Past speed-related research has focused on the operational effects of roadway geometrics along rural two-lane highways using ordinary least-squares regression models. More recent research has focused on the association of traffic flow characteristics on vehicle operating speeds along multilane, limited access highways using a simultaneous equations approach. Few research studies, however, have been conducted to determine the combined association between various geometric design features and traffic flow on operating speeds along multilane highways. This research considers both geometric design and traffic flow parameters, in a simultaneous equations framework, to model the mean operating speed and speed deviation on four-lane highways (two lanes in each direction). Models for both left- and right-lane mean speeds and speed deviations were estimated. The three-stage least-squares estimator was used to investigate the possible endogeneity of mean speed and speed deviation in the system of equations and to account for the contemporaneous correlation between the disturbances across the equations. The results indicate that different geometric design features are associated with mean speed and speed deviation in the left- and right-lane models. As such, it is recommended that future multilane highway speed models consider using a simultaneous equations framework.

This paper focuses on the estimation of highway maintenance marginal costs. Highway maintenance marginal cost has been estimated in the literature using the perpetual overlay indirect approach. This approach assumes that pavement overlay costs dominate maintenance costs and ignores other maintenance activities. This paper focuses on two questions. First, is it acceptable to ignore the less costly activities? Second, if multiple maintenance activities are to be considered, is it acceptable to ignore their interdependence? The results show that less costly maintenance activities cannot be ignored. Furthermore, if multiple activities are to be considered, their interdependence should be taken into account.

Collision prediction models (CPMs) are statistically developed mathematical models that relate the occurrence of traffic collisions to various traits or explanatory variables of locations and zones. This paper describes the use of several recently developed community-based, macro-level CPMs and model-use guidelines in a CPM transferability case study for spatial and temporal transferability. The objective was to test the model-use guidelines in an application that involved transferring CPMs developed using 1996 data for the Greater Vancouver Regional District (GVRD) for use in the Central Okanagan Regional District (CORD) using 2003 data. The GVRD and CORD are regions located 400 km apart in the Province of British Columbia, Canada. The case study was carried out in two parts. First, CPMs were developed using 2003 data for the City of Kelowna following recommended community-based, macrolevel collision prediction model development guidelines. Second, existing CPMs were transferred from the GVRD to the CORD, using the recommended transferability guidelines. An analysis of the results revealed that macrolevel CPM transferability was possible and no more complicated than microlevel CPM transferability. To facilitate the development of reliable community-based macrolevel collision prediction models, it is recommended that CPMs be transferred rather than developed whenever and wherever communities lack sufficient data of adequate quality.

This paper models road traffic demand of container trucks in a Chinese container port. The model is helpful for road planning and traffic management in the port. First, we analyze the attributes of road traffic in the port, to distinguish the differences between port area traffic and surrounding urban area traffic. It is found that port area traffic is generated from vessel operations and fluctuates corresponding to two major factors: operation time schedule and the truck arrival pattern within an operation time window. Since the first factor is available from terminal operators, the truck arrival pattern is the key for modeling port area traffic demand. Second, we conduct a comprehensive survey to collect truck traffic and vessel operation data, which is used to explore the probability distribution of truck arrivals during the consolidation periods. Third, we develop a traffic demand model based on the probability distribution with unknown parameters, and modify this model by taking the effects of some external factors into account. At last, we estimate the parameters with the collected data, and the result indicates that this traffic demand model has a high quality of estimation especially for peak traffic times.

The objective of this research is the analysis of perceived highway level of service in relation to drivers’ personal characteristics and traffic conditions. A field survey was carried out for that purpose, in which drivers were asked to assess the traffic conditions experienced on a freeway segment, just after having actually driven along this segment. At the same time, traffic counts were taking place. Exploratory analysis revealed a large dispersion in drivers’ assessment of level of service for each volume-to-capacity (v/c) value, especially at moderate traffic conditions. Furthermore, the effect of driver’s age, gender, driving experience, familiarity with the road, vehicle capacity, and v/c on perceived level of service was examined by means of nonparametric statistical tests. It was found that, in this sample, only traffic conditions affected drivers’ assessment of level of service. Perceived level of service was then modeled against v/c by means of a piecewise linear regression technique for three scenarios, covering different types of drivers with respect to their tolerance of congestion. The results confirm that the relationship between perceived level of service and traffic conditions has a piecewise linear form with significant differences in slopes and breakpoints. Moreover, they suggest that drivers perceive no more than two or three levels of traffic conditions. More specifically, only low-tolerance drivers appear to distinguish level of service A from B and only high-tolerance drivers appear to distinguish level of service D from E. On the contrary, an impressive reduction of perceived level of service occurs within the v/c interval of 0.55–0.70 in all scenarios. These findings are further discussed in relation to their accordance with the highway capacity manual approach for level of service and its possible adjustments.

