Project Abstract

Traffic congestion interferes with the ability of a transit agency to provide good bus service. By reducing the speed that buses can travel, congestion requires a greater expenditure of resources (particularly the number of buses, drivers, and vehicle hours) to provide a given level of service to the passengers.

The object of this proposed project will be to estimate the actual cost of congestion to New Jersey Transit from the requisite additional resources due to congestion. This will be done through the construction of a model of bus travel time as a function of congestion and other factors and the use of the model to calculate additional bus travel time. From these numbers, the additional vehicle hours, buses, and drivers will be calculated. Using a cost allocation model, the increase in cost will be calculated.

The proposed work will be done by a team of researchers from the Region 2 University Transportation Research Center (UTRC), combining the strengths of researchers at City College of New York with consultants.

Task Descriptions

The basic approach will be to develop one or a series of regression models that will predict bus travel time as a function of traffic volume, V/C ratio, traffic delay, or other measure of congestion; the models will include other predictive variables as needed to increase their accuracy or stability. These models will then be used to estimate round trip travel times for different bus routes under congested and uncongested conditions. The additional buses needed under congested conditions will be calculated. The added costs due to the increase in hours of operations and vehicles will be calculated using costs appropriate to New Jersey Transit.

The project will be organized into the following five tasks:

Task 1: Literature Review A summary of recent literature (since 1996) concerning the impact of highway congestion on bus operations and costs will be produced, including estimation and modeling of bus speeds; causes of bus delay; and other related topics.

Task 2: Data Collection Data on the bus system and routes and data on traffic and the highway system will be collected. In the initial step in this task, members of the study team will meet with New Jersey Transit personnel to go over the type of data that is available and determine how other crucial data will be determined or estimated.

Task 3: Model Development Models that estimate bus travel time as a function of traffic travel time, traffic delay, or other measures of congestion will be developed. Other predictive variables will be included in order to control for the effect of other factors, particularly the impacts of the number of passengers on dwell time, the frequency of bus stops, frequency and timing of traffic signals. Other variables that will be considered include (but are not limited to): service type, street and traffic lane configuration, bus stop locations, number of right and left turns in the bus route, pedestrian volumes, and bus stop location and geometry. Models incorporating lags between increases in congestion and schedule changes will be evaluated. It is anticipated that individual models may be needed for different route types and or different operating environments. Criteria to evaluate the models will include accuracy in estimating current bus travel times, usefulness for future predictions, and the usual statistical measures.

Task 4: Estimation of Impacts The first use of the model will be to estimate the impacts of congestion over the last two years. The travel time model will be used to estimate travel time for New Jersey Transit bus routes under current levels of congestion and under the levels that existed three years previously. Using the total time required for a round trip under the various congestion levels and service types, in combination with the desired or existing headways, the total number of vehicles required for the route can be determined. Using information on the total number of buses, associated drivers, and travel time and a cost allocation model, the increase in cost due to congestion will be determined.

Task 5: Final Report A final report describing the methods and data used to develop the model and the impacts determined with the model will be prepared. Additionally, quarterly reports describing progress and problems encountered in each quarter will be submitted to New Jersey DOT and NJT.

Student Involvement

Both graduate and undergraduate students will participate in the project tasks. Their involvement will include data collection, data analysis, and computer applications.

Relationship with Other Research Activities

None

Technology Transfer Activities

None

Benefits of the Project

The research proposed below would develop models that would be fitted to the operating environment and conditions that New Jersey Transit encounters.