrences and provide targeted insights for policymakers and transportation authorities. Ultimately, the research aims to deepen understanding of regional disparities in work zone crash dynamics, enabling effective resource prioritization and implementation of tailored safety measures. The development of predictive models using RNNs holds promise for enhancing proactive safety planning and resource allocation, ultimately contributing to a nationwide reduction in work zone crashes and advancing the overarching goal of improving road safety for workers and motorists.
]]></description>
      <pubDate>Wed, 19 Nov 2025 15:35:55 GMT</pubDate>
      <guid>https://rip.trb.org/View/2627354</guid>
    </item>
    <item>
      <title>Evaluating the Economic and Safety Trade-offs of Interchange and Access Drive Separation Distances
</title>
      <link>https://rip.trb.org/View/2627344</link>
      <description><![CDATA[The research project will evaluate whether the Iowa Department of Transportation’s (Iowa DOT) minimum separation standards between interchanges and first access points are overly restrictive and potentially detrimental to development opportunities around those interchanges. To achieve this, the project will utilize deep learning techniques to analyze high-resolution aerial photographs to identify interchanges on state-owned roadways, their first driveway access points, and the specific aspects of development status, such as the presence of commercial or residential buildings, vacant land, or agricultural use of the surrounding land. Crash data from the Iowa dataset will be examined to assess safety outcomes about these separation distances. A critical part of the analysis will involve evaluating the economic potential of these lands and estimating the impact of separation standards on land utilization and potential economic growth. 
In addition to state-owned interchanges, the study will identify non-interchange intersections with roadways with similar AADT levels, the number of lanes, if a median is present, and other relevant geometric features to access management. The closest access point will be determined for these intersections, mirroring the approach taken with the interchanges. The crash history for these locations will be retrieved to compare the safety performance of interchanges and non-interchange intersections directly.
This analysis, focusing on interchange and access point separation distances, will help isolate the effect of these separation standards on safety and development, controlling for traffic volume and other features. By examining interchange and non-interchange sites under similar conditions, the research will determine if the minimum separation distances at interchanges are justified or could be adjusted to better balance safety with economic development, potentially informing future policy decisions. The research will also determine the amount of developable land that could be available should the standards be relaxed.
]]></description>
      <pubDate>Wed, 19 Nov 2025 14:36:56 GMT</pubDate>
      <guid>https://rip.trb.org/View/2627344</guid>
    </item>
    <item>
      <title>Hotspot Stability of Freight Vehicle Crashes Involving Vulnerable Road Users: A Spatio-Temporal Perspective</title>
      <link>https://rip.trb.org/View/2625586</link>
      <description><![CDATA[This research will analyze the interaction between two of the most different transportation road users that interact on roads—freight vehicles and vulnerable road users (VRU), i.e., pedestrians and bicyclists. The research objective of this project is to identify the temporal stability of hotspots in (1) non-fatal crashes, (2) fatal crashes, and (3) all crashes (non-fatal and fatal) between freight vehicles and VRU in two U.S. States. This research proposes a novel spatiotemporal analysis to answer whether crash hotspots intensify over time (i.e., the number of crashes increases over time at the same location) or if it stays the same over time.  In terms of processes, the first one is collecting the data on fatal, non-fatal, and all crashes of both States into a single file, cleaning it, and ensuring its validity/accuracy/consistency. Once the data collection is ready, the second process focuses on merging the panel data into a space-time cube. This arrangement will host on a single data array geographical and temporal data of the total number of (1) non-fatal crashes, (2) fatal crashes, and (3) all crashes between freight vehicles and VRU for each State. The third process is calculating a Local Indicator of Spatial Association Statistic (the Gettis Ord*) to identify crash hotspot locations for each year of analysis for each State, and estimate emerging hotspot patterns based on the panel data results. The fourth process will use crash hotspot locations (identified in process three) and data from the County Business Pattern data, the Census Tract Data, and the American Community Survey to compare urban economic and built environment characteristics between different types of hotspots (e.g., recent versus consecutive hotspots), and identify common factors and differences. Specifically, the research team will compute an ANOVA and a post hoc test to identify statistical differences between crash hotspot locations. The last process focuses on visualizing the results on a geographic information system (GIS) software or tables for statistical analysis.  The results of the spatiotemporal analysis will be correlated with urban economic and built environment features to identify common factors in hotspot locations that could have influenced road crashes in both States. These factors include built environment attributes and the number of establishments by industry sector, among others.]]></description>
