Abstract
Recent years have witnessed a sharp decline in national highway trust fund that is primarily used to support infrastructure construction, expansion and retrofit. This is a challenging and critical issue for almost all states, including Pennsylvania. The primary reason is due to the decline of gas tax collected at gas pumps as a result of adaption of high fuel efficiency vehicles. The most recent and ongoing COVID-19 crisis adds additional burden to this lack of public funds for infrastructure, since the overall travel demand declines drastically. Most of states are currently evaluating the implications of tax loss, and proactively developing plan to collect funds in a more equitable and effective manner. The Pennsylvania Department of Transportation has lately given the green light for a proposal that could add tolls on an undetermined number of bridges, as a way to collect funding to support infrastructure. It imposes a challenge how to effectively and accurately evaluate the societal consequences of tolling of various forms, including social equity, congestion delay, emission, fuel use and potential toll revenue. In this research project, we aim to develop a large-scale multi-class network modeling and simulation framework, particularly for DVRPC, that holistically models the spatio-temporal behaviors of private cars, ride-hailing cars, freight trucks, respectively. The result includes the prediction of travel time, travel delay, vehicle-mile-traveled and emissions for each of those vehicle classes and travel modes, either at road and intersection level or averaged at community level by time of day. Potential tolling strategies, such as locations and pricing, can be evaluated and deployed, with the trade-off among tolling revenue, system mobility and social equity.
Description
Recent years have witnessed a sharp decline in national highway trust fund that is primarily used to support infrastructure construction, expansion and retrofit. This is a challenging and critical issue for almost all states, including Pennsylvania. The primary reason is due to the decline of gas tax collected at gas pumps as a result of adaption of high fuel efficiency vehicles. The most recent and ongoing COVID-19 crisis adds additional burden to this lack of public funds for infrastructure, since the overall travel demand declines drastically. Most of states are currently evaluating the implications of tax loss, and proactively developing plan to collect funds in a more equitable and effective manner. The Pennsylvania Department of Transportation has lately given the green light for a proposal that could add tolls on an undetermined number of bridges, as a way to collect funding to support infrastructure. It imposes a challenge how to effectively and accurately evaluate the societal consequences of tolling of various forms, including social equity, congestion delay, emission, fuel use and potential toll revenue.
There exist a number of studies modeling the tolling impact in general networks. However, the main research gaps are: 1) in a general multi-modal transportation system with many different vehicle classes (primarily cars and trucks) and travel modes (primarily private vehicles and ride-hailing vehicles), it is unclear how dynamic tolling impacts the traffic for each class and each mode differently; 2) different communities in a region may be impacted by dynamic tolling differently, leading to social equity issues. It is yet to quantify the influence of tolling to the network performance for each community, and incorporate equity into the decision making of tolling; 3) toll revenue will be used to fund infrastructure projects in the future. Therefore, the pricing and locations of tolls need to optimized to justify the infrastructure projects in the greatest need for a region.
In this research project, we aim to develop a large-scale multi-class network modeling and simulation framework, that holistically models the spatio-temporal behaviors of private cars, ride-hailing cars, freight trucks, respectively. The result includes the prediction of travel time, travel delay, vehicle-mile-traveled and emissions for each of those vehicle classes, travel modes, either at road and intersection level or averaged at traffic zone level by time of day. Potential tolling strategies, such as locations and pricing, can be evaluated and deployed, with the consideration of both system mobility and social equity.
The Philadelphia Metropolitan Region is traffic data rich comparing to other metropolitan areas in the U.S. Various data sets in the Philadelphia region, including traditional traffic sensors (loops, cameras, etc.) and cutting-edge sensors (Bluetooth, GPS probe, ride-hailing vehicle samples, etc.), are available and have been archived for quite long time. The rich data sets allow us to learn travelers’ behavior accurately and develop an in-depth understanding of traffic impact of tolling systems in large-scale networks.
PennDOT is currently looking at ways to implement various tolling strategies to improve the efficiency, reliability and safety of Vine Street expressway corridor, while having the ability to collect public funds for infrastructure construction projects. The tolling along this corridor will involve considerable traffic pattern change, detours and other traffic diversions for both trucks, private vehicles and TNCs. It will also impact bus operations across the bridges to/from Center City. The Center City Bridges tolling presents a great opportunity to conduct research on mobility impact of critical infrastructure tolling and revenue management for public agencies.
