America’s highways have developed dramatically over the last century, evolving into an extensive system of interconnected motorways. Lessons learned during this evolution suggest some future directions and challenges. Federal-aid highway programs played a major role in highway development. Since 1916, cooperative federal-state partnerships and funding arrangements have evolved, continually broadening in scope to address new challenges, including environmental and urban concerns. The 21st century will experience accelerating change and mobility demands. As in the past, more people driving more cars and occupying more land will necessitate better roads and improved public transport. While land use and traffic management policies might reduce demands for highways, they will not substitute for needed investments. While there will be more high-occupancy vehicle toll (HOT) lanes, it will still be necessary to complete missing road links and to alleviate major bottlenecks. Good land management will be needed to maintain environmental quality and control suburban congestion. Limiting strip commercial development, establishing urban growth boundaries, and keying future office development to areas with good transit service will be important. Maintaining future livability and mobility will call for new forms of metropolitan governance, greater public support of transportation, and strong transportation leadership.

Increasing flight frequency on airline routes tends to reduce user delay costs but increase airline operating costs. Flight frequency should vary as demand intensity changes at various times of day. Models are presented here for optimizing the departure times of flights on single airline routes with time-dependent demand. An analytic model is developed to find the departure times that minimize average schedule delay per passenger for given flight frequencies and to identify the algebraic relations among the variables. Using an analytic approach, the departure times of flights that minimize airline operating cost and users’ waiting cost are determined. To minimize the objective function, it is found that the headways should be inversely proportional to the square root of demand intensity near the departure times. Computer models for both cost minimization and profit maximization are developed to solve the problems with relaxed assumptions and verify the numerical results obtained analytically.

Millions of Americans buy light truck vehicles (LTVs) each year to meet family, business, and personal travel needs. The main objective of this research is to study the effect of the increased percentage of LTVs in traffic on fatalities by manner of collision (rear-end), and also to address the impact of crash configuration (car–car, car–LTV, LTV–car, and LTV-to-LTV). This paper presents time series models that incorporate the percentage of LTVs in traffic to analyze and forecast future trends of fatality that result from rear-end collisions. A time series model was estimated to predict future LTV percentages. Future forecasts using the calibrated time series model show that the LTV percentage is expected to increase and reach 45% of the traffic stream in the United States by the year 2010. The crash analysis is based on the fatality analysis reporting system crash database covering the period of 1975–2000. A transfer function time series model with the LTV percentage as an input and annual deaths from rear-end crashes as output indicates that the annual deaths in passenger vehicles involved in rear-end collisions will be 1,004 by the year 2010 (a 5% increase compared to that of the year 2000). The analysis also showed an expected increase in the fatalities of certain configurations based on the type of vehicles involved in the crash, and indicating possible problems if the lead vehicle is an LTV and the following vehicle is a regular car.

The market share of light truck vehicles (LTVs) is increasing rapidly, causing a change in traffic composition in the United States. LTVs include vans, minivans, light duty trucks, and sport-utility vehicles. Users of such vehicles appreciate the extra size, utility, and safety provided. Concerns about the effects of these LTVs on other passenger cars when they both collide are increasing. This paper investigates the effect of the increased percentage of LTVs in traffic on fatalities that result from head-on collisions. It also addresses the impact of crash configuration (car-car, car/LTV, and LTV-to-LTV). Time series models that incorporate the percentage of LTVs in traffic are used to analyze and forecast the future fatality trends that result from head-on collisions. The analysis is based on the fatality analysis reporting system crash database covering the period of 1975–2000. Forecasts from the fitted time series model of head-on collisions showed that during the next ten years, annual deaths in head-on collisions will reach 5,325 by the year 2010, which represents an increase of 8% over the year 2000 figure.

In this paper a comprehensive microscopic model of driver behavior in uninterrupted traffic flow is developed. The model aims to explain or predict the actions of a driver in a variety of driving scenarios ranging from free flow conditions on wide roads to forced flow conditions on narrow two-way roads. The salient features of the proposed theory are (1) a single model (or framework) is used to describe driver behavior in different driving scenarios and (2) the response of drivers through both steering control and speed control are modeled. Driver behavior is simulated in 19 different driving scenarios. Results from these simulations show the efficacy of the proposed model in explaining/predicting driver behavior.

