The serviceability rating of roads has been an interesting and important subject ever since it was introduced to measure the road performance. The present paper analytically addresses this problem—namely, linking a static road profile with the dynamic response of a vehicle to the existing road profile and to the driver's impressions and evaluation of the ride character on the road. Using a quarter car model and applying the spectral density method, analytical expressions have been derived for various response dynamic indices, such as jerk index, acceleration index, average rectified speed (ARV), average rectified slope (ARS), and the international roughness index (IRI), in terms of the pertinent physical parameters of the road and the dynamic characteristics of the quarter car. A general expression for serviceability index of roads is suggested using Fechner's psychophysical law and is verified using field data.

To evaluate curling stresses in concrete pavements requires the input of a temperature differential or even a temperature distribution between the top and bottom of slabs. However, the slab temperature fluctuates throughout a day with many weather factors, such as air temperature, sunshine, clouds, and rain. It is still controversial about which measure during a day is the most appropriate input for curling analysis. The stress model adopted in this study to estimate pavement damage was developed with three-dimensional (3D) finite-element analysis. Field temperature measurements during different times of a day at 14 sites in the United States were input to calculate pavement compound stresses due to curling and wheel loads. By introducing fatigue hypothesis, an algorithm was developed to obtain “equivalent damages” and “effective temperature differential.” The calculated effective temperature differentials were further correlated with local climate data.

This paper presents the results of a study conducted to explore the use of small-scale models of accelerated pavement testing (APT) devices to evaluate the performance of pavements in conjunction with full-scale tests. The motivation for the study was the availability of a model mobile load simulator (MMLS), which had been built originally to illustrate the operation of the full-scale mobile load simulator (MLS) under design at the time. The scaling requirements will be different depending on whether dynamic (inertia), viscous, or gravity effects are important. One must thus decide which one of these effects controls the behavior to try to reproduce it exactly. In the preliminary tests conducted with the MMLS, emphasis had been placed in reproducing accurately the viscoelastic behavior of the asphalt layer. The possibility of obtaining valid results, even if similitude is not maintained in relation to inertia forces, is explored in this paper. The effects of load frequency or velocity and the effects of layer thicknesses are studied. The total thickness of the model pavement, which must be finite, and its effects on displacements and strains are also considered. It is concluded that even when full similitude is not satisfied it is possible to obtain valid results that can be extrapolated to predict prototype performance if one were interested primarily in the behavior of the asphalt layer. Preliminary analyses should be conducted, however, to guide on the selection of the model dimensions.

This paper deals with the back-calculation of elastic modulus of jointed concrete pavement. A series of road-rater tests are carried out on a jointed, Portland cement concrete (PCC) pavement of a partially completed freeway in Taiwan. The deflection data obtained with a road rater are used to back-calculate the modulus of elasticity of the PCC pavement and the coefficient of subgrade reaction. The back-calculation is done using BACK, a computer program developed based on the plate theory. The validity of the BACK program is verified with measured deflections, as well as predictions by ILLISLAB, an existing finite element code that has been proven reliable in predicting pavement responses. The results of this study show that the BACK program is quite reliable in predicting the elastic modulus of PCC pavement and the coefficient of subgrade reaction. However, the results are based on the analysis of PCC pavement over a granular subgrade and may not be directly applicable to those over asphalt-stabilized subbases.

This study considers a plate consisting of one or more layers, resting on a general elastic foundation, and subjected to a nonlinear through-the-thickness temperature distribution, together with arbitrary wheel loads. The resultant bending stress distribution in the plate is presented as the sum of the bending stresses due to the applied loading and to an equivalent linear temperature gradient, plus the pure thermal stresses due to the nonlinear part of temperature distribution. The resulting algorithm has been implemented into finite-element code ILSL2, capable of accommodating one- or two-layered slabs with a temperature distribution through the individual layers described by a linear function, a quadratic function, or explicitly provided by the user at a number of individual depths. Several examples are used to illustrate application of the method.

During the last few years, several teleoperated and machine-vision-assisted systems have been developed in construction and maintenance areas such as pavement crack sealing, sewer pipe rehabilitation, excavation, surface finishing, and materials handling. This paper presents a path-planning algorithm used for a machine-vision-assisted automatic pavement crack sealing system. In general, path planning is an important task for optimal motion of a robot whether its environment is structured or unstructured. Manual path planning is not always possible or desirable. A simple greedy path algorithm is utilized for optimal motion of the automated pavement crack sealer. Some unique and broadly applicable computational tools and data structures are required to implement the algorithm in a digital image domain. These components are described, then the performance of the algorithm is compared with the implicit manual path plans of system operators. The comparison is based on computational cost versus overall gains in crack-sealing-process efficiency. Applications of this work in teleoperation, graphical control, and other infrastructure maintenance areas are also suggested.

