Showing papers by "Dominique Lord published in 2003"

Modeling traffic crash-flow relationships for intersections: dispersion parameter, functional form, and bayes versus empirical bayes methods

Abstract: Statistical relationships between traffic crashes and traffic flows at roadway intersections have been extensively modeled and evaluated in recent years. The underlying assumptions adopted in the popular models for intersections are challenged. First, the assumption that the dispersion parameter is a fixed parameter across sites and time periods is challenged. Second, the mathematical limitations of some functional forms used in these models, particularly their properties at the boundaries, are examined. It is also demonstrated that, for a given data set, a large number of plausible functional forms with almost the same overall statistical goodness of fit (GOF) is possible, and an alternative class of logical formulations that may enable a richer interpretation of the data is introduced. A comparison of site estimates from the empirical Bayes and full Bayes methods is also presented. All discussions and comparisons are illustrated with a set of data collected for an urban four-legged signalized intersection in Toronto, Ontario, Canada, from 1990 to 1995. In discussing functional forms, the need for some goodness-of-logic measures, in addition to the GOF measure, is emphasized and demonstrated. Finally, analysts are advised to be mindful of the underlying assumptions adopted in the popular models, especially the assumption that the dispersion parameter is a fixed parameter, and the limitations of the functional forms used. Promising directions in which this study may be extended are also discussed.

Development of a Procedure for Estimating Expected Safety Effects of a Contemplated Traffic Signal Installation

Abstract: The Manual on Uniform Traffic Control Devices contains warrants for traffic signal installation but cautions that satisfying a warrant does not in itself justify the decision to install a signal and that one should not be installed unless an engineering study indicates that this will improve the overall safety or operation of the intersection. The development of an easily implementable procedure, which is intended to be part of an engineering study, for estimation of the expected safety effects of a contemplated signal installation is reported. These effects can then be considered in conjunction with other impacts in a conventional economic evaluation. The development of the procedure by use of a multijurisdiction database is described, and a detailed illustration is presented. Use is made of the empirical Bayes methodology that of late has been recognized as the state of the art in safety estimation and of the most recent advances in that methodology. Substantial focus is placed on the application of that methodology and on the development of the accident prediction models required to support that application. The development of the procedure is part of an NCHRP project (NCHRP 17-16) that aims to improve the safety warrant for signal installation and, more generally, to determine how safety is considered in the decision to install or not install a signal.

Parameter, functional form, and bayes versus empirical bayes

Abstract: Statistical relationships between traffic crash and traffic flows at roadway intersections have been extensively modeled and evaluated in recent years. This paper challenges the underlying assumptions adopted in the popular traffic crash estimation models for intersections. More specifically, we first challenge the assumption that the dispersion parameter is a fixed parameter across sites and time period. Second, we examine mathematical limitations of some functional forms used by these models, particularly their properties at the boundaries. We also demonstrate that, for a given data set, a large number of plausible functional forms with almost the same overall statistical quality are possible, and discuss an alternative class of logical formulations that may enable a richer interpretation of the data. A comparison of site estimates from the Empirical Bayes and full-Bayes methods is also presented. All discussion and demonstrations are motivated with an urban 4-legged signalized intersection data set collected in Toronto, Canada, for years 1990 to 1995. Finally, we advise the analysts to be mindful of the underlying assumptions adopted in the popular models, especially the assumption that dispersion parameter is a fixed parameter and limitations of the functional forms used. We conclude the paper by discussing several promising directions in which this study may be extended.

Synthesis on the Safety of Right Turn on Red in the United States and Canada

Abstract: The Government of Quebec is currently considering the introduction of right turn on red (RTOR) at signalized intersections in Quebec, with the possible exception of the City of Montreal. The Province of Quebec and New York City are the only jurisdictions in North America where the RTOR is not permitted. Following more than 25 years of intense political debates, the Government agreed in 2000 to commission a policy study, aimed at either adopting or rejecting the RTOR once and for all. In order to fully understand all the issues surrounding the introduction of such a measure, the Ministry of Transportation of Quebec (MTQ) mandated the author of this paper to perform a study on the safety, traffic operational improvements, and other characteristics resulting from RTOR based on information gathered from various agencies in the United States and Canada. The study aimed at collecting crash statistics; gathering and reviewing recent research documents and papers; and surveying the opinions of transportation experts and researchers on this topic. The outcome of this study shows that the RTOR is not a dangerous maneuver at signalized intersections for either vehicles or pedestrians in most circumstances. The people interviewed in this study have corroborated this outcome. In short, the proportion of RTOR crashes is usually very low (less than 0.5% of all crashes in a given jurisdiction) and, in the event of crash, the outcome is generally not severe. Based on the respondents’ comments, it is obvious that many transportation professionals do not consider the RTOR to be a safety problem. The author addresses several issues and unanswered questions on the safety and traffic operational enhancements resulting from RTOR. The final decision as to whether or not introduce the RTOR in Quebec resides with government officials and should be based on the characteristics presented in this research, but should also consider the habits of Quebec drivers and the crossing behavior of pedestrians at signalized intersections.

Traffic Safety Diagnostics and Application of Countermeasures for Rural Roads in Burkina Faso

Abstract: The government of Burkina Faso has recently been making important macroeconomic changes to encourage the economic growth of the country. To maintain this growth, the government has implemented a transportation program to improve road network efficiency and safety. A 2000 study to improve the safety of rural roads in Burkina Faso is described. The primary objectives were to assess traffic safety problems and propose countermeasures to reduce the number and severity of collisions on rural roads. Many rural roads were evaluated on site; all accident data and important socioeconomic variables were collected; and key staff members from various governmental and private agencies were interviewed. The study has shown that traffic safety problems in Burkina Faso are multidimensional, involving inefficient traffic safety management and policy, inadequate road networks, untrained drivers, and defective vehicles. Several traffic safety countermeasures have been proposed for immediate, short-, and long-term application. The most important countermeasures are to create a new institutional framework for improving traffic safety management and train the key personnel responsible for implementing these countermeasures. For the short term, the countermeasures mainly relate to roadway infrastructure improvements and better enforcement tools. For the long term, the countermeasures include a review of current highway traffic laws and their application, evaluation of existing countermeasures, and driver training improvement.

Physical and environmental characteristics of signalized intersections and pedestrian behavior

Abstract: This paper aims to determine some of the physical and environmental characteristics of signalized intersections (SIs) and pedestrian behavior. The dependent variable was defined as whether or not the pedestrian obeyed the green light indicator. The relationship between pedestrian behavior and physical and environmental characteristics of SIs was determined by logistic regression. Results show that some physical and environmental characteristics at SIs such as refuge islands improve pedestrian behavior, whereas a wider street is associated with an increase in the disobedience of the traffic light. The analysis performed in this research was unable to find if the difference in pedestrian behavior between Montreal and Toronto can solely be attributed to the physical environment of SIs.