This paper presents an approach to optimize the short-term traffic prediction performance on freeways using multiple artificial neural network topologies under different network and traffic condition settings. The approach encourages multimodel techniques that are capable of improving the prediction system performance over single-model approaches. Using a mix of neural network topologies, the short-term speed prediction performance was extensively evaluated under different input settings and various prediction horizons (from 5 to 20 min). To enable the networks to learn from historical information, a long-term memory component was introduced to the input patterns to allow the networks to build internal representation of recurrent conditions, in addition to the short-term memory that is encoded in the most recent information. Optimal settings were determined by maximizing the performance under different traffic conditions observed at the target location, as well as upstream and downstream locations. Comparative statistical analysis with naive and heuristic approaches showed that the optimized neural network approach resulted in better prediction performance. The study shows that the optimal settings were consistently more dependent on the long-term memory component as prediction horizon increases.

Recently, a dynamic traffic control system designated as dynamic intersection signal control optimization (DISCO) has been developed. DISCO considers the entire fundamental diagram of traffic flow, which is essential for controlling congested and transient traffic. As a dynamic model, DISCO works with time-variant traffic patterns and derives dynamic timing plans. In this study, we applied DISCO to a congested network in Hong Kong and studied the various control strategies enabled by DISCO. The results showed that DISCO outperformed the existing plans by 30–40% in overall delay reductions. DISCO performed best under the most flexible control scheme of variable-green no-cycle strategy, given a good initial solution for the algorithm. Nevertheless, even good fixed-time plans can produce remarkable results.

This paper investigates the relationship between the as-built roughness of asphaltic concrete pavement and construction characteristics factors using panel (longitudinal) data analysis. This panel study utilizes as-built roughness measurements and pavement and contractor’s characteristics for reconstructed, replaced, and resurfaced pavement projects in Wisconsin from 1996 through 2000. Several panel models, both fixed and random effects models, were developed from the basic model proposed in the study. The research shows that the random effects model provides the best estimate of proposed models on the basis of Akaike’s information criterion (AIC). The results indicate that some pavement and construction characteristics significantly affect as-built pavement roughness. The findings also reveal that contractor’s quality history from past pavement projects is a statistically significant factor that affects as-built roughness and can possibly be converted to contractor’s accumulated performance indicator on pavement construction.

In laboratory testing of asphalt mixtures, the relationship between the number of load repetitions and permanent deformation has been found to include three distinct stages, namely the primary, secondary and tertiary stages. Similar permanent deformation behavior has also been observed in the field accelerated pavement testing data. It is argued that the three-stage permanent deformation behavior is a basic material property and to comprehensively model asphalt layer behavior, it is necessary to develop a model that accurately characterizes this behavior. In this paper, each of the well-known models critically reviewed appears that they are limited to characterizing only the primary stage. Therefore, a new three-stage model is proposed to describe all three stages. Moreover, an algorithm is established to determine the model parameters from typical laboratory data. The algorithm can also be used to identify the transition point between stages, such as flow number. The proposed model and algorithm are demonstrated through laboratory test results. The analysis results match the field performance.

The present study proposes a method for predicting stresses in pavements under vehicular loadings. The proposed method characterizes each pavement layer by a coefficient of lateral stress, which is similar to the commonly used coefficient of lateral earth pressure (K). Stress predictions found by the proposed method are compared with stress predictions calculated by conventional linear elastic techniques. Each conventional solution is shown to represent a special case of the more global proposed solution. Several instrumented test sections are explored to determine coefficients of lateral stress for common paving materials.

