Kyriacos Mouskos

A Linear Model for the Continuous Network Design Problem

This paper introduces a dynamic network design problem model that can be used to compute continuous network improvements. The model assumes system optimum traffic flow conditions and time-dependent demand. A linear programming formulation is introduced based on a dynamic traffic assignment (DTA) model that propagates traffic according to the cell transmission model. The introduced approach is limited to continuous link improvements and does not provide for new link additions. This paper is one of the first attempts to provide an analytical formulation for network design that accounts for DTA conditions. A single destination example network, resembling a freeway corridor, is used to test the model under various congestion levels, loading patterns and budget sizes. Findings from the example demonstrate that the model is simple and effective at capturing the dynamic behavior of traffic and the dependencies among adjacent links.

Simplified formulation of the toll design problem

Much of the existing literature on congestion pricing has assumed that it is possible to toll every link in the network. Whereas this is very interesting theoretically, it is not of much use in practice where only a limited number of links can be tolled (e.g., certain bridges, tunnels, and freeways). To accommodate this situation, a bilevel formulation of the toll design problem that is similar in spirit to bilevel formulations of the continuous network design problem can be developed. Unfortunately, such formulations can be difficult to solve. A simplified, single-level formulation of the toll design problem that may prove to be more applicable in practice is presented.

Evaluation of the TRANSCOM's system for managing incidents and traffic (TRANSMIT)

TRANSCOMs system for managing incidents and traffic (TRANSMIT) is a traffic surveillance and incident detection system that is based on traffic probes equipped with the E-ZPass electronic toll collection tags. The evaluation has been carried out to assess its communication system performance, its incident detection capability, and the traffic-flow parameters estimation, and to identify its current and potential benefits. The TRANSMIT communication system exhibited excellent performance in terms of the transmission rates system wide. The TRANSMIT incident detection algorithm performed very favorably in comparison to the results of the best incident detection algorithms reported in the literature. The link travel time estimates were found to be within the 95% confidence interval. The system offers a unique opportunity for researchers to collect travel-time data in real time, and direct measurements of the space mean speed, as well as incident detection related data.

Transportation Operations Coordinating committee system for managing incidents and traffic : Evaluation of the incident detection System

The Transportation Operations Coordinating Committee’s System for Managing Incidents and Traffic (TRANSMIT) is an operational test that uses vehicles equipped with tags of the E-ZPass electronic toll collection system as traffic probes for traffic surveillance and incident detection. The TRANSMIT incident detection algorithm is based on statistical comparison of real-time estimates of travel times with continuously updated historical travel times for the same time period of the day and type of day (weekday, Saturday, Sunday, or holiday). The probability of detecting an incident and the false-alarm rates produced by TRANSMIT during a 4-month evaluation period (January to April 1996) may be considered excellent for the data collected on the New York State Thruway and satisfactory for the data collected on the Garden State Parkway. The mean time to detection of an incident was not estimated at this stage of the evaluation. The TRANSMIT communication system exhibited excellent performance in terms of the transmission rates systemwide, which were found to be near 100 percent. Only the radio link at Tappan Zee Bridge exhibited a lower transmission rate. A limited probe vehicle test was conducted to determine the detection rate at individual roadside terminals. The lower detection rates observed were site specific rather than systemwide. The performance of the TRANSMIT incident detection algorithm performed very favorably compared with the performances of the best incident detection algorithms reported in the literature.

EFFECT OF MIDBLOCK ACCESS POINTS ON TRAFFIC ACCIDENTS ON STATE HIGHWAYS IN NEW JERSEY

The focus of this research was to conduct a statistical analysis of the effect of various traffic, geometric, and environmental factors on accident rates on New Jersey State highways. The main concern was the effect of midblock access points on accident rates. To identify the characteristics of access-related accidents (section accidents), a comparison study was conducted to investigate the effect of various factors on both section accidents and signalized intersection accidents. Access density was found to be a contributing factor to the occurrence of accidents, but it was not the only factor. About 30 percent of accidents on the study routes in New Jersey are expected to occur between signalized intersections, whereas 7 percent of the accidents are due to maneuvering to and from access points. Accident rates for sections between signalized intersections were better represented by a log-normal distribution than by a normal distribution. The effect of single factors such as access density, median, shoulder, number of lanes, and speed limits on accident rates was also investigated with the Kruskal—Wallis test. Of the above factors, only the number of lanes was found to be nonsignificant to accident rates. Two regression models for four lanes with shoulder, two lanes without shoulder, and four lanes with median were found to have sufficiently good R2 values and their use is recommended with caution. Hourly and subhourly traffic flow data, average operating speed and variance data, and law enforcement data will further improve the analysis of accident data.

