Brian Lee Smith

Comparison of parametric and nonparametric models for traffic flow forecasting

Single point short-term traffic flow forecasting will play a key role in supporting demand forecasts needed by operational network models. Seasonal autoregressive integrated moving average (ARIMA), a classic parametric modeling approach to time series, and nonparametric regression models have been proposed as well suited for application to single point short-term traffic flow forecasting. Past research has shown seasonal ARIMA models to deliver results that are statistically superior to basic implementations of nonparametric regression. However, the advantages associated with a data-driven nonparametric forecasting approach motivate further investigation of refined nonparametric forecasting methods. Following this motivation, this research effort seeks to examine the theoretical foundation of nonparametric regression and to answer the question of whether nonparametric regression based on heuristically improved forecast generation methods approach the single interval traffic flow prediction performance of seasonal ARIMA models.

Exploring Imputation Techniques for Missing Data in Transportation Management Systems

Many states have implemented large-scale transportation management systems to improve mobility in urban areas. These systems are highly prone to missing and erroneous data, which results in drastically reduced data sets for analysis and real-time operations. Imputation is the practice of filling in missing data with estimated values. Currently, the transportation industry generally does not use imputation as a means for handling missing data. Other disciplines have recognized the importance of addressing missing data and, as a result, methods and software for imputing missing data are becoming widely available. The feasibility and applicability of imputing missing traffic data are addressed, and a preliminary analysis of several heuristic and statistical imputation techniques is performed. Preliminary results produced excellent performance in the case study and indicate that the statistical techniques are more accurate while maintaining the natural characteristics of the data.

Multiple-Interval Freeway Traffic Flow Forecasting

Freeway traffic flow forecasting will play an important role in intelligent transportation systems. The TRB Committee on Freeway Operations has included freeway flow forecasting in its 1995 research program. Much of the past research in traffic flow forecasting has addressed short-term, single-interval predictions. Such limited forecasting models will not support the development of the longer-term operational strategies needed for such events as hazardous material incidents. A multipleinterval freeway traffic flow forecasting model has been developed that predicts traffic volumes in 15-min intervals for several hours into the future. The nonparametric regression modeling technique was chosen for the multiple-interval freeway traffic flow forecasting problem. The technique possesses a number of attractive qualities for traffic forecasting. It is intuitive and uses a data base of past conditions to generate forecasts. It can also be implemented as a generic algorithm and is easily calibrated at field locati...

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.

Data-driven methodology for signal timing plan development: A computational approach

Traffic signal systems serve as one of the most powerful control tools available to improve the efficiency of surface transportation travel. A large number of signal systems currently operate using the time-of-day (TOD) approach. In TOD systems, a day is segmented into a number of intervals in which a different timing plan is used. Thus, the challenge in operating a TOD system effectively is to (1) identify appropriate TOD intervals, and (2) develop optimal timing plans for each interval. The existing procedures used by traffic engineers to address these challenges are time consuming and use relatively small sets of data. This research effort developed a new timing plan development methodology that takes advantage of the large sets of archived traffic data (volume and occupancy) that modern systems are equipped to compile. Based upon statistical cluster analysis, this methodology (1) automates the identification of TOD intervals using a high-resolution definition of system state, and (2) provides representative volumes for plan optimization based on the set of archived data. The results of a case study reported in this paper demonstrate that the methodology supports the development of a TOD system that provides benefits when considering performance measures such as delay, when compared to currently used techniques.

A PROTOTYPE CASE-BASED REASONING SYSTEM FOR REAL-TIME FREEWAY TRAFFIC ROUTING

Abstract With the recent advances in communications and information technology, real-time traffic routing has emerged as a promising approach to alleviating congestion. Existing approaches to developing real-time routing strategies, however, have limitations. This study examines the potential for using case-based reasoning (CBR), an emerging artificial intelligence paradigm, to overcome such limitations. CBR solves new problems by reusing solutions of similar past problems. To illustrate the feasibility of the approach, the study develops and evaluates a prototype CBR routing system for the interstate network in Hampton Roads, Virginia. Cases for building the system’s case-base are generated using a heuristic dynamic traffic assignment (DTA) model designed for the region. Using a second set of cases, the study evaluates the performance of the prototype system by comparing its solutions to those of the DTA model. The evaluation results demonstrate that the prototype system is capable of running in real-time , and of producing high quality solutions using case-bases of reasonable size.

