The formation of shockwaves demarking boundaries between different flow states in time-space domains has been well studied and documented in the literature. Previous research concentrated on theoretical and mathematical analyses of shockwaves, providing valuable insights into shockwave phenomena. In the past few years, the traffic signal systems community has witnessed an increasing interest in adaptive control algorithms. These algorithms strive to achieve real-time platoon accommodation by assuming platoon speed and dispersion characteristics in order to predict its time of arrival at the downstream signal. Instead of assuming platoon speed and dispersion characteristics, we propose an on-line analysis of shockwaves at the downstream intersection to achieve the same purpose. This paper introduces a new metric that can detect and quantify shockwaves formed as a result of a suboptimal offset at an intersection. The new metric, the $F^{\prime }$ model, captures the effects of shockwaves using cycle-based tabulation of advance detector occupancy and count. A discriminant analysis is conducted to develop thresholds to which $F^{\prime }$ values can be compared to estimate the offset performance. The metric introduced in this paper can be used for implementing adaptive offset tuning algorithms in a traditional closed loop coordinated traffic signal system.

Police radar is known to have an effect on the speed of drivers. This effect derives from the presence of vehicles equipped with radar detectors in the traffic stream and the reactions that the drivers of these vehicles have to radar. The level of impact on a traffic stream exposed to radar transmissions is a function of the proportion of radar detector equipped vehicles in the traffic stream, i.e., the radar detector density. Microscopic simulation can be used to model this impact. A microscopic simulator extension is proposed that allows determinations of this type to be carried out by adding the capability to model driver response to radar transmissions. This approach can also be applied to the modeling of other driver warning technologies such as changeable message signs. A case study using the proposed methodology to analyze different congestion and radar detector density levels is presented and discussed. The case study shows that the efficacy of using radar as a speed reduction strategy is a function of congestion and radar detector density, with the strategy being most effective for volumes levels between 200 and 1,400 vehicles per hour per lane.

Simple or complex travel patterns, performed as activities, are linked, and various travel modes are used. Individuals try to link several trips as a type of chain to save transportation resources, such as travel distance or time, which consequently produces complex travel patterns. These changes in the urban transportation environment and some studies on activity-based characteristics make it difficult to forecast trips as a result of trip linkage behavior. Two models are established to achieve research objectives. The first is an allocation model for travel modes for activities by the three stage least square (3SLS) estimation method under the restricted conditions. The other is a covariance structure model to analyze direct and indirect effects among latent variables and estimated variables. As a result of 3SLS, the allocation of travel modes is mainly influenced by personal attributes such as gender, education level, and marriage and household attributes, such as the existence of children under 13 years old. In the covariance structure model, it is shown that exogenous estimated variables for personal characteristics explain well the personal latent variable and only household income explains the household latent variable. While the travel behavior latent variable is well effected by travel mode latent variable, it is less effected by the activity latent variable.

Macroscopic emission models use average speed as a sole traffic-related explanatory variable. Research, however, has demonstrated that the use of average speed as a single traffic-related explanatory variable is insufficient in estimating vehicle emissions. The objective of this paper is to attempt to quantify, using simple examples, the impact of vehicle stops on fuel consumption and emissions of hydrocarbons, carbon monoxide, and oxides of nitrogen. This study indicates that the vehicle fuel consumption rate is more sensitive to cruise-speed levels than to vehicle stops. The aggressiveness of a vehicle stop, as represented by the vehicle’s acceleration and deceleration level, does have a significant impact on vehicle emission rates. Specifically, hydrocarbon and carbon monoxide emission rates are highly sensitive to the level of acceleration when compared to cruise speeds in the range of 10–120 km/h. Alternatively, the impact of deceleration levels on all measures of effectiveness is relatively small. Noteworthy is the fact that at high speeds the introduction of vehicle stops involving extremely mild deceleration and acceleration levels can actually reduce vehicle emission rates.