Artificial Neural Network Model for Estimating Temporal and Spatial Freeway Work Zone Delay Using Probe-Vehicle Data

Highway lane closures due to road reconstruction and the resulting work zones have been a major source of nonrecurring congestion on freeways. It is extremely important to calculate the safety and cost impacts of work zones: the use of new technologies that track drivers and vehicles make that possible. A multilayer feed-forward artificial neural network (ANN) model is developed in this paper to estimate work zone delay by using the probe-vehicle data. The probe data include the travel speeds under normal and work zone conditions. Unlike previous models, the proposed model estimates temporal and spatial delays, which are applied to a real world case study in New Jersey. The work zone data (i.e., starting time, duration, length, and number of closed lanes) were collected on New Jersey freeways in 2014 together with actual probe-vehicle speeds. A comparative analysis was conducted; the results indicate that the ANN model outperforms the traditional deterministic queuing model in terms of the accuracy in estimating travel delays. The ANN model can be used to calculate contractor penalty in terms of cost overruns as well as incentivize a reward schedule in case of early work competition. The model can assist work zone planners in designing optimal start and end time of work zone as function of time of day. In assessing the performance of work zones, the model can assist transportation engineers to better develop and evaluate traffic mitigation and management plans.

GENERATING DRIVER POPULATION FOR THE MICROSCOPIC SIMULATION MODEL (CORSIM)

CORSIM (CORridor SIMulator), an interval- based simulation model, was developed by the Federal Highway Administration (FHWA) to analyze traffic operations. In CORSIM, each ve hicle is a distinct object that moves every second, whose kinematic properties (speed, acceleration, and jerk) as well as its status (queued, moving, lane changing) are updated every second. Thus, driving behavior on roadways reflecting real- world traffic operations can be simulated. This paper suggests that the driving behavior repre sented by speed distributions over different driver types should be built in CORSIM to reflect more realistic patterns of driving behavior. The impact of the driver distribution to freeway capacity is demonstrated through simulation analysis.

Evaluation of Transmit's communication system

The System for Managing Incidents and Traffic (Transmit) is a traffic surveillance and incident detection system utilizing existing electronic toll traffic management technology of the E-ZPass system. Its communication system consists of the hardware installed at the roadside terminal (RST) readers to detect E-ZPass tag equipped vehicles (probes), telephone and radio links to transmit corresponding probe data and the computers and related peripherals at the operations information center (OIC) for real time processing of this data. The evaluation of the Transmits communication system was carried out to assess its transmission capability and reliability in terms of data transmission and detection rates. The transmission rates were determined based on the data collected at the RSTs and the corresponding data received at the OIC. The Transmits communication system exhibited excellent performance in terms of the transmission rates system wide, which was found to be near 100%. Only the radio link at Tappan Zee Bridge exhibited a lower transmission rate. A limited probe vehicle test was conducted to determine the detection rate at individual RSTs. The lower detection rates observed were site specific, rather than system wide. However, due to the high number of tag equipped vehicles that are expected in the future, the lower detection rates would not have any significant impacts on the determination of the traffic flow parameters (link travel time and space mean speed) as well as the incident detection capability.

A real-time integrated transport/communication/database architecture to support traffic simulation

This research aims for an establishment of a global center that will advance the knowledge in real-time human and goods transport. The focus of this paper is an approach for addressing the performance and feasibility issues with respect to a reliable and distributed traffic data communication network for optimal and secure routing of traffic. We present communications and database modeling for data acquisition via a high performance system. The proposed architecture is distributed, survivable, and reliable, thus capable of handling stationary and mobile sources of communication systems and integrating databases and dynamic traffic assignment into one comprehensive model. A case study on a visual interactive system for transport algorithms is discussed.