This paper presents the results of numerical simulations of the resilient modulus test used to mechanically characterize the resilient behavior of aggregate materials, commonly used in pavement bases and subbases. The investigation made use of the discrete-element method (DEM) to replicate the particle behavior usually experienced during laboratory sample preparation and testing. The simulations were based on assemblies of circular particles confined between top and bottom rigid boundaries and laterally confined at constant stress. Contact forces and displacements were assumed to obey a linear relationship and shear forces were bounded by a maximum value (Coulomb friction law). Compacted samples were subjected to deviator repeated loads. The investigation showed that the DEM is capable of reproducing the results of the resilient modulus test performed on real granular materials in a qualitative manner. Further, the method predicted the effect of the state of stress depicted by laboratory testing.

In the 1970s, the U.K. Transport Research Laboratory published a series of reports dealing with wearing course slippage due to debonding. At present, however, no accepted nondestructive test (NDT) technique exists for a reliable detection of poor bond. This paper outlines the research undertaken in developing an NDT method for quantifying bond condition beneath thin surfacings (10–50 mm). Laboratory experiments were conducted in the Pavement Test Facility at Nottingham University. A generic type of surfacing material was used and several interface conditions were compared: (1) bonded; (2) debonded; and (3) partially bonded. Different surfacing thicknesses were considered. An instrumented impulse hammer was used to provide the transient excitation, and vertical dynamic responses close to the point of impact were monitored using accelerometers. The results were used to identify suitable parameters for identification of bond condition. Cores taken at test points showed good agreement between the Leutner test results and the predicted condition of the interface. Finite-element (FE) analysis has also been performed to provide understanding of the response of thin surfacings with different degrees of bond. Results from the FE analysis confirm the trends established from the testing.

Dramatic increases in the number and weight of vehicles have resulted in severe rutting on highways in Taiwan. The Taiwan Area National Freeway Bureau constructed an in-service test road in 1996 to investigate the effect of pavement structures and paving materials on pavement performance. A mechanistic-empirical rutting model was developed to account for material property changes in the progression of rutting over time and the cumulated compressive plastic strains of all layers on the basis of traffic loading. In addition, a three-dimensional, dynamic finite-element simulation was established to calculate pavement responses that were input into the rutting model for predicting permanent deformation. Traffic speed showed significant impact on vertical strains at the top of the surface layer, and its influence was consistent with those measured by other researchers. The rutting parameters in the model were calibrated using rut depths obtained from the field test, and the validation outcome showed that their predictions were in good agreement with measured ones. The results of this paper demonstrate that incorporating the mechanistic-empirical approach in the prediction of rutting is viable and provides valuable information on the contribution of each layer to permanent deformation in flexible pavements.

The performance of both flexible and rigid pavement is a function of the constituent materials and their susceptibility to physical and chemical stress and degradation. To that end, specific phenomena that accompany and characterize the interaction of the soil-pavement system with the environment are discussed. Mechanisms of pollution intrusion through the pavement wearing surface, base, subbase, and subgrade soil layer are evaluated. Liquid movement phenomena and processes in the soil mass such as infiltration, percolation, and hydraulic conduction related to the soil-pavement system are examined. Laboratory data that reflect the effects of pollution intrusion on the behavior, stability, and performance of soil-pavement systems are presented and remedial actions on these effects are discussed. Pollution intrusion is manifested by changes in system variables, including solution pH, dielectric constant, ionic composition, temperature, and moisture content. These changes are shown to influence pavement design parameters, such as soil physical properties, compressibility, and stress.

Calculation of the hydraulic gradient due to the upward flow of a large size particles–water mixture in a swaying vertical pipe is a central problem in the design of systems for deep-sea mining of manganese modules. Here, the problem is investigated experimentally and with a mathematical model. An experimental apparatus that mimics the deep-sea mining system was built to investigate the effects of swaying on the hydraulic gradient. The experimental results reveal that the hydraulic gradient in a swaying vertical pipe is greater than in a fixed vertical pipe (for the same flow and solid-loading conditions) as a result of the instantaneous particle-wall collisions and the solids sliding over the pipe wall. A mathematical model for the mixture flows is suggested based on a Lagrangian simulation technique. The model results offer additional insight into the dynamics of this complex flow and agree well with the new laboratory data.