Pedestrian safety is a growing concern for transportation planners and safety engineers at both local and state levels. Continued advancements in data availability, data integration abilities, and analysis methodologies offer new opportunities to identify factors influencing pedestrian safety and to quantify their effects to inform data-driven road safety management. The main objective of this study was to spatially integrate Highway Safety Information System data with multijurisdictional and emerging datasets to analyze two measures of pedestrian safety performance in Charlotte, NC: (1) the severity of a pedestrian crash that has occurred, and (2) the probability that a pedestrian crash will occur on a given roadway segment. To accomplish the objectives, the study explored several high-priority research topics in safety data and analysis, including pedestrian exposure analysis and probe data integration. The research team developed a pedestrian count model to predict pedestrian volumes at locations without pedestrian counts and integrated speed information from probe data to supplement other roadway and contextual transportation data available from several agencies. Pedestrian exposure at a given intersection was found to be significantly influenced by demographic and socioeconomic characteristics, employment, land use, sidewalk presence, transit access, and roadway and intersection characteristics. The project team identified numerous significant factors that influenced pedestrian crash severity and probability, including outputs from the pedestrian exposure model, observed vehicle speeds, traffic volumes, intersection proximity, and other crash-related factors. The results could be used to identify locations that are more susceptible to pedestrian safety issues.

A significant portion of both fatal and total crashes occurs at intersections in the United States. Skew angle may be a significant contributor to these crashes. This paper examines the effects of intersection angle on intersection safety performance. With seven years of crash data from Minnesota and five years of crash data from Ohio, random forest regression data mining and negative binomial regression models were developed to estimate crash modification functions at three-leg and four-leg stop-controlled intersections with two-lane and multilane major legs. Where possible, the results were compared between the two states and used to develop average crash modification function curves. This study shows that over half of the intersection types experience the highest number of predicted crashes when the intersection angle between roadway legs is between 50 degrees and 65 degrees. These results have practical implications for engineers and safety professionals. First, the crash modification function curves supplement and revise the guidance for intersection angle in the Highway Safety Manual and Policy on Geometric Design of Highways and Streets. Second, the functions offer new guidance to agencies planning intersection improvements. Third, the crash modification functions can be used to determine the safety effect of changes in intersection angle.

Background: Previous epidemiological studies have highlighted the high risk of injury to the head, thorax, and cervical spine in rollover crashes. However, such results provide limited information on whole-body injury distribution and multiple region injury patterns necessary for the improvement and prioritization of rollover-focused injury countermeasures.Methods: Sampled cases representing approximately 133,000 U.S. adult occupants involved in rollover crashes (between 1995 and 2013) sustaining moderate-to-severe injuries were selected from the National Automotive Sampling System Crashworthiness Data System database. A retrospective cohort study, based on a survey of population-based data, was used to identify relevant whole body injury patterns.Results: Among belted occupants injured in rollover crashes, 79.2% sustained injuries to only one body region. The three most frequently injured (AIS2+) body regions were head (42.1%), upper extremity (28.0%), and thorax (27.1%). The most frequent multi-region injury pattern involved the head and upper extremity, but this pattern only accounted for 2.3% of all of occupants with moderate or worse injuries.Conclusions: The results indicated that for rollover-dominated crashes, the frequently observed injury patterns involved isolated body regions. In contrast, multi-region injury patterns are more frequently observed in rollovers with significant planar impacts. Identification of region-specific injury patterns in pure rollover crashes is essential for clarifying injury mitigation targets and developing whole-body injury metrics specifically applicable to rollovers. (C) 2016 Elsevier Ltd. All rights reserved.

The National Automotive Sampling System/Crashworthiness Data System (NASS/CDS) remains the best US data source for understanding the magnitude of the opportunities for reducing rollover injuries to the various body regions. However, judicious analysis techniques are required to address the many confounding factors, including but not limited to the consequence of recent safety improvements such as electronic stability control and increased roof strength. To better assess the effect of recent safety improvements, the population of drivers in rollovers in light vehicles model year 2000 and later was examined. To address crash severity, the number of quarter-turns was used. Injuries were separated by body region and the HARM method of aggregating injuries was used to provide added weighting to the more severe injuries. For belted drivers in near-side rollovers, the fourth quarter-turn contained the most HARM and the highest injury risk, especially for chest injuries. For belted drivers in far-side rollovers, most of the chest injury HARM is fairly uniformly distributed between quarter-turns 2, 4, 6 and 8.