This project is a continuation of the research from the Mobility21 projects “Data-driven Network Models for Analyzing Multi-modal Transportation Systems” in FY 2018, and “Mesoscopic car-truck flow modeling and simulation: theory and applications” in FY 2019, both led by PI Qian. It further extends the data-driven multi-modal mesoscopic network modeling and simulation framework on passenger and freight transportation developed in the two previous projects to include the tolling facilities, particularly arbitrary toll locations and costs in general. In the previous two projects, we have developed a simulation framework that can estimate the dynamic origin-destination (OD) demand of multiclass vehicles using traffic counts and speed data, and dynamically simulate the traffic evolution and conditions with consideration of multiclass vehicle interactions. For incorporating tolling into the multi-class simulation framework, we will answer the following questions:
1. How can we determine optimal tolling locations and costs for different time of day and locations?
2. How to simulate the behavior of vehicles of multiple classes and modes in response to dynamic tolls?
Research Approach:
Task 1: Identify various data sources for travel behavior modeling in regional networks
• Request GIS models and regional travel planning model (TIM2.2) from DVRPC where trip information associated in the region can be retrieved.
• Establish a refined GIS model for this research based upon DVRPC model. A stand-alone version of GIS with the following data is necessary for this study, which should include street names, street levels (highway, major arterials, minor streets, alleys, etc.), number of lanes, and speed limit.
• Obtain traffic counts on local streets and highway in the region from DVRPC and the City of Philadelphia. To produce accurate simulation results, 15-min traffic counts by vehicle classifications are necessary.
• Obtain INRIX probe data from RITIS. Those probe data cover highways in the region and major arterials within the region, by vehicle classifications.
• Obtain the modal splits between driving and transit from DVRPC TIM 2.2 model.
• Work with the City of Philadelphia, DVRPC and PennDOT to gather detailed information about all possible tolling plans.
Task 2: Establishing a multi-class dynamic network model for the Philadelphia Metropolitan Region considering tolling
We will use mesoscopic traffic flow models to conduct this research task. We will leverage the dynamic network model that was supported by prior Mobility21 projects to provide estimated day-to-day origin-destination demand among all Traffic Analysis Zones that vary by time of day. The route choices for all travelers in the region will be examined and carefully calibrated using data sets collected in Task 1. The network model is capable of estimating network-wide traffic impact caused by any tolls based upon a generic regional network consisting of freeway and major arterials. It has the capacity of modeling dynamic traffic evolution with the consideration of travel behavior change due to tolling, such as changes in routes and modes. It adopts state-of-the-art traffic models and is much more computationally efficient than other microscopic models that are extremely labor intensive to build.
Task 3: Modeling tolling plans for Center City bridge scenarios
The scenarios of Center City bridge tolling at various locations and at various rates for both cars and trucks will be modeled based on the calibrated regional network model. The simulation adopts the historical traffic demand and their behavioral models from the dynamic network model built in Task 2. For each scenario, the result includes the prediction of travel time, travel delay, vehicle-mile-traveled and emissions for each of those vehicle classes, travel modes, at road and intersection level or averaged at traffic zone level by time of day. More importantly, we will estimate the social equity in terms of accessibility, reliability and travel time/cost change among various pair of main origin/destination points. Ultimately, policy insights of how to make a good tradeoff among tolling revenue, system mobility and social equity will be provided.
The expected outcome of this research is a large-scale multi-modal network modeling and simulation framework that incorporates tolling costs and locations to evaluate revenue, mobility and equity in regional networks. The new simulation framework will be implemented in a prototype web application and will be integrated to an open-source dynamic network analysis toolkit, Mobility Data Analytics Center - Prediction, Optimization, and Simulation (MAC-POSTS).
Upon the completion of this project, we plan to actively seek both industrial and federal funding based on this initial development. Our framework is applicable to any large traffic networks with tolling, charging or tax policies. This generality will attract the attentions from various agencies and private tolling or mobility service companies. Potential funding agencies/collaborators include the Department of Transportation, Federal Highway Administration, National Science Foundation, National Institute of Standards and Technology, and local companies.
