Lina Kattan

A Multinomial Logit Model of Pedestrian–Vehicle Crash Severity

ABSTRACT This article estimated a multinomial logit model to identify the factors determining the severity of pedestrian–vehicle crashes in South Korea. Our results showed that relative to minor crashes, fatal and serious crashes were associated with collisions involving heavy vehicles; drivers who were drunk, male or under the age of 65; pedestrians who were over the age of 65 or female; and pedestrians who were hit in the middle of the road, on high speed roads, in inclement weather conditions, at night, on road links, in tunnels, on bridges, or on wider roads.

Factors contributing to hit-and-run in fatal crashes

Hit-and-run in a collision is a punishable offence as it delays crash notification thereby delaying emergency response which increases the likelihood of traffic fatality. To obtain a better understanding of hit-and-run behavior, a logistic regression model has been applied in this study to identify the factors that might affect the occurrence of hit-and-run in fatal crashes in California, USA. Our results show that roadway functional class, routes, traffic flow, types of roadway section, speed limit, traffic control device, functioning of traffic control device, lighting condition, roadway alignment and roadway profile are important determinants that engineers can target to reduce hit-and-run in fatal crashes. In addition, targeted traffic enforcement should be performed on weekends and nighttime.

Bus rapid transit – a review

Bus Rapid Transit (BRT) has been demonstrated to provide efficient and effective public transport that can even increase transit ridership and attractiveness within defined urban corridors in both developing and developed countries. The merit of the BRT system lies in its ability to provide a high quality public transit service with limited infrastructure and at a relatively low capital and operating cost. In this paper an attempt is made to review BRT characteristics in some detail. An overview of the elements of BRT systems: the running ways, vehicles, stations, as well as operational control, fare collection and passenger information systems is presented. This paper also discusses how these elements allows BRT services to achieve the speed, capacity, reliability, accessibility and operational safety characteristics that distinguishes such systems from regular express bus services. Each built BRT system is unique and depends on how these basic elements are designed and integrated. Environmental, social and economic impacts, users’ perception and modal shift towards BRT have also been critically reviewed. With the presence of local land use and transport policies that support high-density and mixed developments, BRT has been found to attract potential developments around stations and along its corridor, increasing land values and promoting sustainable development. As reported in the literature, the infrastructure costs vary greatly depending on the costs of land acquisition, station design, degree of separation from traffic, technological features and material, labor costs. Thus, BRT capital costs cannot be generalized and must be transferred carefully based on the changes of the input costs that vary from one place to another. The barriers to BRT that block the achievement of significant success in customer satisfaction and ridership gains are mentioned with specific case studies. The review of BRT indicates that in some cases it has similar performance to higher order public transit; however, it still has its unique characteristics that set it apart.

Evacuation behaviour of households and drivers during a tornado

Abstract With increasing concerns over the possibility of tornadoes in highly populated areas in Canada, emergency managers are looking into ways to mitigate the impacts of tornadoes. Given that tornadoes can cause enormous destruction, early warnings and proper evacuation actions are critically important in helping save lives. In this paper, a survey was conducted to analyse the evacuation behaviour of households and drivers during a hypothetical tornado warning situation in the city of Calgary, Alberta. Nearly 500 Calgarians took part in the online survey and provided information on how they would respond to tornado warnings after receiving them. This paper presents the results of the survey. Using probit models, the factors influencing these evacuation decisions are identified and discussed in detail. The results of the household evacuation model show the importance of improving awareness about the safest locations during a tornado. It further highlights the need for targeting the population under the age of 30, who are more likely to take unsafe evacuation actions. The model for evacuation of drivers shows that several factors, such as knowing the difference between a watch and a warning, awareness of safe cover, receipt of warnings through natural environmental cues and the level of education, trigger evacuation actions in avoiding a tornado threat.

Noniterative Approach to Dynamic Traffic Origin-Destination Estimation with Parallel Evolutionary Algorithms

This study focuses on updating time-varying demand matrices by using real observation counts from advanced traffic management surveillance systems. A machine-learning technique using advanced evolutionary algorithms (EAs) is developed instead of the more conventional approaches in the literature. This EA-based demand estimation framework is implemented into a model called the Dynamic Origin-Destination (O-D) Estimator (DynODE). The potential of EAs in the dynamic O-D estimation problem lies in their powerful global search and optimization capabilities. DynODE is integrated with an existing dynamic traffic assignment platform (e.g., DYNASMART-P). The EA-based methods in this study are further augmented with EA parallelization to improve the quality and efficiency of the solution. DynODE mainly addresses offline O-D estimation problems. However, online O-D estimation can be achieved with the parallel version of DynODE with sufficient multiprocessing and parallel computing. The developed approach is rigorous...