Kalman Filter Approach to Speed Estimation Using Single Loop Detector Measurements under Congested Conditions

The ability to measure or estimate accurate speed data are of great importance to a large number of transportation system operations applications. Estimating speeds from the widely used single inductive loop sensor has been a difficult, yet important challenge for transportation engineers. Based on empirical evidence observed from sensor data collected in two metropolitan regions in Virginia and California, this research developed a Kalman filter model to perform speed estimation for congested traffic. Taking advantage of the coexistence of dual loop and single loop stations in many freeway management systems, a calibration procedure was developed to seed and initiate the algorithm. Finally, the paper presents an evaluation that illustrates that the proposed algorithm can produce acceptable speed estimates under congested traffic conditions, consistently outperforming the conventional g-factor approach.

Meeting Real–Time Traffic Flow Forecasting Requirements with Imprecise Computations

This paper explores the ability of imprecise computations to address real-time computational requirements in infrastructure control and management systems. The research in this area focuses on development of nonparametric regression as a means to forecast traffic flow rates for transportation management systems. Nonparametric regression is a forecasting technique based on nearest neighbor searching, in which forecasts are derived from past observations that are similar to current conditions. A key concern regarding nonparametric regression is the significant time required to search for nearest neighbors in large databases. Results presented herein indicate that approximate nearest neighbors, which are imprecise computations as applied to nonparametric regression, may be used to adequately speed the execution time of nonparametric regression, with acceptable degradations in forecast accuracy.

Investigation of Dynamic Probe Sample Requirements for Traffic Condition Monitoring

Many agencies are exploring the use of probe-based traffic monitoring systems to collect information on system performance. Although these systems offer the potential of directly measuring travel times and of significantly reducing the cost per mile of traffic monitoring systems, the issue of identifying the number of samples required to produce reliable and accurate condition information has not been adequately examined. Most existing research into this issue addresses the problem from a very broad perspective, offering a single percentage of vehicles in an area that must be equipped as probes to provide good coverage of the area. This approach ignores variations in flow conditions through time and across different links of the network. The described research addresses the issue of sample sizes for different links and measurement intervals from a rigorous statistical perspective. First, the issue of producing sample sizes on the basis of the central limit theorem (CLT) is empirically examined by using Virginia freeway speed data to assess this methods validity even when the underlying speed population is nonnormal. Applying the CLT to the speed data shows that fluctuations in flow conditions over time and across different locations require varying sample sizes, which the static sample sizes proposed in previous research do not take into account. Finally, the CLT-based sample sizes are compared with a representative static sample size proposed in the literature to illustrate how the CLT method can reduce sample sizes while maintaining the desired confidence and accuracy levels.

CASE-BASED REASONING FOR REAL-TIME TRAFFIC FLOW MANAGEMENT

Real-time traffic management is a promising approach for alleviating congestion. This approach uses real-time and predicted traffic information to develop routing strategies that optimize the performance of highway networks. This article explores the potential for using case-based reasoning (CBR), an emerging artificial intelligence (AI) paradigm, to overcome the limitations of existing traffic-management decision support systems. To illustrate the feasibility of the approach, the article develops and evaluates a prototype CBR routing system for a real-world network in Hampton Roads, Virginia. Cases for building the systems case base are generated using a heuristic dynamic traffic assignment (DTA) model specifically designed for the region. Using a set of 25 new independent cases, the performance of the prototype system is evaluated by comparing its solutions with those of the DTA model. The evaluation results demonstrate the feasibility of the CBR approach. The prototype system was capable of running in real time and produced high-quality solutions using case bases of reasonable size.