Matthew G. Karlaftis

A MULTIVARIATE STATE SPACE APPROACH FOR URBAN TRAFFIC FLOW MODELING AND PREDICTION

Urban traffic congestion is one of the most severe problems of everyday life in Metropolitan areas. In an effort to deal with this problem, intelligent transportation systems (ITS) technologies have concentrated in recent years on dealing with urban congestion. One of the most critical aspects of ITS success is the provision of accurate real-time information and short-term predictions of traffic parameters such as traffic volumes, travel speeds and occupancies. The present paper concentrates on developing flexible and explicitly multivariate time-series state space models using core urban area loop detector data. Using 3-min volume measurements from urban arterial streets near downtown Athens, models were developed that feed on data from upstream detectors to improve on the predictions of downstream locations. The results clearly suggest that different model specifications are appropriate for different time periods of the day. Further, it also appears that the use of multivariate state space models improves on the prediction accuracy over univariate time series ones.

A DEA approach for evaluating the efficiency and effectiveness of urban transit systems

Abstract The need to measure transit system performance along with its various dimensions has led to the development of a large number of quantitative performance indicators. However, depending upon the specific indicator examined, different conclusions can oftentimes be reached regarding performance. Further, although performance and scale economies are closely related issues, they have been generally examined separately in the transit literature. The research reported in this paper uses data envelopment analysis and globally efficient frontier production functions to investigate two important issues in transit operations: first, the relationship between the two basic dimensions of performance, namely efficiency and effectiveness; second, the relationship between performance and scale economies. Using data from 256 US transit systems over a five-year period the results indicate that efficiency and effectiveness are positively related. Further, they imply that the magnitude of scale economies depends on the output specification.

The Gravity Model Specification for Modeling International Trade Flows and Free Trade Agreement Effects: A 10-Year Review of Empirical Studies

The gravity model has been extensively used in international trade research for the last 40 years because of its considerable empirical robustness and explanatory power. Since their introduction in the 1960s, gravity models have been used for assessing trade policy implications and, particularly recently, for analyzing the effects of Free Trade Agreements on international trade. The objective of this paper is to review the recent empirical literature on gravity models, highlight best practices and provide an overview of Free Trade Agreement effects on international trade as reported by relevant gravity model-based studies over the past decade.

Vehicle occupant injury severity on highways: An empirical investigation

Accident severity analysis is important to both researchers and practitioners because of its implications in accident cost estimation, external cost estimation and road safety. Although much research has been done to explore the factors influencing crash-injury severity, few studies have investigated the association between severity and traffic characteristics collected real-time during the time the accident occurred. We apply a random parameters ordered probit model to explore the influence of speed and traffic volume on the injury level sustained by vehicle occupants involved in accidents on the A4-A86 junction in the Paris region. Results indicate that increased traffic volume has a consistently positive effect on severity, while speed has a differential effect on severity depending on flow conditions.

A hybrid hypercube - Genetic algorithm approach for deploying many emergency response mobile units in an urban network

Emergency response services are critical for modern societies. This paper presents a model and a heuristic solution for the optimal deployment of many emergency response units in an urban transportation network and an application for transit mobile repair units (TMRU) in the city of Athens, Greece. The model considers the stochastic nature of such services, suggesting that a unit may be already engaged, when an incident occurs. The proposed model integrates a queuing model (the hypercube model), a location model and a metaheuristic optimization algorithm (genetic algorithm) for obtaining appropriate unit locations in a two-step approach. In the first step, the service area is partitioned into sub-areas (called superdistricts) while, in parallel, necessary number of units is determined for each superdistrict. An approximate solution to the symmetric hypercube model with spatially homogeneous demand is developed. A Genetic Algorithm is combined with the approximate hypercube model for obtaining best superdistricts and associated unit numbers. With both of the above requirements defined in step one, the second step proceeds in the optimal deployment of units within each superdistrict.

Subsidy and public transit performance: A factor analytic approach

The need to measure and evaluate transit system performance has led to the development of numerous performance indicators. However, depending upon the indicator, we oftentimes reach different conclusions regarding transit system performance. The research reported in this paper uses factor analytic methods to generate a set of underlying attributes (factors) that capture the performance of public transit systems in Indiana. Similar to what is reported in the literature, this study finds three attributes that best describe transit system performance: efficiency, effectiveness, and overall performance. Based upon systemsÕ factor scores, the study finds that systems scoring highly on one attribute generally perform well on the remaining attributes. Further, there is an inverse relationship between system performance and subsidies, a finding that supports performance based subsidy allocations.

AN INTERNATIONAL COMPARATIVE STUDY OF SELF-REPORTED DRIVER BEHAVIOR

Abstract Using a large data base of 20,725 questionnaires from 19 European countries, this article uses a combination of factor analysis and tree based regression to determine driver groups with homogeneous self-reported behavior and determine whether regional differences in driving behaviors exist. Self-reported behavior, including speeding, reckless driving, seat belt use, and drinking and driving are examined. The results suggest that speeding and general reckless (dangerous) behavior are related, perhaps capturing a drivers “risk taking” or “pre-trip violations” behavior. Similarly, seat belt use and driving under the influence of alcohol are also related and may represent a drivers “law abiding” tendency or “during-trip violations” behavior. Further, important regional differences and similarities between European drivers are uncovered. Northern European drivers report a significantly higher compliance with drinking and driving laws and seat belt use regulations than do Southern and Eastern European drivers.

Investigating transit production and performance: a programming approach

Although efficiency and productivity are closely related issues, they have been generally examined separately in the transit literature. Using an extensive panel data set, this analysis extends prior research in two directions. First, efficiency rankings and efficient subsets of transit systems are obtained through data envelopment analysis (DEA), a non-parametric linear programming based methodology. Second, based on the results of the DEA analysis, globally efficient frontier production functions, in the context of transit operations in the United States, are built. The results indicate that when jointly considered, there is an improvement on both the theoretical and empirical aspects of examining efficiency and production in transit systems. Further, the results indicate that efficiency and returns to scale findings differ substantially depending on the evaluation methodology used.

Modeling the Effects of Weather and Traffic on the Risk of Secondary Incidents

To extract useful information on variables that are associated with secondary accident likelihood, this article develops neural network models with enhanced explanatory power. Traffic and weather conditions at the occurrence of a primary incident are explicitly considered. Two measures to extract variable significance are introduced: mutual information and partial derivatives. The proposed approach is also compared to other classical statistical approaches of the Logit family. Results suggest that traffic speed, duration of the primary accident, hourly volume, rainfall intensity, and number of vehicles involved in the primary accident are the top five factors associated with secondary accident likelihood. However, changes in traffic speed and volume, number of vehicles involved, blocked lanes, and percentage of trucks and upstream geometry also significantly influence the probability of having a secondary incident. Finally, the incident management implications of the proposed modeling approach are discussed.

Identifying crash type propensity using real-time traffic data on freeways

INTRODUCTION We examine the effects of various traffic parameters on type of road crash. METHOD Multivariate probit models are specified on 4-years of data from the A4-A86 highway section in the Ile-de-France region, France. RESULTS Empirical findings indicate that crash type can almost exclusively be defined by the prevailing traffic conditions shortly before its occurrence. Rear-end crashes involving two vehicles were found to be more probable for relatively low values of both speed and density, rear-end crashes involving more than two vehicles appear to be more probable under congested conditions, while single-vehicle crashes appear to be largely geometry-dependent. IMPACT ON INDUSTRY Results could be integrated in a real-time traffic management application.