Timeline
Task 1: Identify various data sources for travel behavior modeling in regional networks (2 months)
Task 2: Establishing a multi-class dynamic network model for the Philadelphia Metropolitan Region considering tolling (6 months)
Task 3: Modeling tolling plans for Center City bridge scenarios (4 months)
Strategic Description / RD&T
Deployment Plan
We will use the model to simulate traffic evolution in the roadway network as a result of tolling or taxation in the Philadelphia Metro network. This large-scale multi-modal network modeling and simulation framework is capable of quantifying the impact of pricing in general to road traffic and evaluating management strategies about ridesharing. The output will provide the spatio-temporal distributions of both conventional travel modes and trucking mode. We can also obtain the travel time, travel delay, vehicle-mile-traveled and emissions for each road segment and intersection by time of day for each travel mode. More importantly, it allows to quantify the impact to mobility, reliability, tolling revenue and equity, and trade off among those factors. Upon the completion of this project, we plan to actively seek both industrial and federal funding based on this initial development. Our framework is applicable to any large traffic networks with tolling, charging or tax policies. This generality will attract the attentions from various agencies and private tolling or mobility service companies. Potential funding agencies/collaborators include the Department of Transportation, Federal Highway Administration, National Science Foundation, National Institute of Standards and Technology, and local companies.
Each quarter, we will meet with NREL and DVRPC modeling group, to go over the research progress, receive feedback from DVRPC, and improve our models and results.
Expected Outcomes/Impacts
The expected outcome of this research is a large-scale multi-modal network modeling and simulation framework that can quantify the societal impact of tolling and taxation to road traffic, with the presence of private cars, ride-hailing cars, and freight transportation modes. The new simulation framework will be implemented in a prototype web application and will be integrated to an open-source dynamic network analysis toolkit, Mobility Data Analytics Center - Prediction, Optimization, and Simulation (MAC-POSTS). The framework will be tested in the Philadelphia Metro network to validate its accuracy and effectiveness for solving policy questions and supporting decision making in different scenarios.
Expected Outputs
TRID
Individuals Involved
| Email |
Name |
Affiliation |
Role |
Position |
| seanqian@cmu.edu |
Qian, Sean |
CEE |
PI |
Faculty - Untenured, Tenure Track |
| qilingz@andrew.cmu.edu |
Zou, Qiling |
CEE |
Other |
Other |
Budget
Amount of UTC Funds Awarded
$50000.00
Total Project Budget (from all funding sources)
$200000.00
Documents
| Type |
Name |
Uploaded |
| Data Management Plan |
dmp_JBNRfHm.docx |
March 14, 2021, 8:47 p.m. |
| Project Brief |
Qian_DVRPC_2021.pptx |
March 14, 2021, 8:48 p.m. |
| Progress Report |
357_Progress_Report_2021-09-30 |
Sept. 23, 2021, 9:48 p.m. |
| Presentation |
large-scale multi-class network modeling and simulation |
March 16, 2022, 3:31 p.m. |
| Presentation |
large-scale multi-class network modeling and simulation |
March 16, 2022, 3:31 p.m. |
| Progress Report |
357_Progress_Report_2022-03-30 |
March 16, 2022, 3:31 p.m. |
| Publication |
Smart Right-of-Way Permitting System for the City of Pittsburgh: Coordination, Pricing and Enforcement |
April 6, 2022, 5:42 a.m. |
| Publication |
Societal Impacts of a Complete Street Project in a Mixed Urban Corridor: Case Study in Pittsburgh |
April 6, 2022, 5:42 a.m. |
| Publication |
Proactive Management of Mobility Impact of Interdependent Subsurface Utility and Roadway Construction through Incentives |
April 6, 2022, 5:43 a.m. |
| Publication |
Socioeconomic and usage characteristics of transportation network company (TNC) riders |
April 6, 2022, 5:43 a.m. |
| Presentation |
large-scale multi-class network modeling and simulation |
Sept. 29, 2022, 7:57 p.m. |
| Presentation |
large-scale multi-class network modeling and simulation |
Sept. 29, 2022, 7:57 p.m. |
| Presentation |
large-scale multi-class network modeling and simulation |
Sept. 29, 2022, 7:57 p.m. |
| Progress Report |
357_Progress_Report_2022-09-30 |
Sept. 29, 2022, 7:57 p.m. |
| Final Report |
357_-_Final_Report.pdf |
Jan. 5, 2023, 5:17 a.m. |
Match Sources
No match sources!
Partners
| Name |
Type |
| Delaware Valley Regional Planning Commission |
Deployment Partner Deployment Partner |
