At signalized intersections where protected left turn signals are not provided, simultaneous opposing left-turn vehicles may cause insufficient sight distance for each other. There is a lack of literature presenting left-turn sight distance analysis at intersections located on curved highways. In this paper, sight distance calculation models focusing on curved-approach intersections were developed for left-turn maneuvers toward the outside of the curve and toward the inside of the curve, separately. Using those models, it was found that, for left-turn maneuvers toward the outside of the curve, opposing left-turn vehicles might cause serious sight distance problems even for a 3.7-m median. However, for a left-turn maneuver toward the inside of the curve, the curve may mitigate the sight distance problem and the median width plays a more important role in the available sight distance, especially when the curve radius is relatively large. The models and related analyses can be used to lay out intersection design or evaluate the sight distance problem of an existing intersection located on the horizontal curve to ensure safe left-turn maneuvers by drivers.

Sight distance is a key element in highway geometric design. Existing models for evaluating sight distance are applicable only to two-dimensional (2D), separate horizontal, and vertical alignments or simple elements of these separate alignments (vertical curve, horizontal curve). A new model using global positioning system (GPS) data is presented for determining the available sight distance on 3D combined horizontal and vertical alignments. Piecewise parametric equations in the form of cubic $B$ -splines are used to represent the highway surface and sight obstructions, including tangents (grades), horizontal curves, and vertical curves. The available sight distance is found analytically by examining the intersection between the sight line and the elements representing the highway surface and sight obstructions. A profile of available sight distance can be established and used to evaluate sight-distance deficiency. Application of the new model is illustrated using actual GPS data for highway K-177 in Kansas (United States). The model has been tested and verified on most of the highways in Kansas. Software has been developed and can be used for determining the available sight distance on any highway for which GPS data are available.

The relevance of posting the $65\mathrm{km}∕\mathrm{h}$ $(40\mathrm{mi}∕\mathrm{hr})$ minimum speed limit signs on rural interstate freeways in Florida is analyzed by correlating safety and traffic operating characteristics. Analysis of the lower end of speed distribution showed that only 0.14% of the vehicles were traveling at speeds below $65\mathrm{km}∕\mathrm{h}$ . However, a four-year data study of reported crashes showed that the percentage of total crashes involving vehicles with speeds less than $65\mathrm{km}∕\mathrm{h}$ was 9%. The overrepresentation of the slow moving vehicles in the total number of reported crashes raises concerns for the safety implications which could result in the aftermath of raising the minimum speed or scrapping the posting practice as 25 other states have. Further research results indicated that the 15th percentile speed on these freeways sections was $95\mathrm{km}∕\mathrm{h}$ $(60\mathrm{mi}∕\mathrm{hr})$ . It is worth noting that it has been suggested in the literature that the 15th percentile speed should be used as a guide in setting the minimum speed limit on freeways. Further analysis showed that the $16\mathrm{km}∕\mathrm{h}$ $(10\mathrm{mi}∕\mathrm{hr})$ speed range with the most crashes was considerably lower than the $16\mathrm{km}∕\mathrm{h}$ speed range for data collected in the field. This paper further reports on Poisson regression modeling conducted to determine the influence of various speed attributes on crashes occurring on rural interstate freeways in Florida.

The Highway Capacity Manual (HCM) defines uniform delay as a portion of control delay and estimates it based on the flow diagram originally developed for a protected movement. This paper demonstrates the shortcomings of the HCM delay model in the estimation of uniform delay of permitted left turns from an exclusive lane (PLTEL), and proposes a new uniform delay model developed for PLTEL. The control delays estimated by the original HCM model and the one estimated by the proposed uniform delay model for PLTEL were compared against those from corridor simulation. Results showed that the HCM model tended to underestimate uniform delay, and the proposed model for PLTEL showed improved results.

The condition of a road network and its deterioration rate can be estimated by using measurements and statistical models. The purpose of this paper is to provide tools for assessing the accuracy of the condition information based on measured and modeled values. In this study, International Roughness Index (IRI) measurements over 3 years (2000–2002) were used. A logarithmic transformation was applied to the measured IRI values. One half of the data set was used for developing regression models that predict road roughness. These models were validated using the other half of the data set. The comparison of the residual distribution in the logarithmic regression model and measurement accuracy in logarithmic terms facilitates direct consideration of the relative accuracies. The Taguchi loss function was applied to estimating the losses incurred when measured and modeled values were used. The decision of taking new measurements depends on the relative accuracies of the measurement and the modeling, the cost of measurement, and the losses incurred to the road users and the maintaining agency due to untimely maintenance.

The mitigation of severe problems resulting from vehicle collisions with roadside objects has become one of the major research areas in automotive engineering. The literature review shows that few attempts in finite-element computer simulation of vehicle collision with roadside hardware have been conducted. However, limited research has been conducted to enhance the safety performance of traffic light poles when impacted by vehicles. The objective of this paper is to generate information that can be used to enhance energy absorption characteristics of transportation infrastructure involved in vehicle crash accidents. A finite-element computer model, using the available LS-DYNA software, was developed to simulate vehicle collision with a traffic light steel pole in frontal impact. Five configurations of steel pole supports were examined, including embedding the pole directly into the soil. Different types of soil conditions were examined to study their effects on vehicle occupant safety. The study of structural response focused on the energy absorption, acceleration, and deformation of the steel pole and the vehicle. It is demonstrated from numerical simulations that the steel pole embedded directly into the soil is proved to be strong enough to offer protection under service loading and to remain flexible enough to avoid influencing vehicle occupants, thus reducing fatalities and injuries resulting from vehicle impact.

This paper presents the estimation results of an improved nonlinear rutting progression model for asphalt concrete mixes. Rutting is modeled as a function of traffic, temperature, and mix characteristics, including voids filled with asphalt, asphalt content, in-place air voids, surface area, and the densification slope. Mixed effects, an advanced statistical technique that allows the consideration of several sources of randomness and autocorrelated disturbances, is used for the estimation of the model parameters. The estimation results with measurements from WesTrack indicate that the model is statistically significant. Further, the model correctly anticipates the dramatic failure of the WesTrack replacement sections (approximately $10–20\mathrm{mm}$ in a few days). Mixture rutting resistance is linked to the aggregate resistance to densification, as measured by the densification slope in the Superpave gyratory compactor. The model may provide a basis for identification of probable mix performance in the field with the help of only normal quality assurance tests and in-place density measurements. Finally, since the parameters related to asphalt content are binder specific, the specification of the model provides a framework that can be expanded to include other asphalt binder characteristics.

The Vermont Agency of Transportation (Agency) developed pavement design procedures patterned after the American Association of State Highway and Transportation Officials (AASHTO) pavement design model described in the AASHTO 1993 pavement design guide (guide). A consequence of using the structural number approach to pavement design is the need for structural layer coefficients $\left(a_i\right)$ to characterize the component materials of the pavement structure. The agency has developed and tested a method for determining layer coefficients using a falling weight deflectometer (FWD), and the resulting layer coefficients are representative of the in situ behavior of the pavement materials. The agency found layer coefficients determined for unbound subbases to be reasonable, while layer coefficients estimated for asphalt cement concrete materials were generally 25–35% higher than AASHTO’s implied maximum of 0.44. However, a statistical analysis indicates considerable support for the predictive qualities of FWD derived layer coefficients to approximate layer coefficients simulated from the in situ conditions expected to prevail in the final pavement structure.