      <pubDate>Tue, 18 Nov 2025 14:19:08 GMT</pubDate>
      <guid>https://rip.trb.org/View/2625586</guid>
    </item>
    <item>
      <title>Exposing the Dangers of Distance: Mining Crash Narratives to Explore Why Pedestrians Face Severe Injury and Death Far From Home</title>
      <link>https://rip.trb.org/View/2625590</link>
      <description><![CDATA[Pedestrian fatalities in the United States have risen by 83% over the past 15 years, with much of the increase occurring on multilane suburban arterials. In Tennessee, deaths nearly tripled between 2009 and 2022, with studies linking crashes to high-speed midblock locations lacking pedestrian infrastructure. Spatial analysis shows pedestrians are being struck farther from home: in 2014 the median distance between residence and crash site was 1.5 miles, compared to four miles by 2023, while the share of crashes within one mile of home fell from 46% to 30%. Relative to city centers, crash locations remain stable, but the distance between city centers and pedestrian residences has grown, indicating that more crashes involve individuals living farther from urban cores. These shifts suggest pedestrians are traveling into distant, high-risk environments, raising essential questions about why they are walking in such areas and what broader urban trends contribute to this exposure. This study applies a hybrid methodology combining structured crash records with insights from unstructured police narratives from Tennessee’s Integrated Traffic Analysis Network (2014–2024). A home-based approach links pedestrian and driver addresses with U.S. Census block group characteristics, including income levels, vehicle ownership, education, commuting modes, and housing density, to better understand who is involved in these crashes. Artificial intelligence is used to analyze crash narratives for trip purposes such as traveling to grocery stores, bus stops, schools, or workplaces, offering contextual information not captured in standard crash reports. Specifically, this study will locally deploy an open-source large language model (e.g., Gemma or Grok) to serve as a traffic crash analysis agent, capable of addressing questions that help uncover the motivations behind pedestrian trips based on police crash narratives. By conducting all processing locally, this approach ensures the privacy of both pedestrians and drivers is preserved. Together, these methods distinguish between near-home and far-from-home crashes, highlight populations more frequently affected, and examine the role of broader urban development patterns. The findings will support city- and neighborhood-level safety strategies, helping target interventions on hazardous arterials and informing policies for improved safety.]]></description>
      <pubDate>Mon, 17 Nov 2025 16:49:27 GMT</pubDate>
      <guid>https://rip.trb.org/View/2625590</guid>
    </item>
    <item>
      <title>Pedestrian Crashes at High-Speed Intersections: Applying AI for Crash Narrative Analysis to Identify Common and Edge Cases</title>
      <link>https://rip.trb.org/View/2625591</link>
      <description><![CDATA[High-speed intersections, with speed limits of 35 mph or higher, and long crossings are particularly risky for pedestrians. Tragically, in 2022, 983 pedestrians were killed at signalized intersections, representing about 16% of all pedestrian fatalities. Intersections are concerning due to pedestrians' difficulty navigating them and the numerous conflict points they present. The proposed research addresses a critical gap: understanding pedestrian crash descriptors using structured and unstructured (narrative) data on high-speed intersections and identifying simple and complex (or edge) cases—unusual or extreme crashes that deviate substantially from the typical ones. Complex cases represent exceptional circumstances with many contributing factors. Understanding them will help inform current pedestrian safety strategies and help improve how autonomous vehicle algorithms anticipate potential pedestrian conflicts.  The research question is: What are the different types of pedestrian crashes and injuries at intersections, and what are their complexity levels?  