Highway accident severities and the mixed logit model: an exploratory empirical analysis.

Many transportation agencies use accident frequencies, and statistical models of accidents frequencies, as a basis for prioritizing highway safety improvements. However, the use of accident severities in safety programming has been often been limited to the locational assessment of accident fatalities, with little or no emphasis being placed on the full severity distribution of accidents (property damage only, possible injury, injury)-which is needed to fully assess the benefits of competing safety-improvement projects. In this paper we demonstrate a modeling approach that can be used to better understand the injury-severity distributions of accidents on highway segments, and the effect that traffic, highway and weather characteristics have on these distributions. The approach we use allows for the possibility that estimated model parameters can vary randomly across roadway segments to account for unobserved effects potentially relating to roadway characteristics, environmental factors, and driver behavior. Using highway-injury data from Washington State, a mixed (random parameters) logit model is estimated. Estimation findings indicate that volume-related variables such as average daily traffic per lane, average daily truck traffic, truck percentage, interchanges per mile and weather effects such as snowfall are best modeled as random-parameters-while roadway characteristics such as the number of horizontal curves, number of grade breaks per mile and pavement friction are best modeled as fixed parameters. Our results show that the mixed logit model has considerable promise as a methodological tool in highway safety programming.

Differences in rural and urban driver-injury severities in accidents involving large-trucks: an exploratory analysis.

This study explores the differences between urban and rural driver injuries (both passenger-vehicle and large-truck driver injuries) in accidents that involve large trucks (in excess of 10,000 pounds). Using 4 years of California accident data, and considering four driver-injury severity categories (no injury, complaint of pain, visible injury, and severe/fatal injury), a multinomial logit analysis of the data was conducted. Significant differences with respect to various risk factors including driver, vehicle, environmental, road geometry and traffic characteristics were found to exist between urban and rural models. For example, in rural accidents involving tractor-trailer combinations, the probability of drivers' injuries being severe/fatal increased about 26% relative to accidents involving single-unit trucks. In urban areas, this same probability increased nearly 700%. In accidents where alcohol or drug use was identified as being the primary cause of the accident, the probability of severe/fatal injury increased roughly 250% percent in rural areas and nearly 800% in urban areas. While many of the same variables were found to be significant in both rural and urban models (although often with quite different impact), there were 13 variables that significantly influenced driver-injury severity in rural but not urban areas, and 17 variables that significantly influenced driver-injury severity in urban but not rural areas. We speculate that the significant differences between rural and urban injury severities may be at least partially attributable to the different perceptual, cognitive and response demands placed on drivers in rural versus urban areas.

The crash severity impacts of fixed roadside objects.

INTRODUCTION: This study analyzes the in-service performance of roadside hardware on the entire urban State Route system in Washington State by developing multivariate statistical models of injury severity in fixed-object crashes using discrete outcome theory. The objective is to provide deeper insight into significant factors that affect crash severities involving fixed roadside objects, through improved statistical efficiency along with disaggregate and multivariate analysis. METHOD: The developed models are multivariate nested logit models of injury severity and they are estimated with statistical efficiency using the method of full information maximum likelihood. RESULTS: The results show that leading ends of guardrails and bridge rails, along with large wooden poles (e.g. trees and utility poles) increase the probability of fatal injury. The face of guardrails is associated with a reduction in the probability of evident injury, and concrete barriers are shown to be associated with a higher probability of lower severities. Other variables included driver characteristics, which showed expected results, validating the model. For example, driving over the speed limit and driving under the influence of alcohol increase the probability of fatal accidents. Drivers that do not use seatbelts are associated with an increase in the probability of more severe injuries, even when an airbag is activated. IMPACT ON INDUSTRY: The presented models show the contribution of guardrail leading ends toward fatal injuries. It is therefore important to use well-designed leading ends and to upgrade badly performing leading ends on guardrails and bridges. The models also indicate the importance of protecting vehicles from crashes with rigid poles and tree stumps, as these are linked with greater severities and fatalities.

Random parameter models of interstate crash frequencies by severity, number of vehicles involved, collision and location type.

