In the literature, the length of a clothoid spiral usually is determined by three equations. These equations do not lead to the same length. As a result, all equations are applied and the maximum length is chosen. Of these three equations, two are based on actual and unbalanced superelevation with speed. These two equations are theoretically inaccurate. Thus, after selecting a spiral length, designers need to check the length against comfort criteria, even though these two equations are based on the same comfort criteria. Then, the length is adjusted. In fact, all parameters, such as jerk ($\mathrm{g}/\mathrm{s}$), actual superelevation runoff ($\mathrm{mm}/\mathrm{s}$), unbalanced superelevation runoff ($\mathrm{mm}/\mathrm{s}$), rotational runoff ($\mathrm{deg}/\mathrm{s}$), cant gradient, and twist ($\mathrm{mm}/\min$), are interrelated. Jerk criteria can fix the spiral length. In this paper, the writer presents spiral length equations that lead to equal spiral length by exploiting the interrelation between actual and unbalanced superelevation runoffs. With this process, the roll, superelevation gradient, and twist are internalized and designers need not recheck the comfort criteria. A single equation would replace the need for many, saving design time. This paper is a revised, improved version of an earlier, peer-reviewed conference paper by the writer.

A new methodology is introduced for project selection that explicitly addresses issues of achieving maximized overall project benefits by selecting a subcollection of candidate projects for possible implementation at a given budget level while controlling the total risk of the expected project benefits within an acceptable lower bound. The covariance value is utilized to denote the risk of the expected benefits of jointly implementing two projects, in which each project maintains a range of possible benefits and a probability distribution. The summation of all covariance values corresponding to all possible project implementation combinations represents the total risk of the expected benefits. The methodology contains two-step optimizations. First, the Markowitz mean-variance model is employed to establish the lower-bound risk of project benefits for a given budget level. Second, the conventionally accepted zero/one knapsack model for project selection is augmented to incorporate the lower-bound risk established from the first-step optimization as one additional chance constraint. In this way, the overall benefits of projects selected for implementation are maximized while controlling the total risk of the expected project benefits within a lower bound for a certain percentage of time. The proposed model is applied for a 6-year statewide interstate highway project selection and programming. Cross comparisons are made in the consistencies of project selection results generated from the basic knapsack model, proposed two-step enhanced knapsack model, and current state highway programming practice.

An iterative method is proposed to estimate bus route origin-destination (OD) passenger flow matrices from boarding and alighting data for time-of-day periods in the absence of good a priori estimates of the flows. The algorithm is based on the widely used iterative proportional fitting (IPF) method and takes advantage of the large quantities of boarding and alighting data that are routinely collected by transit agencies using automatic passenger count (APC) technologies. An arbitrarily chosen OD matrix can be used as the base matrix required to initialize the algorithm, and the IPF method is applied with bus trip-level boarding and alighting data and the base matrix to produce an estimate of the OD flow matrix for each bus trip. The trip-level OD flow matrices are then aggregated to produce an estimate of the period-level OD flow matrix, which in turn is used as the base matrix for the following iteration. The process is repeated until convergence. Empirical results are conducted on operational bus routes using APC data collected during multiple season-years, where directly observed OD passenger flows are available to represent the ground truth. In all cases in which APC data are available for even a reasonably small number of bus trips, the iteratively improved base method produces better estimates than the application of the traditional IPF method when using a null base matrix, which is commonly adopted in the absence of a priori information without updating. Moreover, the algorithm converges in minimal computational time to the same estimates regardless of the initializing matrices used.

