Ardeshir Faghri

DEVELOPMENT OF A COMPUTER SIMULATION MODEL OF MIXED MOTOR VEHICLE AND BICYCLE TRAFFIC ON AN URBAN ROAD NETWORK

The first step has been taken in the effort to develop a microscopic simulation model of mixed motor vehicle and bicycle traffic over an entire network. The primary objective was to develop the main algorithm of BICycle SIMulator and translate it into a working computer program. To achieve the goal, the gaps in the existing literature were filled by data collected at the University of Delaware campus in Newark. An equation of mixed bicycle and motor vehicle following was developed and used for the representation of longitudinal interaction of vehicles. The program was implemented in C+plus; programming language and tested on real-world data. The resulting software is the first in the world to allow the inclusion of bicyclists in the analysis of urban transportation by the means of microscopic simulation.

Near-Term Travel Speed Prediction Utilizing Hilbert-Huang Transform

: Accurate short-term prediction of travel speed as a proxy for time is central to many Intelligent Transportation Systems, especially for Advanced Traveler Information Systems and Advanced Traffic Management Systems. In this study, we propose an innovative methodology for such prediction. Because of the inherently direct derivation of travel time from speed data, the study was limited to the use of speed only as a single predictor. The proposed method is a hybrid one that combines the use of the empirical mode decomposition (EMD) and a multilayer feedforward neural network with backpropagation. The EMD is the key part of the Hilbert–Huang transform, which is a newly developed method at NASA for the analysis of nonstationary, nonlinear time series. The rationale for using the EMD is that because of the highly nonlinear and nonstationary nature of link speed series, by decomposing the time series into its basic components, more accurate forecasts would be obtained. We demonstrated the effectiveness of the proposed method by applying it to real-life loop detector data obtained from I-66 in Fairfax, Virginia. The prediction performance of the proposed method was found to be superior to previous forecasting techniques. Rigorous testing of the distribution of prediction errors revealed that the model produced unbiased predictions of speeds. The superiority of the proposed model was also verified during peak periods, midday, and night. In general, the method was accurate, computationally efficient, easy to implement in a field environment, and applicable to forecasting other traffic parameters.

Application of GPS in Traffic Management Systems

Integrated Traffic Management Systems (ITMS) need reliable, accurate, and real-time data. Travel time, speed, and delay are three of the most important factors used in ITMS for monitoring, quantifying, and controlling congestion. GPS has recently become available for civil applications. Because it provides real-time spatial and time measurements, it has an increasing use in conducting different transportation studies. This article presents the application of GPS in collecting travel time, speed, and delay information of 64 major roads in the state of Delaware. A comparative statistical analysis was performed on data collected by GPS, with data collected simultaneously by the conventional method. The GPS data proved to be at least as accurate as the data collected by the conventional method, and it was 50% more efficient in terms of manpower. Moreover, the sample-size requirement was determined to maintain 95% confidence level throughout the controlled test. Benefiting from the Geographic Information Systems dynamic segmentation tool, our travel time, delay, and speed information were integrated with other relevant traffic data. This was presented graphically on the Internet for public use. Statistical trend analysis for the data collected in 1997, 1998, 1999, and 2000 are also presented and applications on the overall ITMS are discussed.

Evaluating the Conversion of All-Way Stop-Controlled Intersections into Roundabouts

Roundabouts are becoming increasingly popular in the United States and have been considered for use at intersections facing operational or safety problems. A comparative analysis was performed on all-way stop-controlled (AWSC) intersections and roundabouts on the basis of aaSIDRA software after calibration against local conditions in Delaware and Maryland. A decision support system was developed in the form of a knowledge-based expert system on the basis of the results of the analysis to evaluate the performance of AWSC intersections and their potential success if converted into roundabouts. The results of the comparative analysis are in line with some previous researchers’ findings that AWSC intersections always perform worse than signal-controlled or roundabout type intersections under the same set of traffic conditions and geometric limitations.

Estimation of percentage of pass‐by trips generated by a shopping center using artificial neural networks

Pass‐by trips are trips made as intermediate stops on the way from an origin to a primary trip destination. Accurate estimates of the percentage of pass‐by trips generated by a land use are extremely important for both planners and developers. The traditional method of pass‐by trip estimation is regression modeling with the help of the U.S. Institute of Transportation Engineers (ITE) Trip Generation manual. This paper also uses data from the Trip Generation manual, and focuses on an alternative methodology based on Artificial Neural Networks (ANNs). Use is made of backpropogation, a popular ANN paradigm, and five different architectures of backpropogations are developed, tested and compared against three different regression models — linear, log‐log and log‐linear forms, respectively. The results from the regression and ANN‐based models are compared in terms of the Root Mean Square of Errors (RMSE) of predicted values. It is found that the worst ANN prediction RMSE is lower than the best regression predic...

