A signal timing policy describes operational goals applied when developing a timing plan for a system of coordinated intersections. This research investigates differences in signal timing plans resulting from implementation of five basic signal timing policies (including minimizing delay, minimizing stops, minimizing fuel consumption, maximizing coordination bandwidth, and a “baseline” policy). Further, this research investigates differences arising from application of different techniques to achieve the same policy goal (e.g., different techniques to maximize coordination bandwidth). Volume and other traffic signal data from four traffic signal systems are collected. Various computer tools including PASSER II-90 and TRANSYT-7F are used to develop signal timing plans that emphasize each policy. For the maximum bandwidth policies, several techniques for achieving the policy goal are applied. In all cases, the resulting signal timing plans are transferred to TRANSYT-7F for comparative simulation of detailed systemwide and route-specific measures of effectiveness. Review of these measures of effectiveness provides insights into the relative success of each timing policy, and the impacts of each policy on route-specific and overall system performance. Based on these insights, general guidelines for application of signal timing tools are proposed.

The classification of traffic flow states in China has traditionally been based on the Highway capacity manual, published in the United States. Because traffic conditions are generally different from country to country, though, it is important to develop a practical and useful classification method applicable to Chinese highway traffic. In view of the difficulty and complexity of a mathematical and physical realization, modern pattern recognition methods are considered practical in fulfilling this goal. This study applies a self-organizing neural network pattern recognition method to classify highway traffic states into some distinctive cluster centers. A small scale test with actual data is conducted, and the method is found to be potentially applicable in practice.

FRESIM is a microscopic time-stepping simulation model for freeway corridor traffic operations. To enable FRESIM to realistically simulate expressway traffic flow in Singapore, parameters that govern the movement of vehicles needed to be recalibrated for local traffic conditions. This paper presents the application of a genetic algorithm as an optimization method for finding a suitable combination of FRESIM parameter values. The calibration is based on field data collected on weekdays over a 5.8 km segment of the Ayer Rajar Expressway. Independent calibrations have been made for evening peak and midday off-peak traffic. The results show that the genetic algorithm is able to search for two sets of parameter values that enable FRESIM to produce 30-s loop-detector volume and speed (averaged across all lanes) closely matching the field data under two different traffic conditions. The two sets of parameter values are found to produce a consistently good match for data collected in different days.

The maximum permitted length of trucks in North America and other parts of the world has increased significantly over the past 20 years. This has led to concern regarding the interaction of trucks with other elements in the traffic stream. There are many highway infrastructure design criteria that need to be reassessed in light of recent evidence on the behavior and properties of trucks. One of them is the passing sight distance (PSD), which is crucial for a safe overtaking maneuver. In addition, to overcome the increased congestion of modern roads, the development of the Intelligent Transportation System also requires an on-line overtaking model, which is able to provide the desired trajectory for the control. This paper presents a mathematical model that enables the determination of the safe PSD and calculates the desired trajectory for overtaking on two-lane highways. In addition to vehicle length, 10 other parameters that have an effect on the PSD are also included in this model. Specifically, the effects of various input parameters have been examined. Simulation results from this model are compared with those suggested by the American Association of State Highway and Transportation Officials (AASHTO) to test the validity of the model. Also, we explain why the standards given in the AASHTO study of 1994 could be thought to be conservative.

Road and railway transition curves have again become the subject of important scientific research and serious traffic engineering analyses because of high driving speeds and demands for automatic drive (i.e., car pilots). This article shows, in an original way, how to define or reconstruct the track of a vehicle when passing the elements of a continuous curve (straight lines, circles) so the track (curve) suits all requirements to which transition curves must be adapted. The praxis whereby a known mathematical curve (e.g., a cubic parabola, lemniscate, or chlotoid) was assumed as a transition curve and its suitability was analyzed has been passed over. On the basis of assumed kinematics models of the motion of a vehicle along joint alignment elements with a changing radius of curvature, the writers have analyzed different transition curves resulting in safe, comfortable, and economic driving. The curve resulting from a parabolic velocity chart of front wheel rotation during such movement which found the most suitable has been named POLUŠA; a geometrical analysis of POLUŠA has been performed and a manual for practical use completed.

This paper presents the formulation and the application of a continuum-based finite-layer approach to evaluate pavement strain response under actual traffic loading. The model incorporates important pavement response parameters such as the dynamic tire-pavement load variations and corresponding complex contact stress distributions (normal and shear), vehicle speed, and viscoelastic material characterization. Results of a parametric study in which two typical thin and thick pavement sections were subjected to traffic loading moving at different speeds are included. As an important application of the proposed method, pavement strain responses generated by wide-base and dual tires have been presented. This study reveals that contradictory to the past studies, the impact of tire-pavement contact shear stress on tensile strain at the bottom of asphalt concrete is insignificant.

