SWUTC Research Project Description

Title of Project:  Robust Pricing of Transportation Networks Under Uncertainty

Project Number:  169206

Principal Investigator:
S. Travis Waller
(512) 471-4539
P.I. Affiliation:  University of Texas at Austin
stw@mail.utexas.edu

Project Monitor:
Arash Mirzaei
616 Six Flags Drive
P.O. Box 5888
Arlington, TX  76005-5888
(817) 695-9261

Project Status:  Active

Date Started:  9/1/08

Estimation Completion Date:  8/31/09

Estimated Cost - Current Fiscal:  $39,000

Estimated Cost - Total Planned:  $39,000

Project Summary:
Project Abstract:
Both public and private entities are concerned with the impacts of future toll revenue, and the effects of tolled facilities on system congestion.  Due to the inherent complexity of transportation systems, it is impossible to predict travel demand and congestion conditions exactly, and simplistic attempts to account for this consistently underestimate true levels of congestion.  Thus, in the context of roadway pricing, there is a need to develop mathematical models which explicitly account for both demand and supply uncertainty in both the short-term and long-term time scales.  This project will develop these models, which will be suitable either to determine the best pricing policies to maximize revenue or minimize congestion, or to evaluate alternative toll policies according to these metrics.  Thus, these models will produce more accurate predictions of toll revenues and congestion levels than are available using current methods. 

Project Objectives:
The three primary objectives of this research are as follows:

• Review existing methodologies related to modeling supply- and demand-side uncertainty mathematically.
• Incorporate these techniques into pricing models, allowing both optimal toll policies to be found (satisfying goals such as revenue maximization or congestion minimizatin), and allowing alternative toll policies to be evaluated according to a variety of measures of effectiveness.
• Examine the suitability of these models for a variety of applications, such as prediction of toll revenues, transportation planning, and alternative analysis involving BOT projects.

Task Descriptions:
Task 1.  Review of Past Research on Supply and Demand Uncertainty
A careful study will be made of existing research on uncertainty in transportation systems, both of its causes and its effects.  This will include examination of both travel demand modeling literature, research into traffic operations, and into travelers’ behavior in the face of information and uncertain conditions.  These will be of use in determining the most appropriate and useful scope for this project.

Task 2.  Identify Relevant Sources of Uncertainty and Measures of Effectiveness
Guided by the literature identified in Task 1, the uncertain factors which most impact congestion pricing policies will be identified.  Further, a suite of metrics will be identified to quantify the effects of a given toll policy in an uncertain setting (including, but not limited to, the mean and standard deviation of travel times and revenues).  Together with the identification of key factors, the scope of the project will be explicitly delineated based on the state-of-the-art in modeling.

Task 3.  Develop Demand-Side Modeling Techniques
In this task, the research team will create mathematical models which allow variations in both short-term and long-term demand to be accounted for in developing toll policies.  These models will draw on concepts such as deterministic or stochastic user equilibrium, first- and second-best congestion pricing, and day-to-day evolution of flows, and will generate optimal (or approximately optimal) tolling schemes in response to demand uncertainty.

Task 4.  Develop Supply-Side Modeling Techniques
In parallel with Task 3, the research team will also develop independent techniques to account for disruptions in roadway capacity.  In addition to the concepts of equilibrium and pricing used in Task 3, research on safety and incident management may also be included to ensure that the models accurately reflect these phenomena.  These models will be unified in the succeeding tasks, but developing them in isolation allows greater study of the different impacts of these factors.

Task 5.  Integrate Demand- and Supply-Side Uncertainty into Pricing Models
This task combines the results of Tasks 3 and 4 by developing two models for short-term and long-term pricing policies, allowing simultaneous consideration of the effects of supply and demand uncertainty.  The primary challenge will be incorporating the interaction between these two effects, which must be done in a rigorous and realistic manner.

Task 6.  Incorporate Responsive Tolling and User Information
Building on Task 5, in this task, the research team will explore the impacts of responsive tolling and user information, extending the short-term and long-term models to account for these possibilities, allowing identification of opportunities to improve system conditions, revenues, or other measures of effectiveness.

Task 7.  Simulation of Results
To examine the applicability and transferability of these models, the research team will perform simulation tests of the models developed in Tasks 5 and 6.  The exact scope of these tests will depend on the models which are developed, but a network representing a major metropolitan area (such as the cities of Austin or El Paso) will be used.

Task 8.  Final report
A comprehensive final report will be written, describing all research work performed.

Index Terms:
Tolls, Rates, fares and tolls, Congestion pricing, Traffic congestion, Transit station congestion, Revenues, Transit operating agencies, Public transit, Research projects