This study introduces a new methodology for signal timing optimization that is carried out by adjusting green splits of a.m. peak, p.m. peak, and rest of the day timing plans for each signalized intersection in the urban street network without changing the existing cycle length and signal coordination to minimize total vehicle and pedestrian delays per cycle. It contains a basic model that handles vehicle delays only and an enhanced model that simultaneously addresses vehicle and pedestrian delays using two different pedestrian delay estimation methods. Both models are incorporated into a high fidelity simulation-based regional travel demand forecasting model for detailed traffic assignments. A computational study is performed for methodology application using data on Chicago metropolitan area travel demand, traffic counts, geometric designs, and signal timing plans for major intersections in the Chicago central business district (CBD) area. A sensitivity analysis is conducted in the application of the enhanced model to examine the impacts of assigning different weights to vehicle and pedestrian delays on intersection vehicle travel time and delay reductions after signal timing optimization. The computational experiment reveals that after systemwide signal timing optimization, vehicle delays in the CBD area could reduce by 13% when considering only vehicle delays and by 5% when simultaneously considering vehicle and pedestrian delays.

As one of the most popular unconventional interchange designs, diverging diamond intersection (DDI) has received increased attention over the past decade. Through a reverse operation of traffic movements between its two crossover intersections, DDI can accommodate more traffic movements within each phase. To design an effective signal plan for DDIs, one needs to address the following two critical issues: (1) how to select the common cycle length and green splits at each crossover intersection under different geometric conditions, and (2) how to coordinate a DDI’s two crossover intersections with its adjacent conventional intersections. To contend with these issues, this paper presents an optimization model with the objective of maximizing intersection capacity to yield the optimal green splits and cycle length. Also, in view of the potentially large left-turn traffic volumes from the freeway off-ramps, this study has further modified a model to provide progressions to both left-turn and through traffic paths. Using simulation software as an unbiased tool, this study has conducted extensive simulation comparisons between the optimized signal plans and the results from signal optimization software under various traffic scenarios. The experimental results confirm the promising properties of the proposed signal models for DDI, especially if the traffic progression between two crossover intersections is the major concern.

Recently observed increases in the numbers of permitted and illegal overweight trucks travelling over U.S. highways have raised concerns over their contributions to the reduction in the service lives of pavements and bridges and the costs of maintaining, upgrading and replacing the highway infrastructure system. Cost allocation studies are used by transportation officials to help their asset management processes and to establish truck traffic and permitting policies taking into consideration information on the composition of overweight trucks and their permit categorization. In recent years, cost allocation studies have heavily relied on data assembled by weigh-in-motion (WIM) systems which provide information on traffic counts, truck axle configurations and weights for various highway classes and economic regions. However, WIM data by themselves do not provide information on the numbers of illegal overweight trucks because many of the overweight trucks may have been issued permits that allow them to operate on a yearly basis or on a trip-by-trip basis. The object of this paper is to develop a data mining procedure to identify and classify overweight vehicles whose axle weights were collected by WIM systems into different permit and illegal categories. The first step of the proposed data mining procedure establishes a set of rules that are satisfied by different types of permit trucks. These rules are inferred from a review of available permit databases. In a second step, a search algorithm is used to check each vehicle in a WIM database and identify whether it violates any of a jurisdiction’s legal weight limits. Subsequently, every overweight truck’s axle weights, axle spacings, total length and gross vehicle weight are checked to verify whether these characteristics match any of the criteria established when setting up the data mining rules. The overweight vehicles are separated based on the likelihood of having been issued a particular type of permit or if they are potentially illegal. The validity of the algorithm is demonstrated by analyzing truck data collected at a WIM site in upstate New York. The algorithm’s output showed reasonable agreement when compared to the results of a truck survey performed by the New York State Department of Transportation (NYSDOT). A parametric analysis is executed to assess the sensitivity of the results to the level of accuracy of the WIM system. It is concluded that the algorithm produces statistically robust overweight truck and permit categorizations which can eventually help highway agencies establish rational permit issuance policies, weight enforcement strategies, and cost allocation studies.