A nine-year (1999-2007) continuous panel of crash histories on interstates in Washington State, USA, was used to estimate random parameter negative binomial (RPNB) models for various aggregations of crashes. A total of 21 different models were assessed in terms of four ways to aggregate crashes, by: (a) severity, (b) number of vehicles involved, (c) crash type, and by (d) location characteristics. The models within these aggregations include specifications for all severities (property damage only, possible injury, evident injury, disabling injury, and fatality), number of vehicles involved (one-vehicle to five-or-more-vehicle), crash type (sideswipe, same direction, overturn, head-on, fixed object, rear-end, and other), and location types (urban interchange, rural interchange, urban non-interchange, rural non-interchange). A total of 1153 directional road segments comprising of the seven Washington State interstates were analyzed, yielding statistical models of crash frequency based on 10,377 observations. These results suggest that in general there was a significant improvement in log-likelihood when using RPNB compared to a fixed parameter negative binomial baseline model. Heterogeneity effects are most noticeable for lighting type, road curvature, and traffic volume (ADT). Median lighting or right-side lighting are linked to increased crash frequencies in many models for more than half of the road segments compared to both-sides lighting. Both-sides lighting thereby appears to generally lead to a safety improvement. Traffic volume has a random parameter but the effect is always toward increasing crash frequencies as expected. However that the effect is random shows that the effect of traffic volume on crash frequency is complex and varies by road segment. The number of lanes has a random parameter effect only in the interchange type models. The results show that road segment-specific insights into crash frequency occurrence can lead to improved design policy and project prioritization.

A simultaneous equations model of crash frequency by severity level for freeway sections.

This paper presents a simultaneous equations model of crash frequencies by severity level for freeway sections using five-year crash severity frequency data for 275 multilane freeway segments in the State of Washington. Crash severity is a subject of much interest in the context of freeway safety due to higher speeds of travel on freeways and the desire of transportation professionals to implement measures that could potentially reduce crash severity on such facilities. This paper applies a joint Poisson regression model with multivariate normal heterogeneities using the method of Maximum Simulated Likelihood Estimation (MSLE). MSLE serves as a computationally viable alternative to the Bayesian approach that has been adopted in the literature for estimating multivariate simultaneous equations models of crash frequencies. The empirical results presented in this paper suggest the presence of statistically significant error correlations across crash frequencies by severity level. The significant error correlations point to the presence of common unobserved factors related to driver behavior and roadway, traffic and environmental characteristics that influence crash frequencies of different severity levels. It is found that the joint Poisson regression model can improve the efficiency of most model coefficient estimators by reducing their standard deviations. In addition, the empirical results show that observed factors generally do not have the same impact on crash frequencies at different levels of severity.

A multivariate tobit analysis of highway accident-injury-severity rates.

Relatively recent research has illustrated the potential that tobit regression has in studying factors that affect vehicle accident rates (accidents per distance traveled) on specific roadway segments. Tobit regression has been used because accident rates on specific roadway segments are continuous data that are left-censored at zero (they are censored because accidents may not be observed on all roadway segments during the period over which data are collected). This censoring may arise from a number of sources, one of which being the possibility that less severe crashes may be under-reported and thus may be less likely to appear in crash databases. Traditional tobit-regression analyses have dealt with the overall accident rate (all crashes regardless of injury severity), so the issue of censoring by the severity of crashes has not been addressed. However, a tobit-regression approach that considers accident rates by injury-severity level, such as the rate of no-injury, possible injury and injury accidents per distance traveled (as opposed to all accidents regardless of injury-severity), can potentially provide new insights, and address the possibility that censoring may vary by crash-injury severity. Using five-year data from highways in Washington State, this paper estimates a multivariate tobit model of accident-injury-severity rates that addresses the possibility of differential censoring across injury-severity levels, while also accounting for the possible contemporaneous error correlation resulting from commonly shared unobserved characteristics across roadway segments. The empirical results show that the multivariate tobit model outperforms its univariate counterpart, is practically equivalent to the multivariate negative binomial model, and has the potential to provide a fuller understanding of the factors determining accident-injury-severity rates on specific roadway segments.

A study of factors affecting highway accident rates using the random-parameters tobit model.

A large body of previous literature has used a variety of count-data modeling techniques to study factors that affect the frequency of highway accidents over some time period on roadway segments of a specified length. An alternative approach to this problem views vehicle accident rates (accidents per mile driven) directly instead of their frequencies. Viewing the problem as continuous data instead of count data creates a problem in that roadway segments that do not have any observed accidents over the identified time period create continuous data that are left-censored at zero. Past research has appropriately applied a tobit regression model to address this censoring problem, but this research has been limited in accounting for unobserved heterogeneity because it has been assumed that the parameter estimates are fixed over roadway-segment observations. Using 9-year data from urban interstates in Indiana, this paper employs a random-parameters tobit regression to account for unobserved heterogeneity in the study of motor-vehicle accident rates. The empirical results show that the random-parameters tobit model outperforms its fixed-parameters counterpart and has the potential to provide a fuller understanding of the factors determining accident rates on specific roadway segments.