At the planning stages of a highway development process a number of candidate highway alignments are explored. While computerized highway optimization models can seek optimized horizontal and vertical alignments by minimizing aspects of total highway cost, selection of the final alignments is often influenced by environmental impact assessment and public and political participation. In order to ensure that there are no adverse impacts on the environment and the living conditions of the residential community, the Federal Highway Administration has initiated a context-sensitive design approach to highway design. A criteria-based decision support system is developed in this research for selecting highway alignments using genetic algorithms and geographic information systems, which can assist in estimating trade-offs among alignment-sensitive cost, alignment circuity, floodplain and wetland impacts, and intersections with existing highways. A procedure for selecting notably different alignments is proposed by evaluating alignments obtained at intermediate search generations by genetic algorithms. A set of different alignments can provide enough leeway to explore trade-offs among cost and environmental and roadway impacts. An example using Maryland’s geographic database is presented.

A modeling framework is developed to assess the performance of driver information systems. The two main components of the framework are a route choice module and a traffic simulation module. In the route choice module, three types of driver information systems are studied: the pretrip information service, the in-car route guidance system, and the variable message signs. The traffic simulation module controls the dynamics in a road network as traffic evolves over time in response to driver information. The dynamic properties of the framework are demonstrated in a test network and a simplified Singapore road network. The results suggest that there could be an optimal level of subscription to driver information or market penetration. Subscription to real-time driver information services is recommended, but a high level of subscription may result in extra delay. Variable message signs are not as effective as either pretrip information or in-car route guidance at their respective optimal levels of market penetration.

A two-phase fluids model for mixing traffic flow on freeways has been proposed, where vehicles are decomposed into two parts (i.e., phases)—slow moving and fast moving–denoted by subscripts $i=1$ and 2, respectively. Based on the fact that both phases should be at rest under traffic jam conditions, it is assumed that the vehicular speeds for both phases are functions of the global traffic density, so that traffic flux can be expressed explicitly, considering that the speed of the second phase may be decreased when the mass fraction of the first phase becomes large. In addition to the relation to global density of traffic, it is assumed that the speed of vehicles in the second phase also depends on the mass fraction for the first phase. By neglecting the traffic generation rate, the governing equations from the mass conservation law were solved numerically with the Yee-Roe-Davis second-order symmetrical total variable diminishing algorithm. Two cases were considered: first, that there exists a soliton in the initial distribution of global density; second, that there is an initial uniform global density. Both cases were allowed to have a soliton in the initial density for the slowly moving vehicles. The numerical results indicate that the evolution of a soliton in traffic is quite different from those in water wave problems. The initial solitary-wave perturbation has been distorted dramatically. It was found that the presence of a soliton just in the slowly moving vehicles can increase global density, which means that traffic mixing can be viewed as a source of density wave production. Under very congested traffic flow, the speed for moving vehicles fast approaches that of vehicles in the first part.

A comprehensive evaluation is presented of the single-station wavelet energy neural network freeway incident-detection algorithm of Karim and Adeli. Quantitative performance measures of detection rate, false alarm rate, and detection time as well as the qualitative measure of portability are investigated for both urban and rural freeway conditions. Further, the performance of the algorithm is compared with that of California algorithm 8. This research demonstrates the portability of the wavelet energy algorithm and its excellent performance for urban freeways across a wide range of traffic flow and roadway geometry conditions, regardless of the density of the loop detectors. Rural freeways present additional challenges in that flow rates are low and detector stations are spaced further apart. Considering the difficulty in automatic detection of incidents on rural freeways, the new wavelet energy algorithm performs well on such freeways. The algorithm is fast as it detects an incident on urban freeways in less than 2 min and on rural freeways in less than 3 min.

This paper presents results of an in-service performance evaluation of the bullnose median barrier used in the state of Iowa. The data were collected in a four-county portion of Iowa during a 24-month data collection effort from 1997 to 1999. The bullnose barriers in the data collection area were usually installed consistent with Iowa DOT policy and represent good quality installations. Their collision performance was assessed in terms of collision characteristics, occupant injury, and barrier damage. While the data set for this study was small, examination of the collisions indicates that impacts at and near the nose often resulted in unacceptable vehicle behavior and severe occupant injuries. The data suggest that there is a need to develop median treatments that prevent vehicles from penetrating the system and contributing to injuries of vehicle occupants.

