Luis F. Miranda-MorenoLiping FuFrank F. SaccomannoAurelie Labbe...
p.1-8页
查看更多>>摘要：Many types of statistical models have been proposed for estimating accident risk in transport networks, ranging from basic Poisson and negative binomial models to more complicated models, such as zero-inflated and hierarchical Bayesian models. However, little systematic effort has been devoted to comparing the performance and practical implications of these models and ranking criteria when they are used for identifying hazardous locations. This research investigates the relative performance of three alternative models: the traditional negative binomial model, the heterogeneous negative binomial model, and the Poisson lognormal model. In particular, this work focuses on the impact of the choice of two alternative prior distributions (i.e., gamma versus lognormal) and the effect of allowing variability in the dispersion parameter on the outcome of the analysis. From each model, two alternative accident estimators are computed by using the conditional mean under both marginal and posterior distributions. A sample of Canadian highway-railway intersections with an accident history of 5 years is used to calibrate and evaluate the three alternative models and the two ranking criteria. It is concluded that the choice of model assumptions and ranking criteria can lead to considerably different lists of hazardous locations.
查看更多>>摘要：A linear optimization model that maximizes the safety benefits of improvements on an existing highway within specific budget constraints is presented. This model works in conjunction with crash prediction models that predict the expected number of crashes for highways by using a base model and productive accident modification factors (AMFs). The base model predicts the expected number of crashes for a base highway. The AMFs modify the prediction on the basis of the actual characteristics of the highway. If there were multiple alternative improvements for several highway features, then there would be many combined alternatives, each with a certain cost and a certain degree of improvement in highway safety. The proposed model uses linear optimization to find the combined alternative that has the best safety improvement within the available budget. The mathematical optimization model is studied in a general form as well as a detailed form in the context of an existing crash prediction model, the crash prediction module of the interactive highway safety design model. The procedure for building the optimization problem is described. A C programming code was developed to build the linear optimization problem. A test case study is defined, and the problem is built and solved by using the CPLEX optimization solver. The variation of the safety measures versus improvement costs is studied, and the results are discussed.
查看更多>>摘要：Comparing samples from two or more populations is among the most common statistical tasks that engineers and scientists perform. Typically a t-test or one-way or two-way analysis of variance is used to compare the means of different populations. Although these tests are useful for describing differences in means for various populations, they are of limited use for comparison of other population parameters, such as quantiles (i.e., percentiles). Quantile comparisons are useful for examining the changes that occur to portions of the population away from the median. Comparison of speed percentiles is especially important to the traffic engineering profession to determine the effect of various treatments on speeds of faster drivers (e.g., 85th percentile speeds). Simply testing the mean speed values does not indicate whether the differences occur because of slower traffic driving more slowly, faster traffic driving more slowly, or both. Furthermore, are positive effects on faster drivers being masked by contrasting effects on slower drivers? The main obstacle to providing confidence intervals and tests for quantiles is calculating reasonable estimates of variances for the sample quantiles that are far from the median. This paper describes a nonparametric double bootstrapping procedure for direct comparison of quantiles of two or more sample populations. The first bootstrap simulation is used to produce estimates of standard errors for the desired quantiles and thereby overcome the inability to make reasonable variance estimations. The second layer of bootstrap simulations is used to determine the threshold cutoff values based on a desired level of confidence for the test of hypothesis. The cutoff values also may be used to form confidence intervals. The steps of the procedure along with an example of its use are provided.
查看更多>>摘要：Research into the application of freeway loop detector data for traffic safety has gained momentum in recent years. The incompleteness of data from loop detectors has been a common problem in both the development and the implementation of models. The effect of individual crash precursors, obtained one at a time from a series of loop detectors, on relative risk of crash occurrence was examined through within-stratum one-covariate logistic regression models. The hazard ratio (resultant change in log odds of observing a crash by changing the covariate by one unit) was used as the measure of risk. The log of coefficient of variation in speed expressed as percentage, standard deviation of volume, and average occupancy expressed as percentage were found to be the most significant individual covariates affecting the odds of crash occurrence at a crash site. It was also observed that these parameters calculated at a 5-min level (as opposed to a 3-min level) are more significantly associated with crash occurrence. Hazard ratios corresponding to these covariates observed at a series of stations during six 5-min slices were plotted as a contour variable. The location and time of measurements of these parameters with respect to the location and time of the crash were used as ordinate and abscissa, respectively, in the contour plots depicting spatiotemporal variation of crash risk. The chart corresponding to the log of coefficient of variation in speed demonstrated the most clear patterns of increasing risk as the time and location of the crash are approached. On the basis of these spatiotemporal patterns, a methodology with which to identify freeway black spots in real time is proposed. This information could be used by traffic management centers to take preventive measures to avoid crashes or to prepare law enforcement and emergency vehicles for the impending situation.