The National Highway Traffic Safety Administration (NHTSA), an agency of the United States Department of Transportation (US DOT), provides crashworthiness data sets serving as a starting point for highway research. Although the information has provided guidance to safety researchers and decision-makers for vehicle damage and occupant injury research, crash environment fields have been inadequate to serve the growing surveillance burden placed upon the Federal Highway Administration (FHWA), another agency of the US DOT. Recent advances in NHTSA data reporting, however, have allowed for supplemental data acquisition, at no added cost.

National Automotive Sampling System Crashworthiness Data System (NASS/CDS)1995-2009 was the basis for evaluating safety changes in the vehicle fleet with model year. The analysis shows that the mean AIS 3+ HARM for belted drivers in 1996-2009 model year vehicles had decreased injuries in side and frontal crashes. Head injuries were the most reduced body region in frontal crashes, decreased by 40%; however, there was an increase in lower extremity injuries. Chest injuries were most reduced in side crashes, decreased by 25%. For rollovers, reduction in injuries for all body regions were observed in 2000 to 2009 model years compared to models up to five years earlier. For the most recent model years, a larger fraction of the AIS 3+ HARM occurred in severe rollover (more than 7 quarter-turns).

Based on the National Automotive Sampling System Crashworthiness Data System since the 1988-1992 model years, there has been a reduction in the MAIS 3+ injury rate and the Mean HARM for all crash modes. The largest improvement in vehicle safety has been in rollovers. There was an increase in the rollover injury rate in the 1993-1998 model year period, but a reduction since then. When comparing vehicles of the model year 1993 to 1998 with later model vehicles, the most profound difference was the reduction of rollover frequency for SUV's - down more than 20% when compared to other crash modes. When considering only model years since 2002 the rollover frequency reduction was nearly 40%. A 26% reduction in the rate of moderate and serious injuries for all drivers in rollovers was observed for the model years later than 1998. The overall belt use rate for drivers of late model vehicles with HARM weighted injuries was 62% - up from 54% in earlier model vehicles. However, in rollover crashes, the same belt use rate lagged at 54%.

Target pre-crash scenarios, crash modes, and occupant injury mechanisms are statistically described for crash imminent braking (CIB) and advanced restraint system (ARS) applications based on pre-crash sensing. Vehicle-object and vehicle- vehicle crashes are distinguished between single- impact and multiple-impact crashes. This analysis focuses on light vehicles of model year 1998 or higher that suffered frontal damage from the first most harmful event. An in-depth examination of candidate crash cases from target crashes was conducted to understand crash mechanisms and circumstances as well as occupant injury scenarios. Consideration was given to pre-crash conditions for CIB applications and to injury source for ARS applications. Results will be used in subsequent research to assess candidate CIB and ARS technologies, develop system functional requirements, devise test procedures, and estimate safety benefits.

Crash scenarios are prioritized for crash-imminent braking and advanced restraint applications aimed at reducing injuries when the crash becomes unavoidable using forward-looking pre-crash sensors. Characteristics of priority crash scenarios will form a basis for the development of performance specifications and test procedures for these two applications. Target crashes involve at least one light vehicle (passenger car, van or minivan, light pickup truck, or sport utility vehicle) of model year 1998 or higher with frontal damage from the first impact. Crash prioritization is based on the number of fatalities and functional years lost. Data sources include the Fatality Analysis Reporting System, General Estimates System, and Crashworthiness Data System. Crash scenarios are broken down by vehicle-object and vehicle-vehicle crashes. Pedestrian, pole, or tree impacts prevail in vehicle-object crashes. Dominant vehicle-vehicle crash scenarios include opposite direction and rear-end pre-crash scenarios with front-to-front and front-to-back impacts, respectively.

Previous research has described the frequency of injuries to body regions suffered by belted, non-ejected occupants of vehicles involved in rollover crashes. While these studies have characterized the crash and occupant outcomes based on contact of the belted occupant with the vehicle interior, these studies have not delved into the specific types of injuries within the body regions. Atkinson et al. (2000) described in detail head and neck injury types resulting from rollover crashes in order to determine what injuries should be predicted in rollover simulations. Ridella and Eigen (2008) investigated injury mechanisms in belted, rollover-involved occupants from the NHTSA’s Crash Injury Research Engineering Network (CIREN) database, but could not project their conclusions to the national level. The aim of this study is to determine distributions of specific injury types in rollover crashes of belted, non-ejected occupants from recent years of the National Automotive Sampling System - Crashworthiness Data System (NASS-CDS) database.