This paper presents a microscopic traffic simulation approach to the capacity impact analysis of weaving sections caused by the installation of exclusive bus lanes (XBLs) on urban expressways. Three typical configurations of XBL in Beijing are identified, including median bus lane with off-on-ramp, curbside bus lane with on-off-ramp, and curbside bus lane with off-on-ramp. Then, the weaving section length, headway, mainline volume, as well as off-ramp and on-ramp volumes for general traffic are considered as parameters in designing simulation scenarios. Finally, the microscopic simulation is executed, by which the capacity of general traffic for weaving sections is estimated and the impact of the weaving section length and headway on the capacity is evaluated. It is found that the weaving section length and headway have different impacts on the capacity of general traffic in different configurations. The results also provide that the weaving section length and headway are more sensitive in the on-off-ramp than in the off-on-ramp scenario with the curb bus lane.

Travelers’ perception of reliability of a road network is typically based on factors contributing to both recurring and nonrecurring congestion. However, literature documents little research to integrate, estimate and assess congestion using both these disparate congestion components. Integrating recurring and nonrecurring congestion components to estimate congestion helps address questions such as “How reliable are roads in the transportation network?” “Which path is more reliable to reach the destination from an origin within an on-time window during a certain time of the day?” The focus of this paper is to develop and illustrate the working of a methodology to estimate travel time and its variations, travel delay index due to crashes and their severity, congestion score and reliability of each link in the network. Traffic volume, link capacity, travel speed, crashes and their severity, and estimated time taken for normal traffic conditions to restore after a crash are used in the computations. Temporal variations in travel time and crashes for each link are also studied as travel demand and crash occurrence depend on the time of the day. Sensitivity analysis is also conducted to examine and assess reliability of links based on variations in weights to integrate recurring and nonrecurring congestion components. Data collected for the city of Charlotte in Mecklenburg County, North Carolina are used to demonstrate the methodology.

Explaining pedestrians’ crossing behavior along entire trips may contribute toward more efficient and targeted planning of pedestrian facilities in urban areas and more accurate consideration of pedestrian safety. Although existing research on pedestrians’ crossing behavior is extensive, most related studies examine pedestrians’ crossing decisions at local level and focus on a particular set of determinants (roadway environment, traffic conditions, or human factors). On the contrary, crossing behavior along entire trips is seldom explored. This paper presents a theoretical framework for modeling pedestrians’ crossing behavior along a trip, addressing a large part of the difficulties involved in collecting the necessary data and setting up a modeling framework. First, a topological approach of pedestrian trip characteristics and crossing decisions is proposed, allowing consideration of distinct patterns of crossing behavior along a trip. Moreover, specific techniques from the family of discrete choice models are proposed for determining the number and location of pedestrians’ crossings, accounting for the hierarchical and dynamic nature of pedestrians’ decisions along a trip. Finally, a field survey method is presented, allowing collection of detailed information about pedestrians’ crossing decisions along urban trips, including data on roadway, traffic, and individual characteristics, as well as the interactions between pedestrians and motorists. Preliminary results from a pilot implementation of the proposed framework are promising, and a full-scale application for testing and validation is in progress.

This paper presents an investigation into car-following interactions on two-lane rural highways. Empirical data from several study sites on two-lane and four-lane rural highways in the state of Montana were used in this investigation. Three separate yet relevant analyses are presented in this paper. The first analysis involves a preliminary characterization of time headways and their distributions and shows how they are affected by passing restrictions on two-lane highways. The second analysis presented in this paper involves an investigation of speed-headway relationship to establish a headway threshold that could be used in defining free-moving vehicles on two-lane highways. The third and last analysis investigates the “following by choice” phenomenon on two-lane highways using data from two-lane as well as four-lane study sites. The study found that the car-following interaction generally ceases when headways exceed a value of approximately six seconds. Also, a significant proportion of drivers choose to maintain relatively short headways while following other vehicles on two-lane highways regardless of passing restrictions.

At the Pacific Highway port of entry between the United States and Canada, typical delays are known to regional carriers and internalized into schedules. Due to their relative infrequency, the largest crossing times are not internalized into schedules and cause significant disruptions to regional supply chains. This technical note describes the recent patterns of very long crossing times (defined as more than 2 h or the largest 1% of crossing times) and explores the relationship between arrival volume and crossing time. To do so, this study uses commercial vehicle crossing time data from GPS technology and volume data from the British Columbia Ministry of Transportation. Results show a weak correlation between border crossing time and arrival volume when considering individual observations, but a stronger correlation when data are aggregated. Results show a high percentage of crossing time can be attributed to sources other than primary booth delay, particularly for the most disruptive, very long crossing times.