Improving transportation safety remains the U.S. Department of Transportation's (USDOT's) top priority. The overarching objective is to reduce transportation-related fatalities and injuries by addressing driving behavior and vehicle-related and infrastructure safety issues. USDOT uses a data-driven approach to identify risk factors and develop countermeasures and assess their effectiveness.
The United States and much of the world have made considerable progress in improving safety across all modes of transport— strides made possible through technological advances such as more effective safety belts, regulatory actions such as vehicle safety standards, effective law enforcement, and public outreach. Despite the progress, transportation, including highways and the other modes, accounts for about one-third of the accidental deaths in the United States and is the leading cause of death for people between the ages of 5 and 24 [USHHS CDC 2012].
Transportation fatalities in 2011 were 34,388, a decline of 22.5% over 2000, while transportation injuries were 2,237,029 after a decline of 30.5%. This is in contrast to the preceding 1990-2000 period when fatalities declined 6.3% and injuries declined 1.6%.
Transportation Fatalities: 1990–2011
Transportation Injuries: 1990–2011
These decreases in the number of fatalities and injuries were observed despite U.S. Census data that show a 24.9% increase in the U.S. population—from 249 million in 1990 to nearly 312 million in 2011 [USDOT BTS 2013].
The majority of transportation fatalities and injuries occurred on the Nation's highways, which carry most of the passenger and freight traffic in the United States. Even though 2011 was the safest year on the highways since 1949 in terms of the number of traffic fatalities [USDOT NHTSA 2011], on average 89 people died and over 6,074 per day were injured on the Nation's highways.
While new and emerging technologies like vehicle-to-vehicle communications and next-generation air traffic control systems offer great promise, no solution would be complete without high-quality data and robust statistical analysis. Recognizing the importance of data and statistical analysis in improving transportation safety, Volume 10 of the Journal of Transportation and Statistics is focused on safety research. This Special Issue features six compelling studies that explore the frontier of applied statistical analysis and modeling to offer potentially life-saving insights for both researchers and policymakers. The authors are from some of the leading transportation research institutes in the world, and their exemplary work is indicative of the scope and depth of their expertise. Furthermore, the papers featured in this issue are the product of a multidisciplinary approach to transportation research—in a world where the lines between academic fields, industries, and business sectors are becoming less defined, such a perspective is crucial.
Transportation Fatalities by Mode: 2011
Research published in this Special Issue ranges from assessing crash-risk in roadway corridors where the absence of crash-related incidents has skewed the perception of danger, but not the tangible threat to a comprehensive analysis of the National Highway Traffic Safety Administration's Fatality Analysis Reporting System (FARS) data that sought to uncover previously unseen correlations between the different types of crash-related deaths and the factors that led to those deaths. In all of these studies, the underlying current driving their research is the idea that somewhere in the ever-expanding sea of data are answers capable of saving lives.
This special issue of the Journal includes six papers:
Modeling School Bus Crashes Using Zero-Inflated Model
When a school bus crashes, it is almost always breaking news. While motor-vehicle crashes during the morning or evening commute are a relatively common occurrence across much of the country, school bus crashes are rare events, and when children are injured or worse, it can be devastating for a community. This study explores the potential of the zero-inflated negative binomial (ZINB) model to shed light on previously unknown risk factors that could threaten the safe transport of children on specific segments of the roadway.
Crash Injuries in Four Midwestern States: Comparison to Regional Estimates
This study looks into factors that contribute to the most deadly motor-vehicle crashes in Iowa, Kansas, Missouri, and Nebraska, and why the magnitude of outcomes associated with factors such as adverse weather or seatbelt use in these four states varies so greatly from previous regional estimates. The findings raise questions about current methodologies used to guide new safety measures as well as the absence of a standard framework for crash reporting.
Investigation of the Impact of Corner Clearance on Urban Intersection Crash Occurrence
Signalized intersections contain numerous crash risk factors that have been subject to extensive study. However, there has been little research on corner clearance—the distance between a corner of two intersecting roads and the first driveway—which poses a unique safety risk to drivers exiting from such driveways. This study analyzed crash count data collected from all major, signalized intersections in Las Vegas and North Las Vegas, Nevada, to determine how corner clearance impacts roadway safety. The results provide several key findings that could support future measures to reduce risks associated with corner clearance.
Application of the Bayesian Model Averaging in Predicting Motor Vehicle Crashes
Reliable statistical models underpin the validity of roadway safety research. Typically, analysts will apply multiple models during a study and then apply the one that provides the single "best fit" for the relevant data. This methodology is inherently limited because it does not incorporate the uncertainties presented by the disparate models. In this study, the authors explore the efficacy of applying Bayesian Model Averaging to account for this problem.
Lane Width Crash Modification Factors for Curb-and-Gutter Asymmetric Multilane Roadways: Statistical Modeling
This study is the result of an analysis of crash frequency on multilane, urban roadways and the possible correlation of asymmetrical lanes to both frequency and severity. Asymmetric lanes occur when then the outside lane is wider than the inside lane. The authors' conclusions point to simple changes in roadway design which could reduce the number and severity of crashes along corridors identified as being at-risk.
A Multidimensional Clustering Algorithm for Studying Fatal Road Crashes
Building on existing research on correlative relationships linking fatal crash factors, this study applies a specialized theoretical method, called "graph-cuts," to analyze all fatal car crashes occurring in the prior 2-, 5-, and 10-year spans. This approach searches for clusters that indicate subtle correlations that emerge in a comparative analysis of the historical crashes to the 84 enumerated parameters that can describe a fatal crash event. Using this method, the authors found strong correlations between certain parameters that had not been reported in prior studies.
U.S. Department of Health and Human Services. Center for Disease Control. Deaths: Preliminary Data for 2010. National Vital Statistics Reports Vol. 60, No. 4, Jan. 11, 2012. Available at www.cdc.gov.
U.S. Department of Transportation. Bureau of Transportation Statistics. 2013. The American Landscape. Pocket Guide to Transportation. Available at https://www.bts.dot.gov/bts/publications.
U.S. Department of Transportation. National Highway Traffic Safety Administration. 2011. DOT Estimates Three Percent Drop Beneath 2009 Record Low. Press Release. April. Available at www.nhtsa.gov.