To answer this question, the team will use data from crashes from Tennessee’s Integrated Traffic Analysis Network (TITAN) and Wisconsin’s WisTransPortal, for which the research team has access to police crash narratives. The team will separate pedestrian crashes at high-speed signalized intersections (with speed limits of 35 mph or higher) and compare them to those at intersections with speed limits of less than 35 mph. The team will use AI to develop high-quality, detailed crash descriptors from the narratives of police reports and quantitative crash data. Natural language processing and feature extraction techniques will categorize pedestrian crashes into specific types based on detailed pre-crash actions, human errors, and circumstances obtained from structured and unstructured data. The study will identify edge cases and relevant safety countermeasures (e.g., conflict reductions) while providing a nuanced understanding of crash circumstances (relative to current practice).   The study will create a unique and comprehensive crash database that can provide deep insights into the range of injuries, crash attributes (e.g., crash location within the intersection or pedestrian and driver actions), precrash positions, driver and pedestrian impairment, and roadway conditions, and design (e.g., visibility, number of lanes, pedestrian crossing facilities). The study will apply rigorous analysis methods, including unsupervised learning techniques to identify complex cases and inference-based frequentist methods to quantify key correlates of crash injuries. Cluster analysis, specifically through hierarchical or k-means techniques, will differentiate complex crash cases from more common ones, effectively isolating extreme cases deviating from typical patterns.   In addition to highlighting the issue of pedestrian crashes at intersections and their correlates, a unique aspect of this study is the identification of complex cases. By doing so, the study aims to uncover the underlying patterns and risk factors that contribute to complex and unusual pedestrian crashes at intersections. Rather than focusing solely on the common crash situations, considering a wide range of possibilities and using a unique database helps to understand and address common and rare cases for high-speed and low/medium-speed intersections (especially relevant, given the adoption of vehicle automation and higher safety standards), ensuring a safer environment for vulnerable road users.]]></description>
      <pubDate>Mon, 17 Nov 2025 16:29:12 GMT</pubDate>
      <guid>https://rip.trb.org/View/2625591</guid>
    </item>
    <item>
      <title>Mining Police Crash Report Narratives: A Natural Language Processing Approach to Identify Bus-Stop Related Crashes</title>
      <link>https://rip.trb.org/View/2625594</link>
      <description><![CDATA[Transit riders are a particularly vulnerable population, as they often walk to and from bus stops, wait in areas where multiple transportation modes interact, and cross the road at least once during a round trip. Prior studies have identified a significant relationship between transit elements (i.e., stops, corridors, and ridership levels) and pedestrian crash locations. National databases like the Fatality Analysis Reporting System (FARS) reported 196 transit bus stop-related pedestrian crashes (2014–2022), while the Crash Report Sampling System (CRSS) reported 93 (2016–2022). This small sample appears to contradict rising pedestrian crash trends in the U.S., suggesting potential underreporting due to inconsistent definitions, lack of standardized fields for transit bus stop-related crashes, or variation in how crashes are coded. To address this gap, artificial intelligence methods like natural language processing (NLP), specifically named entity recognition (NER), can extract transit bus stop-related details from police crash report narratives. NER will be applied to Minnesota and Tennessee datasets to identify such crashes. The model will be trained, validated, and tested for generalizability using metrics like precision and recall. Results will be cross-analyzed with national databases (FARS, CRSS) to test the hypothesis that transit bus stop-related crashes are underreported. Misclassified cases will also be analyzed to identify patterns. While NER has been widely used to improve crash data quality, it has not been applied to identify transit bus stop-related crashes specifically. This approach could streamline data collection, reduce manual review time, and enhance the accuracy of pedestrian crash data. By addressing a critical gap in crash reporting, this work will improve the ability to study risks faced by transit riders and inform safety improvements at bus stops.]]></description>
      <pubDate>Mon, 17 Nov 2025 15:03:26 GMT</pubDate>
      <guid>https://rip.trb.org/View/2625594</guid>
    </item>
    <item>
      <title>Identification of Factors Contributing to the Decline of Traffic Fatalities in the United States</title>
      <link>https://rip.trb.org/View/2611417</link>
      <description><![CDATA[NCHRP Report 928 describes a comprehensive analysis of the factors associated with fatality rates in states, especially as they relate to the substantial drop in fatalities occurring in the years from 2008 to 2011. The analysis included predictors representing a wide variety of potentially related factors including travel, demographics, the economy, vehicle safety systems, and state spending on several categories of infrastructure and safety improvements. Understanding the broad array of factors that influence traffic safety in the United States is particularly important for state highway safety planning.

