The abilities of two line-source dispersion models to predict road-edge concentrations of carbon dioxide along several roads are tested. The comparison of the predicted CO₂ concentrations from both CALINE4 and HIWAY-2 using a new power-based emissions module are compared with eight days of experimental data collected at three sites in Sydney through a recent field study carried out by the Commonwealth Scientific Industrial Research Organization (CSIRO) on behalf of the Road and Traffic Authority (RTA) in Sydney to quantify air pollution near roads and highways in Sydney. The measurements were made at locations up to 60 m downwind from the roadside and to heights 10 m above the ground. The comparison of the measured CO₂ concentrations and those predicted on the basis of appropriate traffic and meteorological data shows that both CALINE4 and HIWAY-2, when used in conjunction with the new vehicle emissions module, satisfactorily estimate road-edge concentrations of CO₂. As such, both the two dispersion models, CALINE4 and HIWAY-2, in conjunction with the new power-based emissions module, are potentially useful tools for highway planners.

In this paper, a new transition curve to be used in the design of the horizontal geometry of routes has been examined. This curve joins two straight lines with a single element. It is an alternative curve to the well-known classical transition curve, which is formed by spiral curve–circular arc spiral curve. In order to compare the suggested transition curve with the classical transition curve in terms of the vehicle-road dynamics, the curvature and superelevation functions of both curves are derived, and funcions of the lateral change of acceleration of both curves are obtained and illustrated graphically for the seven selected motion models. The comparisons that take into account the discontinuities in the form of jumps in the graphics of the lateral change of acceleration have shown that the new transition curve is superior to the classical transition curve for all seven motion models.

Several variables are usually considered when fund is allocated for highway improvement programs. Among these variables are the project improvement alternatives for each highway, the different locations available, and the annual maintenance budget. The different improvements and location alternatives provide the flexibility needed to optimize the costs or benefits of improvement projects within the constraint of a given budget. Integer programming techniques and computer programs are the basic requirement for solving the fund allocation problems. These programs and the appropriate computers, despite the decrease of hardware and software costs, may not always be available, especially in developing countries. There can also be a shortage of professionals who can use the integer programming techniques to solve the problem. This paper presents a developed tabulated manual procedure to get the optimum or “nearly optimum” solution for the highway maintenance fund allocation problems. A sample project solved by the computerized integer program is given, and the result is compared with that obtained by the developed tables.

The main objective of this paper is to introduce a systematic approach to determine the need for and location of truck escape ramps (TERs). The approach utilizes variables concerning the design vehicle, the road, and the driver. These variables are included in either a stochastic or deterministic form. Vehicle variables include vehicle type, gross vehicle weight, and brake type. Road variables include the profile (length and steepness of grade), horizontal curvature (cornering speed), and the availability of a brake-check area at the top of the grade. Driver variables include information about the driver's actions at the brake-check area, brake checking, and gear selection. A number of supplementary variables are also included, such as accident history and the presence of objects in the path of a runaway vehicle. The approach also utilizes a number of existing vehicle-road performance models to generate relevant data concerning the descent speed, cornering stability, and brake temperature. The application of the methodology is illustrated with examples from highways in mountainous terrain of Jordan and western Canada. The examples selected are diverse and were chosen to illustrate the generalized nature of the methodology.

In a study of effects of wheels with very heavy loads but very low tire inflation pressures on unbound, single-layer granular pavements, trafficking tests were carried out on full-scale indoor pavements at the Pavement Test Facility, Transport Research Laboratory, Crowthorne, U.K. Two values of wheel load, two values of tire inflation pressure, and three subbase thicknesses were used. Testing was continued to 10,000 passes and a total of 4,704 measurements of rut depth made at intervals, providing a statistically significant data set. The development of ruts, the rut cross sections, and the effects of halving the tire inflation pressure or nearly doubling the wheel load are presented. Comparisons are made to a rut prediction equation derived by the U.S. Army Corps of Engineers Waterways Experiment Station in similar work, and conclusions are drawn.

The main objective of this study was to investigate the performance of safety evaluation methods and estimate the safety effectiveness of street bumps, closing of median openings, and divisional islands in an urban environment. Traffic accident data were obtained from the municipality of Irbid, Jordan, and its police records. In the analysis the performances of simple before-and-after and empirical Bayesian methods were compared with results obtained by comparison group methods. The results indicate that the simple before-and-after method overestimated the effectiveness of safety improvements and led to erroneous conclusions at specific locations, as well as at the aggregate level. The comparison group method and the empirical Bayesian method provided results comparable to each other. Based on the results of the comparison group and the Bayesian methods, the study indicates that street bumps and closing of median openings at intersections are effective in reducing total accidents. In contrast, implementation of divisional islands did not provide a significant reduction.

This paper presents the results of finite element computer simulations of vehicular impacts with a roadside crash cushion. DYNA3D is used to model the energy-dissipating response of the Narrow Connecticut Impact Attenuation System under high-speed impacts with heavy and light vehicles. The finite element predictions are compared with the experimental results of full-scale crash tests involving automobiles impacting the device at 97 km/h. It is demonstrated that computer simulations can effectively model the dynamic response of a crash event and provide valuable measures of occupant risk. Computer simulations allow the investigation of crash scenarios involving nontracking, braking, and turning vehicles. Such full-scale crash events are very difficult to conduct in the field. It is recommended that these simulation tools be used more extensively in the design of roadside safety features so that the number of expensive full-scale crash tests required to develop future hardware can be minimized.

