In 1964, Carey and Irick defined the present serviceability index (PSI) by assuming that the PSI can be expressed on a scale of 0 to 5 as the sum of a logarithmic function of slope variance, a quadratic function of mean rut depth, a square root function of cracks and patches, and a random error term using the American Association of State Highway Officials (AASHO) road test data. These functional forms have been used widely although they yield intercept values for PSI over 5 and some other nonphysical results for both rigid and flexible pavements; similar difficulties appear in the work of others performed over the years since then. To resolve these difficulties, we propose a new methodology dealing with the test data by applying Fechner's psychophysical law. It is found that the AASHO road test data, road test data obtained in Texas, road test data from Canada, and data from the international road roughness experiment are organized well by the simple summation of logarithmic terms of roadway characteristic variables. The methodology can be further applied to handle similar types of road test data.

In the face of an aging road infrastructure and shrinking highway maintenance budgets, optimizing the allocation of limited highway resources is paramount to the Saskatchewan Department of Highways and Transportation (SHT). In order to make informed infrastructure management decisions, SHT has identified the need to accurately quantify vehicle operating costs on different roads in diverse operating conditions. SHT had previous experience with the HDM III model, but found they could not support HDM III because of the extensive data requirements. As a result, in 1991 SHT initiated a review of available road user cost models in the hope of identifying a model that would meet the specific needs of SHT. No such model was found primarily because accurately predicting vehicle operating costs on most Saskatchewan roads is difficult because of the inherent uncertainty associated with the parameters that influence vehicle operating costs. Therefore, a mechanistic-probabilistic vehicle operating cost (PVOC) model was developed to estimate vehicle operating costs of different vehicles operating on different roads in Saskatchewan conditions. The PVOC model computes a probabilistic estimate of vehicle operating costs, including the expected value and the associated uncertainty of the estimate. The PVOC model was validated for several vehicle types operating on Saskatchewan roads and was found to be within 20% of published industry averages and within 9% of historic Saskatchewan truck fleet data.

Trip distribution analysis involves determination of the pattern of trip interchanges between land-use parcels or traffic zones using an appropriate analytical technique. The different analytical techniques available for trip distribution can be broadly grouped into two types: growth-factor methods and synthetic methods. The gravity model is one of the commonly used synthetic models for trip distribution. This paper deals with the estimation of travel deterrence for trips made for different purposes and by different modes through calibration of two types of gravity model formulations, using the data set pertaining to a medium-size city with a complex travel environment in India. Stratified data of home-based trip distribution for three trip purposes, and trips made by five different modes, were used for calibrating the model formulations. Models of type I formulation were calibrated by the cell-by-cell comparison procedure and those of type II formulation as per the Bureau of Public Roads (BPR), United States of America method. The parameter values obtained for the models of both formulations were found to be logical and statistically significant. Modeling of the stratified trip distribution, apart from yielding models of better predictive power, has provided insight into the relative effect of travel deterrence on trips made for different purposes and by different modes.

The objective of this study is to determine the minimum risk route for transporting a specific hazardous material (HM) between a point of origin and a point of destination (O-D pair) in the study area which minimizes risk to population and environment. The southern part of Quebec is chosen as the study area and major cities are identified as points of origin and destination on the highway network. Three classes of HM, namely chlorine gas, liquefied petroleum gas (LPG), and sulphuric acid, are chosen. A minimum risk route model has been developed to determine minimum risk routes between an O-D pair by using population or environment risk units as link impedances. The risk units for each link are computed by taking into consideration the probability of an accident and its consequences on that link. The results show that between the same O-D pair, the minimum risk routes are different for various HM. The concept of risk dissipation from origin to destination on the minimum risk route has been developed and dissipation curves are included.

The American Association of State Highway Officials (AASHO) road test, conducted during the period of 1958 through 1960, was a factorial test of pavement durability that considered layer depths, axle load, and number of load applications as the primary variables. These data were processed using traditional statistical techniques. The AASHO formula is the resulting databased model of the road-test data. In the present paper, we reexamine the AASHO road-test data, using the Monte Carlo Hierarchical Adaptive Random Partitioning (MC-HARP) neural-network model developed by Banan and Hjelmstad (1995), and show that an MC-HARP model can represent the data far better than the AASHO formula can. We conclude that the MC-HARP neural network may be an appropriate tool for the development of databased models of pavement performance in the future.