查看更多>>摘要：Many studies have shown that intersections are among the most dangerous locations of a roadway network. Therefore, there is a need to understand the factors that contribute to traffic crashes at such locations. One approach is to classify intersections and quantify the effects that configuration, geometric characteristics, and traffic volume have on the number of crashes at signalized intersections. This paper addresses the different factors that affect crashes, by type of collision, at signalized intersections. It also looks into the quality and completeness of the crash data and the effect that incomplete data have on the final results. Data from multiple sources were cross-checked to ensure the completeness of all crashes, including minor crashes that were usually unreported or were not coded into crash databases. The tree-based regression methodology was adopted in this study to cope with multicollinearity between variables, missing observations, and the fact that the true model form was unknown. The results showed a significant discrepancy in the factors that were found to affect the different collision types and their influence in each model. The two most significant differences in comparison with the total crash model as a base case were found to be in the models of head-on and left-turn crashes. The results also showed that the important factors were relatively consistent for rear-end, right-turn, and sideswipe crashes when minor crashes were considered. However, angle and head-on crashes showed significant changes in the model structure when minor crashes were added to the data set because these types of crashes were less stable. Finally, different roadway characteristics were correlated with different types of crashes.
查看更多>>摘要：Before effective remedial treatments can be implemented at hazardous intersections, it is often necessary to identify the causal factors affecting accident frequency. However, a problem often encountered in safety studies is the underreporting of accidents. This biased reporting may affect the selection of the parsimonious model. This study analyzes the factors affecting road accident frequency at three-legged signalized intersection approaches in Singapore, with special emphasis to underreporting. The annual accidents at 104 three-legged signalized intersections are modeled as the sum of observed Poisson outcome of accident reports. The reporting mechanism is introduced as a probit equation. The model shows that several geometric, traffic, and traffic control factors significantly affected the accident occurrence. The total approach volumes, left-turn volumes, existence of unprotected left-turn slip roads, number of signal phases per cycle, use of permissive right-turning phase, and sites with sight distances less than 100 m or greater than 300 m appear to increase accident occurrence. However, the presence of right-turn channelization, provision of an acceleration section for left turning, existence of a surveillance camera, availability of median railings, and presence of an approach gradient greater than +5% may reduce the occurrence of intersection approach accidents. Moreover, the study shows that the reporting rate may drop because current law enforcement requires that only injury accidents and accidents at intersections in a residential area be reported.
查看更多>>摘要：The future of traffic management and highway safety lies in proactive traffic management systems. Crash prediction models that use real-time traffic flow variables measured through a series of loop detectors are the most important component of such systems. A previous crash prediction model was developed with the matched case-control logistic regression technique. Although the model achieved reasonable classification accuracy, it remained open to improvement because of the limited study area, sample size, and transferability issues. Therefore, the previous work had been extended. Multivehicle freeway crashes under high- and low-speed traffic conditions were found to differ in severity and in their mechanism. The distribution of 5-min average speeds obtained immediately before the crash from the loop detector station closest to the crash shows two approximate mound-shaped distributions. This distribution is used as the basis to separate the models for crashes occurring under the two speed conditions. The results show that, as expected, variables that entered in the final models (for crashes under high and low speeds) were not the same. However, they were found to be consistent with the probable mechanisms of crashes under the respective speed conditions. A possible implementation of the separate models with the use of the odds ratios and with the balancing of the threshold between achieving high classification of crash potential and the false alarm situation is presented.
查看更多>>摘要：This paper presents a study evaluating the power model of the relationship between speed and road safety. The power model states that a given relative change in the mean speed of traffic is associated with a relative change in the number of accidents or accident victims by means of a power function. An extensive review of relevant literature has been made, and evidence from 98 studies containing 460 estimates of the relationship between changes in speed and changes in the number of accidents or accident victims has been synthesized by means of meta-analysis. The results are broadly supportive of the power model. It is concluded that speed has a major impact on the number of accidents and the severity of injuries and that the relationship between speed and road safety is causal, not just statistical.
查看更多>>摘要：According to recent national statistics, red light running crashes represent a significant safety problem at signalized intersections. To examine the overall characteristics of red light running crashes, this study used the 1999 to 2001 Florida crash database to investigate the crash propensity related to traffic environments, driver characteristics, and vehicle types. The quasi-induced exposure concept and multiple logistic regression technique were used to perform this analysis. The results showed that traffic factors including number of lanes, crash time, weather, highway character, day of week, urban or rural location, speed limit, driver age, alcohol or drug use, physical defect, driver residence, and vehicle type were significantly associated with the risk of red light running crashes. Furthermore, it confirmed that there were significant interaction effects between the risk factors, including crash time and highway character, number of lanes and urban or rural location, weather condition and driver age, driver age and gender, alcohol or drug use and gender, and type of vehicle and gender.
查看更多>>摘要：This paper examines risks associated with peak period lane closure during construction or maintenance work on urban freeways. In accordance with recently implemented policy by the Colorado Department of Transportation, lane closure would be allowed if reserve capacity were available. A relatively minor accident in the work zone caused substantial delays during the peak period that virtually paralyzed traffic in the Denver, Colorado, metropolitan area. This occurrence caused reexam-ination of the existing lane closure policy. Generally speaking, if a contractor is allowed greater flexibility in establishing work schedules, including the ability to work through peak periods, a lower bid can be expected. This paper compares savings in the cost of construction related to allowing lane closure during peak periods with the cost of potential incident-related delays in the framework of a quantitative risk analysis.
NETL
NSTL