Development of Integrated Traffic Monitoring System for Delaware

The establishment of a comprehensive statewide traffic counting program is discussed. The program comprises automatic traffic recorder (ATR), automatic vehicle classification (AVC), and weigh-in-motion (WIM) sites for the state of Delaware. The program was undertaken to review, establish, and implement effective statistical and procedural methods. The second phase of a two phase project, which implements the methodologies that were derived in the first phase, is presented. Using descriptive analysis and seasonal grouping, the number and location of sites needed for each of the three types of traffic monitoring devices were determined. Existing field data from Delawares current ATR locations allowed for a statistical determination of the necessary number and road-type group distribution for the ATR sites. The absence of field data for AVC and WIM sites, however, necessitated alternative methods for determining the number and location of the traffic monitoring devices. As a result, a combination of statistical analysis and engineering judgment must be used for the establishment of any statewide traffic monitoring system.

Experimental investigation of spatial breakdown evolution on congested freeways

Although urban and suburban freeways are designed to enable smooth flow with high speed, they experience traffic congestion during peak day periods especially when traffic demand is high but also during unexpected incidents. Traffic breakdown is identified when there is a rapid drop in speed level and a simultaneous steep rise in density level during the phase transition from free-flow to breakdown flow. The recovery from traffic breakdown is achieved after speed and density return to their pre-breakdown level in the uncongested conditions. This paper evaluates the spatial dependency of breakdown and investigates the consistency of time of breakdown between weekdays, based on data collected from a suburban section of Interstate 66 in Virginia. The identification of location and weekday effects is conducted by experimental design techniques. The spatial dependency of the transition times is evaluated through raw data in the speed–time plane and further examined by the shockwave linear propagation principle. Both approaches are generally consistent and produce similar results.

Artificial neural network-based approach to modeling trip production

Accurate and reliable estimates of trip production of a study area are important for an accurate forecast from the four-step travel demand forecasting procedure. In the trip generation step, trip production estimates are considered more accurate, and trip attractions are adjusted while keeping the productions constant. This means that more accurate trip production rates will result in more reliable forecasts. Improving the accuracy of forecasts requires an extensive and reliable data base or improvement in the modeling techniques. Since data base enhancement is costly and time-consuming, an alternative methodology is proposed and examined for trip production prediction using artificial neural network (ANN) concepts and techniques. The data base used was made available by the Delaware Department of Transportation. The data were collected for 60 sites throughout Delaware between 1970 and 1974 and are based on field counts and home interviews. Twenty-six regression models were calibrated on these data. In addition, 18 ANN architectures were developed, and their predictions were compared with those from regression models. Comparisons indicate that the ANNs have the capability to represent the relationship between the trip production rate and the independent variables more accurately than regression analysis at no additional cost of increasing the data base.

DEVELOPMENT AND EVALUATION OF A STATISTICALLY RELIABLE TRAFFIC COUNTING PROGRAM.

This paper reviews the existing procedures to compute the seasonal factors, growth factors, and the number of automatic traffic recorder stations (ATRs) for developing a statistically reliable traffic counting program. The review is based on, (i) theoretical evaluation of the procedures, (ii) analysis of the results obtained from applying these procedures to extensive real life data, and (iii) experience of past researchers with these methods. The paper primarily deals with existing statistical procedures for determining seasonal factors, since it is believed that these factors play the most important role in estimating Annual Average Daily Traffic (AADTs) from short counts. The procedure of cluster analysis (which is most commonly used to determine seasonal factors) is critically reviewed. The shortcomings of this method motivated the development of a different statistical procedure, based on regression analysis, to determine seasonal factors. It is shown that this newly developed method gives better res...

Applying Problem-Oriented and Project-Based Learning in a Transportation Engineering Course

AbstractA growing consensus in the literature on how to improve the quality of learning suggests that adaptions to problem-based learning (PBL) methods could firstly improve learning experiences, problem solving, skill acquisition and a reasonable level of skill transfer from a student perspective, and secondly develop better professional competence and preparation in these needed highway capacity analysis skills. This paper describes the educational basis of problem-oriented and project-based learning (POPBL) approaches that have been developed as an integral part of a four-year undergraduate engineering degree program at the University of Delaware and incorporated into the transportation curriculum, using real world cases to teach students how to think like expert practitioners. A seven-phase POPBL conceptual framework and the experiences of employing POPBL are presented to demonstrate how it operates in transportation engineering education.