The permeability of a highway base layer is measured in the laboratory using a permeameter. The determination of coefficient of permeability (hydraulic conductivity) K utilizes Darcy's law, which assumes that the Reynolds number is less than some limit. Reynolds number increases as the pore sizes in the sample increase and as the mean velocity of the water flowing through the permeameter increases. Standard test procedures attempt to limit Reynolds number by limiting hydraulic gradients to a small number of values correlated to common situations. However, with open-graded bases, the standard values may produce insufficient control. In this study, 150 permeameter tests were conducted with dense-graded bases, open-graded bases, and open-graded, asphalt-treated bases. Analysis of the results led to an easy-to-apply equation that can be used before testing highway base samples to set test conditions (hydraulic gradient) that will produce laminar flow at a sufficiently low Reynolds number.

The results of a laboratory study on the influence of standard Proctor and West African standard compactive efforts as well as compaction delays up to 3 h on the compaction and strength characteristics of lateritic soil treated with a maximum of 8% lime (by dry weight of soil) is presented. The results obtained show that the compaction and strength properties of the lime-treated soil decreased with increases in compaction delays. The decreases that were generally greater at higher lime contents were more prominent in specimens compacted at the energy of the West African standard. The determination of properties of lime-treated soil at no compaction delay defines optimum properties of the soil-lime mixtures, while compaction and strength properties determined following compaction delays define the minimum that can be achieved in the field for the specified elapsed times between mixing and compaction.

The American Association of State Highway and Transportation Officials (AASHTO) model of calculating sight distance at intersections without control (Case I) has been modified to explicitly incorporate the design speeds of both intersecting roads in computing the required sight distance for each road. It is shown in this paper that using design speeds of both roads in the modified model will result in an inadequate sight distance for most vehicles. To address this issue, this paper presents a revised model that is based on a small percentile speed of the intersecting road (such as the 1 or 5 percentile). Using the revised model, design requirements of intersection sight distance (ISD) are established for passenger cars and trucks. The results show that the practice of using design speed in ISD analysis may result in significant underestimation of ISD requirements, particularly when the speed of the approaching vehicle on the intersecting road is extremely low. Furthermore, the amount of underestimation increases with an increase in the difference between the design speeds of the intersecting roads. For intersections with sight-distance restrictions, the standard solution is to reduce vehicle speed. However, it is interesting that, for some obstruction locations, it is necessary to increase the speed to satisfy sight-distance needs. An analytical method and a worksheet for finding the required (reduced or increased) speed are presented.

Existing symmetrical and unsymmetrical (two-arc) vertical curves may not be feasible when the highway alignment is subject to physical constraints (e.g., vertical clearance) and the beginning and end of the vertical curve are fixed. This paper presents a new three-arc vertical curve that accommodates such constraints. The unsymmetrical three-arc curve connects two tangents with unequal lengths and consists of three parabolic arcs smoothly connected at the points of common curvature. Since there are generally many three-arc curves that satisfy the constraints, the optimal three-arc curve that maximizes the smoothness of the curve was found. This was achieved by minimizing the sum of the absolute differences in the rates of change of grades at the two points of common curvature using an exhaustive search. Existing traditional and equal-arc unsymmetrical curves were found to be special cases of the three-arc curve. Another interesting special case is the three-equal-arc symmetrical curve, which is directly solved in closed-form. Sight distance characteristics of the three-arc curve were analyzed and compared with those of existing curves. It is shown that the three-arc curve improves sight distance for certain conditions. Application of the new curve is illustrated using numerical examples. The proposed vertical curve should be useful in the design of constrained vertical alignments such as complex freeway interchanges.