From 2008 to 2011, the United States, along with several other countries, experienced a substantial drop in annual traffic fatalities. State departments of transportation are keenly interested in capturing the key contributing factors to this decline so that the information can be used to help focus resources on effective countermeasures in future years. Safety countermeasures are implemented in a wide variety of ways—changing behavior (e.g., through driver education), changing vehicles (e.g., with Electronic Stability Control and other safety technologies), and changing the environment (e.g., improving roadways, laws, and enforcement). Safety can be influenced by factors other than safety efforts themselves, so it can be difficult to know which changes are responsible for overall reductions in fatalities.

Under NCHRP Project 17-67, the research conducted by the University of Michigan, with support from Texas A&M was asked to provide a multidisciplinary analysis of the relative influence of the types of factors that contributed to the national decline in the number of highway fatalities and rates in the United States during the years of 2008–2011. The research team used the Haddon matrix to identify many potential travel, demographic, economic, vehicle, and infrastructure influences on fatalities. Data on these factors were then collected from data sources publicly available at the state level (e.g., FHWA Highway Statistics). Annual state-level measures of these factors were compiled into a database covering the years from 2001 to 2012 and matched with fatalities from the Fatality Analysis Reporting System (FARS) database.

These data were analyzed using statistical methods to predict fatalities in states as well as changes over time. Because the statistical models closely approximated the reduction in fatalities from 2008 to 2011, the factors could then be evaluated in terms of their individual (and combined) contributions to fatalities. This report covers the scope of the problem, the data obtained to measure each factor, the statistical models, and the interpretation of results to understand how different factors play a role in total fatality counts. The knowledge gained from this process can be used to predict future fatality levels for planning at the state level and to provide insight into factors influencing these levels and actions that might reduce them. 
]]></description>
      <pubDate>Mon, 20 Oct 2025 17:48:26 GMT</pubDate>
      <guid>https://rip.trb.org/View/2611417</guid>
    </item>
    <item>
      <title>Exploring Completeness and Accuracy of Driver Crash Reporting of Property Damage Only Crashes</title>
      <link>https://rip.trb.org/View/2487369</link>
      <description><![CDATA[This project proposes an exploration of reprising an electronic citizen crash report for Property Damage Only (PDO) crashes to determine if the data provided is found to be adequately complete and accurate to supplement crash data in the absence of police reported PDO crashes. The study proposes to simulate PDO crash exposure among a population of drivers, with no experience in crash reporting, and analyze their inputted data in the MNCrash crash reporting system (or replica) to determine their degree of completeness and accuracy. Citizen provided crash data will be compared to data provided by trained police officers who are presented the same simulated crash scene and compared to a gold standard crash report (validated by multiple experts). Findings may result in the human-centered design and testing of an abbreviated electronic PDO crash report for citizen reporting and recommendations regarding its use.
]]></description>
      <pubDate>Wed, 08 Oct 2025 10:25:11 GMT</pubDate>
      <guid>https://rip.trb.org/View/2487369</guid>
    </item>
  </channel>
</rss>