Dynamic response of concrete pavements subjected to moving loads is analyzed by using the three-dimensional (3D) finite-element method in conjunction with Newmark integration scheme. The dynamic vehicle-pavement-foundation interaction effects are considered in the 3D finite-element algorithm. The moving vehicle loads are modeled as lumped masses each supported by a spring-dashpot suspension system and having a specified horizontal velocity and acceleration. Concrete pavements are considered to respond elastically and are represented by a series of brick elements. The present formulation considers a linear viscoelastic foundation model (Kelvin model) consisting of a system of discrete linear springs and dashpots. The interaction between concrete pavements and underlying soil foundation is considered. The present results are compared with the available theoretical results to validate the accuracy of the present algorithm. A study is conducted to evaluate the effects of various parameters on the dynamic response of concrete pavements subjected to moving loads.

Thermal cracking of asphalt pavements is a serious problem in Canada and the northern parts of the United States. On many occasions, either at the design stage or during service, highway agencies demand a forecast of pavement performance rating, which is highly sensitive to the intensity of cracking. There are two failure modes of thermal cracking: low-temperature cracking and thermal-fatigue cracking. Low-temperature cracking is caused by accumulated thermal stresses in the pavement layer during cold winters or spring thaws. Thermal-fatigue cracking is caused by daily cyclic thermal loading. Classical probabilistic approaches have been applied to pavement design systems, including thermal cracking predictions, during the past three decades. Advances in reliability analysis, however, have proven that classical reliability methods are inconsistent, and mandate that current design procedures should be revised accordingly. This paper presents an improved reliability model for predicting thermal cracking. The proposed model accounts for the variability in the component design variables and the correlation between the two failure modes. The model results were verified using Monte Carlo simulation, and the sensitivity of the predicted intensity of cracking to various design variables was examined.

A construction method for rigid pavement utilizing waste plastic for forms to create cells is analyzed for stress resulting from application of a standard highway equivalent single axle load (ESAL). The finite-element method is used. The analysis parametrically studied varying slab, base, subbase, and subgrade thickness and elastic properties and two different tire pressures. Stresses are compared to those in a solid slab under the same loading conditions. Although the stresses in the cellular rigid pavement are higher than in the solid slab, the maximum principal stresses are shown to be within a range that would allow the use of this method for rigid pavement construction. Depending on requirements for load resistance and fatigue criteria, significant savings of material and cost can result. The method reduces material use and the energy required to produce it by approximately 25 and reuses waste product. The construction method is also applicable to parking areas, walks, revetments, canal linings, floor slabs-on-grade, and similar construction.

The finite-element study of the effect of temperature variation on plain-jointed concrete pavements is presented. Temperature variation causes curling and thermal-expansion stresses. Curling stresses result from temperature gradients through a slab depth. Thermal-expansion stresses are induced due to uniform changes in temperature that cause the slab to expand. The developed three-dimensional (3D) model consists of four slabs separated by longitudinal and transverse joints. The interaction between the ground and the concrete slab along with interaction at the joints were modeled using interface elements. These elements gave the model the capability to solve for partial contact between curled slabs and the ground to investigate the effect of compressive stresses that may develop at the joints during curling, and to study the influence of friction between slabs and the ground. The data obtained using the finite-element model has shown reasonable agreement with the results obtained from three computer models: KENSLABS, ILLI-SLAB, JSLAB, and the analytical solution proposed by Bradbury. The best correlation was obtained with JSLAB. The model was used to perform parametric studies on curling and thermal-expansion stresses to study the effect of superposition of both stresses and to address the effect of uniform temperature changes on joint opening. Another simpler model using nine layers across the depth of a pavement slab was used to introduce the effects of nonlinear temperature distribution. The results of the parametric studies are presented and compared with other solutions. The arithmetic addition of positive curling stresses and thermal-expansion stresses were less than those stresses obtained by superposition. In some cases, the calculated joint openings were higher than the allowable joint opening. Nonlinear temperature distribution caused higher tensile stresses than the linear distribution of temperature. The difference in tensile stresses between the two distributions was approximately 3–13 of the modulus of rupture of concrete.

Turned-down guardrail terminal sections were analyzed using LS-DYNA3D, a dynamic, nonlinear, large deformation finite-element modeling package with capabilities for simulating vehicular impacts into roadside hazards. A baseline simulation was conducted on the existing turned-down approach terminal section, as well as on various retrofit options. Simulation findings were used to identify potential design changes and to select the two most promising options. Modified designs were subjected to six low-speed and one high-speed full-scale crash tests with a minisize automobile. The crash testing verified that one of the retrofit designs would allow the W-beam rail to drop down without causing vehicle rollover when impacted end on by a minisize automobile. However, the revised design did not meet recommended safety standards when one of the weakened posts failed to break away as designed. It is believed that by revising the post installation procedures, the modified turned-down terminal section would perform satisfactorily.