The Commercial Motor Vehicle Safety Act (CMVSA) of 1986 established uniform federal standards for testing and licensing to ensure the “fitness” of persons who operate commercial motor vehicles. The purpose of the research reported on here was to determine if the results of such standardized testing could be directly linked to safer commercial vehicle operation in Michigan. The relationships between Commercial Driver License (CDL) test performance, driver demographics, driver attitudes, driver knowledge, and driver safety-related history were studied. These relationships were used to examine questions such as: are alternate forms of the CDL test equivalent; are there problems with testing bias; is the CDL test “good” at differentiating between drivers with safe or unsafe driving habits; can CDL test scores be related to driver safety; and, does the CDL test penalize drivers who have good driving records, but poor verbal skills? The analysis resulted in the identification of several negative aspects of the CDL test. First, the alternate forms of the CDL are not equivalent. Second, educational, racial, and, gender biases were identified in the CDL test structure. Finally, the CDL test structure may penalize drivers with “safe” driving records, but poor verbal skills. Several positive aspects were also identified. First, the CDL study manual is aiding drivers in meeting the minimum knowledge requirements for the operation of commercial vehicles. Most importantly, it appears that, on the whole, the CDL test is an effective tool for identifying safe drivers.

Rapid suburbanization of housing and employment has produced severe traffic congestion in North American cities. One response to this problem in the greater Toronto area (GTA) has been to identify urban forms that are more supportive of public transport and require less vehicle kilometers of travel to support. The analytical tools used to assess the travel implications of different urban forms normally use travel demand parameters that are uniform across an area. This has yielded misleading estimates of travel demands. This paper describes analyses of the intraregional differences in travel behavior in the greater Toronto area. The analyses described are at two spatial scales: the suburb (municipality/planning district) level and the much finer traffic analysis zone level. The analysis units were grouped into high-growth, developing, and low-growth categories at both spatial scales. The analyses reported in this paper show that household characteristics and travel behavior are quite similar for both established and redeveloping zones in the older, stable suburbs. Significant differences in travel characteristics exist between the older, established zones and the growing zones in the developing suburbs. Household trip rates are shown to vary with household size, car ownership, and whether a household is located in a stable or growing suburb. Accessibility to public transport is shown to affect trip behavior differently in growing areas than in established areas. The paper concludes by discussing the public policy and transport systems analysis implications of the results.

Simulation is becoming an essential tool for planning, design, and management of airport facilities. A simulation of aircraft at gates at an airport can be applied for various periodically performed applications, relating to the dynamic behavior of aircraft at gates in airport terminals for analyses, evaluations, and decision supports. Conventionally, such simulations are implemented using an event-driven method. For a more efficient simulation, this paper proposes a network-based method. The basic idea is to transform all the sequence and constraint relations of aircraft at gates into a network. The simulation is done by calculating the longest path to all the nodes in the network. The effect of the algorithm of the proposed method has been examined by experiments, and the superiority of the proposed method over the event-driven method is revealed through comprehensive comparisons of their overall simulation performance.

Polyethylene is the predominant pipe material for trickle irrigation laterals and submains because of its flexibility and resistibility to the sun. Laboratory experiments have been conducted to determine the proper values of the Hazen-Williams friction coefficient for commonly used polyethylene plain pipes of different diameters. The commonly recommended values of the Hazen-Williams coefficient range from 130 to 140 regardless of the pipe diameter. Results of this investigation indicate that the average Hazen-Williams coefficients are 129, 136, 144, and 148 for pipe diameters of 13, 16, 19, and 22 mm, respectively. The differences between the measured friction losses and those calculated based on the average Hazen-Williams coefficients are small and vary from 8 to 14% at pipe diameters of 13 and 22 mm, respectively. The differences in calculating friction losses based on the commonly recommend values of the Hazen-Williams coefficient (130 and 140) and those based on the average coefficients reach 27% at a pipe diameter of 22 mm.

Purposely or accidentally detonating explosives near pressurized gas pipelines may have severe consequences on them, ranging from permanent deformation to gas loss or even violent rupture. Owing to lack of analytical prediction models, a shortcut is proposed in this article by which conservative estimations can be obtained of the safety distance of a pipeline from an explosion source, when explosive quantity is known. Alternatively, the maximum allowable explosive quantity can be calculated, provided distance is given. If both explosive quantity and distance are given, stresses on the pipeline may be estimated. Experimentally determined ratios of circumferential to longitudinal strains found in the literature are used. This ratio depends on soil peak particle velocity in contact with the pipeline. For total stress estimations on the pipeline, either biaxial or Huber-Hencky-Mises stresses are used, the latter being more conservative. Validation of proposed method is obtained against experimental observations and results. In this prediction model the characteristics of the explosion source, of the soil, and of the pipeline are taken into account.