Rod E. Turochy

New Procedure for Detector Data Screening in Traffic Management Systems

Automated monitoring of traffic conditions in traffic management systems is of increasing importance as the sizes and complexities of these systems expand. Accurate monitoring of traffic conditions is dependent on accurate input data, yet techniques that can be used to screen data and remove erroneous records are not used in many traffic management systems. Procedures that can be used to perform quality checks on the data before their use in traffic management applications play a critical role in ensuring the proper functioning of condition-monitoring methods such as incident detection algorithms. Tests that screen traffic data can be divided into two categories: threshold value tests and tests that apply basic traffic flow theory principles. Tests that use traffic flow theory use the inherent relationships among speed, volume, and occupancy to assess data validity. In particular, a test that derives the average effective vehicle length from the observed traffic variables detects a wide range of erroneous data. A new data-screening procedure combines both threshold value tests and traffic flow theory–based tests and can serve as a valuable tool in traffic management applications.

Effect of Load Spectra on Mechanistic-Empirical Flexible Pavement Design

The transition from equivalent single-axle loads to load spectra represents a significant change in traffic characterization as many agencies begin the migration from empirically based pavement design methods to mechanistic-empirical (M-E) approaches. A key issue lies in determining appropriate representative load spectra that can be used for design. Previous studies have examined differences between load spectra and have largely found them statistically different and site-specific. This study proposes an approach to evaluating different load spectra in terms of practical effects on resulting flexible pavement thickness design. Load spectra from 12 sites in Alabama were evaluated against a statewide load distribution. Customized M-E design software featuring Monte Carlo simulation and layered elastic analysis generated pavement response distributions for a range of three-layer pavement structures. The responses were transformed through calibrated transfer functions into predictions of pavement performance...

Measuring Variability in Traffic Conditions by Using Archived Traffic Data

The predictability of transportation service is of great importance to travelers. Whereas most transportation performance measures deal more directly with congestion, such as through delay measures, few quantify the level of predictability of travel. A new measure that effectively measures predictability of transportation service, the variability index, was developed and demonstrated. The variability index is a good example of the application of data mining in large transportation databases. Conceptually based on multivariate statistical quality control (MSQC), the variability index is computed by measuring the size (spatial volume) of the confidence regions defined by MSQC by using large sets of archived traffic data. In other words, experience (archived traffic data) is mined to measure the level of variability experienced by time and location. A case study application of this procedure demonstrates how this measure can clearly identify times of day and days of the week that experience relatively high degrees of variability in traffic conditions—or less predictable service for the traveler.

Use of Data from Specific Pavement Studies Experiment 5 in the Long-Term Pavement Performance Program to Compare Virgin and Recycled Asphalt Pavements

The Specific Pavement Studies Experiment 5 (SPS-5) in the Long-Term Pavement Performance program was designed to study the effects of overlay rehabilitation type on typical distress measures. The rehabilitation treatments compared overlay thickness, overlay type, and surface preparation before rehabilitation. The thicknesses used were 50- and 125-mm overlays. The overlay types were virgin asphalt mix and recycled asphalt that contained approximately 30% reclaimed asphalt pavement (RAP). Surface preparation consisted of either milling or not milling the existing pavement before rehabilitation. Eighteen states and provinces in North America built SPS-5 projects between 1989 and 1998. Seven distress parameters from these test pavements were analyzed, including international roughness index (IRI), rutting, fatigue cracking, longitudinal cracking, transverse cracking, block cracking, and raveling. Analyses were conducted to determine which factors affected overlay performance as measured with the above parameters. Further statistical testing compared the performance of the virgin mix sections directly with equivalent sections that contained 30% RAP. Overlays with mixes that contained 30% RAP were found to perform as well as overlays with virgin mixes in terms of IRI, rutting, block cracking, and raveling. Thicker overlays improved pavement performance, except for rutting. Milling before rehabilitation decreased IRI, fatigue cracking, and transverse cracking but increased rutting.

Applying quality control to traffic condition monitoring

The problem of traffic congestion on urban freeways has led to the development of traffic management systems responsible for monitoring and responding to traffic conditions. Many systems now archive the traffic data collected at many locations throughout their coverage areas. The archived data, along with advances in computing power, make more complex condition monitoring methods feasible. Multivariate statistical quality control (MSQC) has the potential to evaluate current conditions with respect to normal conditions based on historical data. The state of the highway system can be assessed using Hotellings T/sup 2/ to measure the distance of current data from the center of the region defined by historical data. This measurement and additional calculations can provide an assessment of normality across a range of conditions, rather than the binary output typical of incident detection algorithms. Extensions on basic MSQC can provide additional insight into causes of abnormal traffic conditions.