This paper presents a new procedure for back-calculating the properties of a concrete pavement with two slab layers using deflection measurements from nondestructive deflection testing. The two slab layers are first represented by an equivalent single slab layer based on the principle of equivalent flexural rigidity. In the back-calculation analysis, the pavement deflection response under load is evaluated using an analytical model for a slab supported on an elastic foundation. Employing a closed-form back-calculation algorithm, it is shown that the back-calculated equivalent radius of relative stiffness and subgrade modulus are unique and independent of the characteristics of the two slab layers. It follows that the equivalent flexural rigidity for the two slab layers is also unique. The next step employs a trial-and-error algorithm to back-derive the moduli of the two slab layers by matching the equivalent flexural rigidity. An upper bound and a lower bound to the modulus can be established for the slab layer with the higher modulus. Similarly, an upper bound to the modulus can be found for the slab layer with the lower modulus. Four numerical examples are presented to demonstrate the use of the proposed back-calculation method.

A method of analysis using fully deformable blocks is proposed for modeling unbound granular layers used in flexible pavements. An assemblage of discrete blocks of aggregates is employed to approximate the load transfer mechanisms of the real particulate nature of granular materials. The unbound aggregate bases were modeled as particulate media composed of blocks of aggregates, which are able to transfer both shear and normal compressive loads through the interfaces. Using a finite-element approach, the discrete blocks are modeled as single quadrilateral elements connected only by surrounding interface elements at each block interface. Realistic granular layer particle characteristics including translation, sliding, and even separation are permitted for each block. An iterative procedure for equilibrium is employed in the model, and the constitutive relations used in the analysis are described for determining the behavior modes of the interface elements. The granular base discrete block representation is then used in a conventional flexible pavement section having other layers modeled as continua. Important findings are presented related to the shear resistance of the granular bases.

In order to realize the thickness design of flexible pavements at a given probability level, supplementary analysis of the generalized CBR equation was carried out. Investigation includes verification of the equation parameters, analysis of the model transformation, and determination of the corresponding confidence intervals. Suggested approximations of the constructed confidence limits for the generalized CBR equation curve enable the implementing of the thickness design for flexible pavements, especially airfields, with a given probability level. For a practical application of the probabilistic approach, a computer program was prepared. An example of thickness design for the flexible pavement at a given probability level is presented. The results of the work provide controlled employment of the generalized CBR equation, thus increasing the reliability of the predictions and conclusions in the design and evaluation of flexible pavements.

A survey of existing underground heat distribution systems was conducted to compare the failure characteristics of steel conduits with fiberglass reinforced plastic conduits. Pressure testing was carried out in order to set an objective standard for performance. Plastic conduits were found to have a failure rate 4–5 times as high as steel conduits. The implications of this for a comparative life-cycle cost analysis are very serious.

An analytical model was developed to compare the life-cycle costs of two types of underground heat distribution systems: directly buried conduits and concrete trench installations. The costs of initial construction, routine maintenance, operation, repair, and replacement are represented in the model. Comparisons are made to explore the influence of failure rates, heat loss assumptions, the discount rate, and the policy on partial section replacement. Failure rate estimates are obtained from the literature and also from survey results obtained by the U.S. Army. Life-cycle cost computations indicate that, when initial construction costs are the same, concrete trench installations have a cost advantage of $18–$42 per foot over buried steel conduit when other cost components are taken into account over a 25-year lifetime. The cost advantage of concrete trench over fiberglass reinforcement plastic (FRP) buried systems is $32–$133 per foot. The variation in the cost advantage reflects the variation in the estimate of the failure rate for buried steel and FRP systems. Routine maintenance was crucial for the buried systems to avoid dramatic increases in operation and replacement costs that would make the cost advantage of concrete trench installation even greater than stated above. In contrast, the low incidence of failure associated with concrete trench systems made their life-cycle costs relatively insensitive to maintenance policy.

A system to identify all left turns, through movements, and right turns at a signalized intersection is described. The complete system requires detection of vehicle departures from the intersection, right turn detection, and information on present phasing from the signal controller. The system is demonstrated with videotape data for one departure leg of a signalized intersection. Actual vehicle movements are compared to both a manual application of the algorithm and an application using a wide area detection system (WADS). The manual application indicated that the algorithm should correctly reflect vehicle movements. The WADS application also supported the feasibility of the system. The information from such a system could be used to improve the information components of real-time traffic adaptive signal control systems, dynamic traffic assignment, and estimates of origin-destination matrices.