The authors hope to provide an intermediate method of data extraction, taking advantage of the improvements in child passenger data collection and recording. The authors also wish to highlight the importance of appropriate usage of the data. As suggested in the Eigen 2007, the enhanced data set SAS files, also known as the 30-file data set, will be contrasted with the 11-file data set format, the traditionally available NASS CDS SAS files, and analysts will be referred to the NHTSA web site for supplementary information. Further, frequently asked questions will be addressed to provide uniform information dissemination to all users. The primary data source will be the National Automotive Sampling System (NASS) Crashworthiness Data System (CDS). As conclusion, the authors propose a three-step extraction methodology to be used until the enhanced data files can be released. This includes traditional data extraction to retain weighting factors, extraction of the enhanced variables, attributes, and associated graphics, and manually integrating the two data sources. The full text of this paper may be found at: http://www-nrd.nhtsa.dot.gov/pdf/esv/esv21/09-0550.pdf For the covering abstract see ITRD E145407.

This report presents the results of crash analyses that defined and prioritized target crashes for advanced restraint systems based on pre-crash sensors. These analyses targeted the driver and front-seat passenger 13 or older, traveling in light vehicles of model year 1998 or newer that sustained frontal damage. The focus was on occupants who suffered an injury level 3 or higher on the Maximum Abbreviated Injury Scale. Crash analyses included data queries of national crash databases from the 1997-2006 Crashworthiness Data System (CDS), 2006 General Estimates System, and 2002-2006 Fatality Analysis Reporting System. Data queries were part of the top-down analysis that identified key crash scenarios and concomitant injuries. Crash scenarios incorporated pre-crash scenarios, impact modes, and occupant restraint use. Severity of priority crash scenarios was quantified in terms of the number of fatalities and functional years lost. Moreover, injured body regions and related injury levels were expressed for each crash scenario. Based on results of top-down analysis, cases were selected from the CDS for further examination to identify the injury sources and their link to crash scenarios and severities. Only belted occupants were considered. The steering wheel had the highest contribution rate to injury in chest, head, and upper extremity body regions. Injury to the abdomen was caused predominantly by the seat belt. The instrument panel caused the highest rate of injury to the lower extremity.

Target pre-crash scenarios, crash modes, and occupant injury mechanisms are statistically described for crash imminent braking (CIB) and advanced restraint system (ARS) applications based on pre-crash sensing. Vehicle-object and vehicle-vehicle crashes are distinguished between single-impact and multiple-impact crashes. This analysis focuses on light vehicles of model year 1998 or higher that suffered frontal damage from the first most harmful event. An in-depth examination of candidate crash cases from target crashes was conducted to understand crash mechanisms and circumstances as well as occupant injury scenarios. Consideration was given to pre-crash conditions for CIB applications and to injury source for ARS applications. Results will be used in subsequent research to assess candidate CIB and ARS technologies, develop system functional requirements, devise test procedures, and estimate safety benefits. The full text of this paper may be found at: http://www-nrd.nhtsa.dot.gov/pdf/esv/esv21/09-0248.pdf For the covering abstract see ITRD E145407.

In response to evolving sensor and occupant retention technologies, the National Highway Traffic Safety Administration (NHTSA) will soon begin cooperative research to develop test procedures for advanced occupant restraints. It is believed that these restraints will be real-time adaptive to a variety of crash types and severities, as well as address such problems as improving belt effectiveness in front-front crashes to higher than the current 50% fatality reduction level and possibly making the belts and air bags better suited for rollover and offset crashes. The research will address: identification of potential improvements in current restraints, identification of minimum performance and objective testing, as well as performance metrics, and calculation of benefits inherent in such improvements. To complete these tasks, the identification of a target crash population, estimations of the effectiveness of advanced restraints from test and evaluation, and benefits calculation based upon the target population and the effectiveness estimates is necessary. This paper serves as an initial analysis of the advanced restraint system target population. The Crashworthiness Data System (CDS) of the National Automotive Sampling System (NASS) was chosen for the initial work owing to its complete crash, vehicle, occupant, and injury reporting in the U.S. In addition, in order to maintain a focus on recent vehicle designs and performance, the most recent eight years of data were used for an occupant population that contains only belted drivers and passengers. By analyzing this population, attention was focused on the current performance of restraints in order to identify opportunities for restraint improvement. Restrained occupants with Maximum Abbreviated Injury Scale (MAIS) groupings of 3+ (serious injuries and higher) were quantified. Disaggregations of the primary direction of force, impact area, and injury types, among others, were computed across all crash types in order to develop an understanding of the requirements for advanced restraint prototype designs.