The static elastic layer model is customarily used for analyzing deflection measurements and back-calculating pavement layer moduli. Estimation of in-situ layer moduli by means of a mechanistically based iterative technique is known as back-calculation. Despite the fact that the falling weight deflectometer (FWD) load induces a dynamic load, dynamic/impact analysis routines are seldom, because of the mathematical complexity. Dynamic deflection prediction models are developed in this study, which can be used in back-calculation routines. The first step in accomplishing this is to select an appropriate analysis model for which ABAQUS—a general purpose finite-element program—is used. Static and dynamic deflection responses from the ABAQUS model of an uncracked pavement are validated. Following this first step, cracks and joints along with other features of the pavement are modeled in addition to nonlinear behavior of pavement materials. Deflection responses of cracked pavement under a series of increasing loads, assuming linear and nonlinear behavior of base, subbase, and subgrade materials, are computed and compared. After finalizing the model, a synthetic deflection database, specifically with FWD load, is developed relying on a fractional factorial design layout in which thicknesses, layer moduli, and cracks are allowed to vary over a range. Making use of this database, regression equations that predict surface deflection bowls in terms of layer moduli and thicknesses are developed. Deflection equations are validated using field data from two in-service pavements.

A study was undertaken to determine the drainability characteristics of several types of unbound granular materials that are used in pavement bases. From the literature, testing concerns were identified and analyzed. Hydraulic conductivity and effective porosity were determined for aggregates from two sources of crushed stone and two sources of gravel. For each material two open gradations were tested in a rigid wall permeameter. One dense gradation of each material, at two levels of compaction, was tested in a flexible wall permeameter as well. Efforts were made to reduce possible flow along the sides of the rigid wall permeameter and to eliminate problems with air bubbles. The average hydraulic conductivities for the open and dense-graded materials were 3.6 × 10⁻¹ and 3.0 × 10⁻⁴ cm/s (1,021 and 0.85 ft/d), respectively. The effective porosities for the open and dense-graded materials were 68 and 27% of the nominal porosities, respectively. Four regression equations representing a wide range of hydraulic conductivity were developed to estimate hydraulic conductivity based on the results of this study and others reported in the literature.

The relative effectiveness of commercially available road dust suppressants in abating fugitive dust emission and loss of fines from unpaved road surfaces was assessed in a field-based research project. The dust suppressants studied, lignin derivatives and chloride based compounds, were used on unpaved road test sections during the severe dusty months of 1993 and 1994 in Colorado, which fell in the late spring to autumn period. To measure the relative effectiveness of the different dust suppressants, comparative fugitive dust emission studies were conducted on unpaved road test sections using the Colorado State University Dustometer, a dust-sampling device developed during this research. In addition, total aggregate loss from the surfaces of the test sections was measured. Based upon the prevailing costs, analyses were performed to determine the economics of using the different dust suppressants. The research indicated that all three dust suppressants studied reduced fugitive dust emission from the unpaved roadways by 50–70%. The treated test sections retained 42–61% more aggregate than the untreated control test section. The cost savings of retaining aggregate on the treated test sections more than offset the costs of the dust suppressants, resulting in an estimated cost savings of 28–42% over the untreated control test section.

The objective of this research project is to determine the service lives of three different types of joint sealants—polyvinylchloride (PVC)-coal tar, rubberized asphalt, and silicone—and to recommend maintenance strategies to improve the performance of concrete pavements in Utah. The silicone material has been performing well for over 10 years. The PVC-coal tar material had failed in the test sections that are an average of 15 years old, and the rubberized asphalt material has aged beyond its expected service life in the test sections over 20 years old. Survival curves of the three sealants were developed with respect to their ages, and their performance correlated well with their ages. The conical spalling at the bottom of the joints increased after 10 years of service. Based on the limited field data, it is recommended that Utah Department of Transportation (UDOT) should consider the use of a narrower joint width of 6 mm instead of the current 9-mm design, use silicone or rubberized asphalt material, and implement a maintenance plan with a 10-year cycle for recutting and resealing of the joints to 9 × 9 mm.

Current failure criteria of asphalt pavements are either empirical or assume linear elastic material response and use a single load level to relate the number of load repetitions to fatigue failure. To better understand the crack propagation properties of asphalt pavements, laboratory tests and nonlinear analysis were performed to evaluate the low-temperature fracture parameters of conventional asphalt concrete and asphalt-rubber mixture. Two approaches based on nonlinear fracture mechanics, the compliance approach and the R-Curve approach, were used. Beam specimens were prepared with different binder contents and tested under three-point bending flexural conditions at two test temperatures. A closed-loop servohydraulic test system was used with the crack mouth opening as the control parameter. Nonlinear fracture parameters were obtained at different stages of crack propagation. Results show that the asphalt-rubber mixture has higher fracture toughness and consequently larger resistance to cracking than asphalt concrete. Also, the asphalt-rubber mixture is less sensitive to temperature than asphalt concrete. Increasing the binder content increased the toughness values for both asphalt concrete and asphalt-rubber mixture in most cases. The R-Curve approach provides a good measure of characterizing the fracture behavior of asphalt mixtures.