This study was conducted to investigate the visibility effectiveness of hemisphere and corner cube pavement markers on various curved road driving conditions. The study consisted of a field experiment to investigate visual threshold and an evaluation study to investigate visibility effectiveness. In the field experiment, a U-shaped test track with a curvature of 30 m radius was set up to investigate the visibility threshold. Two marker types, the hemisphere and the corner cube, and two directional driving conditions, right-turn and left-turn conditions, were considered. The results indicate that there is no difference between marker types, but there are differences between turning directions. The left turn has a significantly higher value and is therefore determined to be the resultant visual threshold. The resultant visual threshold is $1.70\times {10}^{-6}\mathrm{lx}\text{,}$ and this is significantly higher than the recommendation of the Commission Internationale de l’Eclairage (CIE). The evaluation study was to evaluate the visibility effectiveness of these two markers on eight curved roads with different radii based on the resultant visual threshold. The evaluation indicates that both pavement markers meet the 3 s preview time as suggested by CIE. However, it also suggests that the corner cube is superior on roads with radii greater than 60 m, whereas the hemisphere is much better for the roads with radii smaller than 60 m.

An integrated methodology is proposed for the design of polluted water control systems for road infrastructures, based on hazard analysis concepts. The case of a simultaneous rainstorm and road accident with the dispersion of dangerous substances is considered. The risk function is computed with regard to the objective evaluation of severity of the event and to the stochastic calculation of the associate probability. A logical framework for the evaluation of severity is proposed by taking into consideration four possible circumstances for the dispersion of pollutants: surface and ground water, and natural and urban soil. A set of check lists have been devised for the assessment and design of the environmental protection control systems. The methodological approach is completed with the formulation of a stochastic model representing the statistical combination of the occurrence of the hydrological processes and the road accident dynamics. The stochastic model estimates the expected volume of contaminated rainfall for any defined probability of occurrence of the polluting event, i.e., return time of the hazard, in a coherent statistical setting. The proposed integrated methodology has been applied to a real case presently under construction: the Salerno–Reggio Calabria Highway located in Southern Italy.

Ground-penetrating radar has been used successfully to predict layer thicknesses in flexible pavements. Currently, a time-domain method is used to measure the two-way travel time between reflections from which the layer thickness is predicted. In this method, the dielectric constants are assumed to be real and are calculated from simplified equations. This paper presents a new method to measure layer thicknesses in flexible pavements. The method is based on predicting the reflected frequency spectrum through a multiple reflection model. The dielectric constants of the materials are assumed to be constant over the entire frequency range of the radar system, although dielectric and conduction losses are considered in the model. An optimization procedure is used by first calculating the reflected frequency spectrum using several values for the thicknesses and the complex dielectric constants, then choosing the solution with the minimum root-mean-square error between the predicted and measured frequency spectrums. The method was used in three pavement sections with known layer thicknesses. A percentage error of less than 3.6% was found between the measured and predicted values.

Subgrade soil characterization in terms of resilient modulus $\left(M_R\right)$ is essential for pavement design/evaluation. Due to the complexity of the laboratory $M_R$ test procedure, highway agencies have been exploring nondestructive tests, such as the falling weight deflectometer. In this study, 12 test sections with subgrade reflecting a range of soil types (fine- and coarse-grain soils) were tested. Falling weight deflector measurements were conducted directly on the prepared subgrade, followed by thin-wall Shelby tube samples from depths up to 1.5 m for laboratory $M_R$ tests. Deflection measurements were repeated at the same locations after completion of pavement construction. Results showed that the backcalculated moduli of the prepared subgrade are in good agreement with the laboratory values. Due primarily to confinement and overburden pressure, the backcalculated subgrade modulus, based on deflection measurements on the finished pavement, increases approximately 40 and 100% for fine- and coarse-grain soil, respectively. The modulus increase is substantiated by complementary studies of 20 Mississippi long-term pavement performance sections. In light of these results, the AASHTO factor of 0.33 to adjust for the backcalculated moduli, computed from deflection measurements on finished pavements, warrants further scrutiny and revision.