Heteroscedasticity modeling in transportation engineering is primarily conducted in short-term traffic condition forecasting to generate time varying prediction intervals around the point forecasts through quantitatively predicting the conditional variance of traffic condition series. Until recently, the generalized autoregressive conditional heteroscedasticity (GARCH) model and the stochastic volatility model have been two major approaches adopted from the field of financial time series analysis for traffic heteroscedasticity modeling. In this paper, recognizing the pronounced seasonal pattern in traffic condition data, a simple seasonal adjustment approach is explored for modeling seasonal heteroscedasticity in traffic-flow series, and four types of seasonal adjustment factors are proposed with respect to daily or weekly patterns. Using real-world traffic-flow data collected from highway systems in the United Kingdom and the United States, the proposed seasonal adjustment approach is implemented and validated. Empirical results show that the proposed model can effectively capture and hence model the seasonal heteroscedasticity in traffic-flow series. In addition, through a comparison with the conventional GARCH model, the proposed approach is shown to consistently generate improved performances in terms of prediction interval construction. Potential applications are discussed to explore the value of heteroscedasticity modeling in transportation engineering studies.

In recent years, a variety of tools have been developed to assist in the analysis of flexible pavement service life in the United States. Although the output from many of the approaches is similar, each tool varies in the amount of testing effort, materials, and equipment necessary to arrive at the resulting predictions. Building upon the findings from research conducted in the National Cooperative Highway Research Program (NCHRP) Project 9-22B, this study focused on comparing flexible pavement performance predictions using three emerging analysis tools. One of the pavement analysis software packages, which considered the standard of, predicts performance in terms of incremental distresses and damage accumulation during a pavement’s service life. The other two programs are performance-related specification (PRS) tools that predict pavement performance in terms of service life factors. Three different asphalt mixtures were evaluated including a conventional mixture and two unconventional mixtures modified with emerging material technologies. The conventional mixture was used in a full-depth reconstruction project that featured low-level truck traffic. The two unconventional mixtures were placed as overlays on two structures featuring heavy truck traffic and the same substructure. Results from all three programs indicated that rutting predictions were similar among different analysis tools. However, the fatigue cracking predictions showed the greatest differences due to how the different analysis tools model climatic conditions. A comparison of the different analysis tools also suggested that for the mixes evaluated, one of them could be used effectively with Level 2 hot mix asphalt (HMA) inputs, in lieu of requiring the more time-intensive Level 1 HMA inputs, to predict the service life of flexible pavements. By performing a parametric analysis, it was established that this tool was sensitive to traffic variations and changes in structural support in combination with higher traffic. Certain modifications to the volumetric properties of the HMA layers produced significantly different performance predictions using this particular tool. The results of this study showed that the three analysis tools are comparable and produce equivalent performance predictions. Thus, the decision of which analysis tool to use should be based on a transportation agency’s assessment of its testing capabilities, budget level, and project complexity. In addition, the approach developed in this study can be enhanced in the future to form the basis of a decision-making procedure to help direct the amount of effort and funding necessary in the pavement and mixture design phase.

Transportation agencies’ motor vehicle count programs tend to be well established and robust with clear guidelines to collect short-term count data, to analyze data, develop annual average daily traffic (AADT) adjustment factors, and to estimate AADT volumes. In contrast, bicycle and pedestrian traffic monitoring is an area of work for most transportation agencies. In most agencies, there are a low numbers of counting sites and limited agency experience to manage a city-wide or state-wide system of collecting, processing, and using nonmotorized data. Short duration counts are used to estimate longer duration volumes such as AADT. Because bicycle or pedestrian short-term counts vary dramatically over time and significantly more than motorized vehicle counts, the direct application of motorized vehicle AADT estimation methods may be inadequate. The goal of this paper is to present a methodology that will enhance, if needed, existing AADT estimation methods widely employed for motorized vehicle counts. The proposed methodology is based on the analysis of AADT estimation errors using regression models to estimate a correcting function that accounts for weather and activity factors. The methodology can be applied to any type of traffic with high volume variability but in this research is applied to a permanent bicycle counting station in Portland, Oregon. The results indicate that the proposed methodology is simple and useful for finding ideal short-term counting conditions and improving AADT estimation accuracy.