Managing speed at school and playground zones

Since speeding is one of the major causes of frequent and severe traffic accidents around school and playground areas, many jurisdictions have reduced the speed limits in these areas to protect children who may be at risk. This paper investigated the speed compliance, mean speed and 85th percentile speed at selected school and playground zones in the City of Calgary in Alberta. Our results showed that the mean speed was lower and the rate of compliance was higher in the school zone compared to the playground zone, 2 lane roads relative to 4 lane roads, roads with fencing, traffic control devices and the presence of speed display device or children, and zones that were longer (> 200 m). Accordingly, this study provided recommendations to improve the effectiveness of school and playground zone speed limits.

Variable speed limit: an overview

This paper provides an overview of various variable speed limit (VSL) strategies developed over the last two decades. It first explains the theoretical background of VSL systems and how they can be used as dynamic traffic control devices to improve traffic conditions and increase safety. An overview of VSL control strategies, ranging from early rule-based approaches to the most advanced network-wide coordinated approach, is provided, followed by a discussion of their associated potential benefits and areas for improvement. The likely environmental benefits of VSL strategies are also discussed, along with some further related aspects of VSL application, such as driver compliance to VSL and the integration of a differential speed limit (DSL) with VSL. Finally, the paper presents a critical review, pointing out some key issues that were identified from the literature review and recommends some directions for future research that need to be addressed as recommendations for best practices to underpin further refinement, development, and application of VSL.

Demand and Supply Calibration of Dynamic Traffic Assignment Models: Past Efforts and Future Challenges

The calibration of dynamic traffic assignment (DTA) models has evolved rapidly over the past decade and has been fueled by the need for applications ranging from long-term planning to real-time traffic operation. Numerous formulations and solution approaches based on either analytical or simulation-based approaches have been introduced. The aim of this paper is the documentation of the existing DTA model calibration approaches for future reference. The literature on the calibration of DTA models can be categorized on the basis of the two major components that need to be calibrated: demand and supply model estimation. Travel behavior modeling and origin–destination demand estimation problems are considered in the determination of demand models. Supply models simulate traffic dynamics, queue formation, dissipation, and spillback in either a microscopic or mesoscopic context. Early DTA model calibration efforts were based on iteration between the two demand and supply components. Recent frameworks have focused on the simultaneous calibration of both components. Therefore, different solution approaches have been addressed with various functional needs and degrees of robustness. This paper summarizes the current understanding of calibration and estimation of all input parameters for a DTA model, reviews the existing literature, and highlights the gaps that need to be addressed in future research.

Injury Risk in Collisions Involving Buses in Alberta, Canada

With 2000 to 2007 crash data, this study investigated the factors that contributed to injuries in collisions that involved at least one bus in the province of Alberta, Canada. Crashes of all types of buses (e.g., school, transit, intercity) were considered. Four logistic regression models were calibrated: single-vehicle collisions on highways, single-vehicle collisions on nonhighway locations, two-vehicle collisions on highways, and two-vehicle collisions on nonhighway locations. The analysis showed that weather conditions were a significant contributing factor in all four types of collisions, although crashes in adverse weather conditions resulted in fewer injuries. The type of collision, characteristics of collision partner, driver age of collision partner, and weather conditions had a significant effect on the level of severity of collisions on both highway and nonhighway locations. Other factors were shown to affect injury risk only in one particular situation. For instance, for highway-related collisions, the age of the collision partner had a significant effect on levels of accident severity, whereas the age of the bus driver did not. In addition, for highway collisions, the severity was higher for head-on crashes, bus–bus crashes, bus–truck crashes, bus–motorcycle crashes, older buses, crashes on grade and in sags, and crashes during dark and sun glare, whereas accident probability decreased with larger outside shoulder width. For nonhighway locations, crashes occurring near tunnels, overpasses, and signalized intersections were shown to result in a higher probability of injury. The results showed that single-bus collisions involving pedestrians at nonhighway locations had higher injury risk than collisions involving objects.

A Probe-Based Variable Speed Limit System

This article presents the results of an evaluation of a candidate variable speed limit (VSL) system that takes the space mean speeds (SMSs) from probe vehicles as its main input. The presented algorithm extends the capability of a model predictive control (MPC) VSL model with the input of SMS derived from vehicle probes and their corresponding space-based densities, instead of spot-based densities and detector speeds. The developed probe-based VSL was evaluated on an 8-km stretch of Highway 2, in Calgary, Alberta, Canada, using Paramics microsimulation software. The performance of the probe-based VSL algorithm was examined under various traffic conditions and probe vehicle parameters. The findings of the probe-based VSL analysis support previous findings taking input from point detectors and showing that the efficiency of the VSL strategies was efficient only for a limited range of traffic conditions. The results of the analysis show the effectiveness of developing advanced freeway control algorithms that use the main input parameters from vehicle probes. The performance of the probe-based VSL system was also comparable to, and in some cases outperformed, the point-based detector algorithm. However, the algorithm was shown to be effective only for a high frequency of information updates and under a relatively high percentage of vehicles probe penetration percentage.