Showing papers by "Dominique Lord published in 2015"

Validation of Crash Modification Factors Derived from Cross-Sectional Studies with Regression Models

Abstract: Crash modification factors (CMFs) can be used to capture the safety effects of countermeasures and play a significant role in traffic safety management. As an alternative to the before-and-after study, the regression model method has been widely used for estimating CMFs. Although before-and-after studies are considered to be superior, the use of regression models for estimating CMFs has never been fully investigated. Consequently, the conditions in which regression models could be used for such a purpose were examined. CMFs for three variables—lane width, curve density, and pavement friction—were assumed and used for generating random crash counts. Then CMFs were derived from regression models by using the simulated crash data for three different scenarios. The results were then compared with the assumed true values. The study results showed that (a) when all factors affecting traffic safety are identical in all segments except those of interest, CMFs derived from regression models should be unbiased; (b)...

Michigan urban trunkline intersections safety performance functions (SPFs) development and support.

Abstract: This study involves the development of safety performance functions (SPFs) for signalized and stop-controlled intersections located along urban and suburban arterials in the state of Michigan. Extensive databases were developed that resulted in the integration of traffic crash information, traffic volumes, and roadway geometry information. After these data were assembled, an exploratory analysis of the data was conducted to identify general crash trends. This included assessment of the base models provided in the Highway Safety Manual (HSM), as well as a calibration exercise, which demonstrated significant variability in terms of the goodness-of-fit of the HSM models across various site types. Michigan-specific SPFs were estimated, including simple models that consider only annual average daily traffic (AADT). More detailed models were also developed, which considered additional geometric factors, such as posted speed limits, number of lanes, and the presence of medians, intersection lighting, and right-turn-on-red prohibition. Crash modification factors (CMFs) were also estimated, which can be used to adjust the SPFs to account for differences related to these factors. Separate SPFs were estimated for intersections of only two-way streets and for those where at least one of the intersecting streets was one-way as the factors affecting traffic safety were found to vary between these site types. Severity distribution functions (SDFs) were also estimated, which can be used to predict the proportion of injury crashes which result in different injury severity levels. The SDFs may include various geometric, operation, and traffic variables that will allow the estimated proportion to be specific to an individual intersection. Ultimately, the results of this study provide Michigan Department of Transportation (MDOT) with a number of methodological tools that will allow for proactive safety planning activities, including network screening and identification of high-risk sites. These tools have been calibrated such that they can be applied at either the statewide level or within any of MDOT’s seven geographic regions, providing additional flexibility to accommodate unique differences across the state. The report also documents procedures for maintaining and calibrating these SPFs over time, allowing for consideration of general trends that are not directly reflected by the predictor variables.

Application of the Hyper-Poisson Generalized Linear Model for Analyzing Motor Vehicle Crashes.

Abstract: The hyper-Poisson distribution can handle both over- and underdispersion, and its generalized linear model formulation allows the dispersion of the distribution to be observation-specific and dependent on model covariates. This study's objective is to examine the potential applicability of a newly proposed generalized linear model framework for the hyper-Poisson distribution in analyzing motor vehicle crash count data. The hyper-Poisson generalized linear model was first fitted to intersection crash data from Toronto, characterized by overdispersion, and then to crash data from railway-highway crossings in Korea, characterized by underdispersion. The results of this study are promising. When fitted to the Toronto data set, the goodness-of-fit measures indicated that the hyper-Poisson model with a variable dispersion parameter provided a statistical fit as good as the traditional negative binomial model. The hyper-Poisson model was also successful in handling the underdispersed data from Korea; the model performed as well as the gamma probability model and the Conway-Maxwell-Poisson model previously developed for the same data set. The advantages of the hyper-Poisson model studied in this article are noteworthy. Unlike the negative binomial model, which has difficulties in handling underdispersed data, the hyper-Poisson model can handle both over- and underdispersed crash data. Although not a major issue for the Conway-Maxwell-Poisson model, the effect of each variable on the expected mean of crashes is easily interpretable in the case of this new model.

Analysis of Injury Severity in Pedestrian Crashes Using Classification Regression Trees

Abstract: Texas is considered to be an “opportunity” state by the Federal Highway Administration (FHWA), due to the high number of pedestrian crashes. Data from the Fatality Analysis Reporting System (FARS) show that the number of pedestrian fatal crashes in Texas is the third highest in the U.S and is significantly higher than the national average. The research team explored the Texas Department of Transportation (TxDOT) Crash Record Information System (CRIS) database to identify characteristics of crashes involving pedestrians in Texas. A Classification and Regression Tree (CRT) analysis of all pedestrian crashes was conducted to find the significant factors influencing the severity of crashes involving pedestrians in Texas. The classification tree identified that light condition, road class, traffic control, right shoulder width, involvement of a commercial vehicle, pedestrian age, and the collision manner, have the most influence on the severity of pedestrian crashes.

Pedestrian fatal crashes on freeways in texas

Abstract: Over the five-year period of 2007 through 2011, 2,232 fatal pedestrian crashes were recorded in Texas. 21% of these crashes were found to have occurred on controlled-access facilities (i.e. freeways). This is an alarmingly high number for a location where pedestrians are least expected. This study analyzed crash reports and police officer narratives to understand the characteristics and contributing factors associated with fatal pedestrian crashes on freeways. The contributing factors identified include pedestrian and driver alcohol use and dark conditions. Eighty percent of the crashes occurred after dark, almost half of which were at a location with no lighting. Intoxicated pedestrians were involved in twenty eight percent of crashes, with an average BAC of 0.20. To alleviate this problem, there may be a need for conducting a “Don’t Drink and Walk” campaign to educate the general public of dangers of walking while intoxicated, especially at night. A quarter of the crashes involved unintended pedestrians, i.e. those who were out of their vehicle due to a previous crash or a stalled vehicle. Motorists should be educated to not work on their vehicle in traffic and not to try and cross the freeway to reach a shoulder or median. It is best to wait in the vehicle with seat belt and hazard lights on until emergency services arrive.