PRIORITIZING PROPOSED TRANSPORTATION IMPROVEMENTS: METHODS, EVALUATION, AND RESEARCH NEEDS

Since the costs of proposed transportation projects typically exceed available funding, the process by which projects are selected and prioritized for development is scrutinized by a wide variety of stakeholders. The U.S. Congress and state legislatures are increasingly involved in selecting projects, and thus some projects do not evolve from the transportation planning process. Oversight panels, special interest groups, and the public are applying pressure to clarify the process by which proposed projects are prioritized and to minimize the impact of politics. A rational tool that lays out a clear procedure can assist decision makers in selecting and prioritizing potential improvements, make the process more transparent, and strengthen the link between planning and programming. Several such tools have been proposed, and many state and local transportation agencies are using some of them. These instruments vary widely in objectivity, the factors considered in evaluating potential transportation improvements, and output formats. The role they play in programming decisions also varies, but their use makes available more information to interested parties, including those vested with the authority to make decisions about transportation programming.

Quality Assurance of Hot-Mix Asphalt: Comparison of Contractor Quality Control and Georgia Department of Transportation Data

Quality assurance is the process by which highway construction elements are sampled and tested to ensure compliance with specifications and other project requirements. The results of contractor-performed tests, originally performed for quality control purposes, are increasingly used in the acceptance decision in many states. The Georgia Department of Transportation (GDOT) uses contractor-performed tests in the acceptance decision on acceptable corroboration of GDOT-performed tests. Statistical analyses were performed to assess differences between tests conducted on hot-mix asphalt concrete by GDOT and its contractors during the 2003 construction season. Measurements of gradation and asphalt content taken by both parties were compared both across all projects and on a project-by-project basis for projects large enough to meet sample size requirements for this type of analysis. Both tabular and graphic representations of data are used to interpret the results. Statistically significant differences occur in some cases; these differences are much more common when comparing variability of these measurements than with the means. At the project level, on most projects in which statistically significant differences occur, the GDOT value typically is larger.

Comprehensive Analysis of Wrong-Way Driving Crashes on Alabama Interstates

Crash data on Alabama Interstates were collected for a 5-year period from 2009 to 2013. True wrong-way driving (WWD) crashes were identified from the hard copy of crash reports and existing maps. The crash data contained 18 explanatory variables representing the driver, the temporal, vehicle, and environmental information. A Firth’s penalized likelihood logistic regression model was developed to examine the influence of the explanatory variable on the dichotomous dependent variable (type of crash, i.e., WWD versus non-WWD). This model was an appropriate tool for controlling the influence of all confounding variables on the probability of WWD crashes while considering the rareness of the event (i.e., WWD). A separate model that used the standard binary logistic regression was also developed. Two information criteria (the Akaike information criterion and the Bayesian information criterion) obtained from both developed models indicated that for this database, Firth’s model outperformed the standard binary logistic model and provided more reliable results. With Firth’s model, explanatory variables including month of the year, time of day, driver’s age, driver’s mental and physical condition, driver’s residency distance, vehicle age, vehicle damage, towing condition, airbag deployment status, and roadway condition were found to characterize WWD crashes. Using the obtained odds ratio, this paper discusses the various effects of the identified variables and recommends several countermeasures policy makers can use to address the WWD issue on Alabama Interstates.

Modeling Retroreflectivity Performance of Thermoplastic Pavement Markings in Alabama

AbstractPavement markings play an important role in the roadway system because they provide information for drivers to follow the road. Markings must be replaced when their retroreflectivity falls below an acceptable level. The rate of degradation of marking retroreflectivity can be influenced by many factors, such as type of material and traffic volume. The efficiency of pavement marking maintenance programs could be improved with statistical models that estimate the degradation of retroreflectivity over time. A case study based on data collected by the Alabama DOT was undertaken to address these issues. The purpose of the research reported in this paper was to observe how existing models fit the data and develop new statistical models that could predict retroreflectivity over time for Alabama conditions. Modeling of retroreflectivity performance over time for thermoplastic markings was executed for 15 projects that had measurements of retroreflectivity for the same locations in years 2007–2010. The mode...

Relating short-term traffic forecasting to current system state using nonparametric regression

Traffic management systems have been developed as an intelligent transportation systems application in major metropolitan areas to improve the safety and efficiency of the highway network. Monitoring of traffic conditions is a key function of these systems; short-term forecasting of traffic conditions is a potentially valuable function still under development. These systems typically archive extensive amounts of traffic data that can be mined to provide bases for monitoring and forecasting applications. Multivariate statistical quality control provides a means to describe the extent to which an observation deviates from a definition of normal, with respect to historical means and variances. A promising set of non-parametric regression techniques that use the nearest neighbor concept to locate past observations that are similar to current conditions for use in short-term forecasting has been developed. Information on deviation from normal of current and historical observations can be used to enhance these forecasting procedures. This paper documents several approaches to linking condition monitoring with forecasting, such as in the nearest neighbor selection process, in the determination of proximity of neighbors to the current observation, and in their relative contributions to the forecasts.