Login

Project

#58 Integration of parking data across mixed-density suburban communities in Pittsburgh


Principal Investigator
Sean Qian
Status
Completed
Start Date
Sept. 1, 2017
End Date
Dec. 31, 2019
Project Type
Research Advanced
Grant Program
Private Funding
Grant Cycle
2017 Smart Mobility Challenge
Visibility
Public

Abstract

Dormont and Mount Lebanon are two adjacent and critical municipalities in the southern suburbs of Pittsburgh, which happen to share parking, transit, traffic signaling, and congestion management challenges across a shared Business Rt 19 corridor. We propose to work collaboratively with them over time to develop innovative solutions in a shared corridor, but will initially focus on parking data analysis and management. We will produce a web application that can provide parking information across the municipal boundaries, and consider pricing changes to enhance revenue.    
Description
Dormont and Mount Lebanon are two adjacent and critical municipalities in the southern suburbs of Pittsburgh. They comprise a significant portion of domiciled commuters to the city, but also share portions of a key commuting corridor to downtown - business Rt. 19 (aka West Liberty Avenue / Washington Rd) - that is used by many residents in the south hills.  Rt. 19 spans approximately one mile in the two communities.

The two municipalities also have geographic and demographic profiles that are promising for such a case study.  Business Rt. 19 bisects each community, and Dormont density on par with downtown - with about 9,000 residents in 0.7 square miles.  Mount Lebanon is much larger (33,000 people in 6 square miles) and has a density similar to Dormont on the adjacent end but is more suburban as Rt. 19 progresses towards the south.  In both communities, Business 19 forms the core central business district of the community. 

This one mile stretch has approximately 20 signalized intersections, and varies from one-lane bidirectional traffic with shoulder parking on both sides to two-lane bidirectional traffic.  Parking in both communities is at a premium during various times of the day, leading to additional traffic problems for both residents as well as visitors. The two communities use different vendors for parking payment and data collection services (MeterFeeder and Streetline) and have access to multiple years of historical data. They also have different technologies and vendors for their existing signal management efforts.

This corridor can become very congested during morning and afternoon rush hours, which also disrupts the efficiency and patronage of local businesses. There is relatively high use of public transit (20-30% in the two communities), of which there are 5 existing “T” light rail stations near or along Rt. 19. The presence of this existing infrastructure creates another compelling reason for supporting analysis in this corridor.

A final justification for the social aspects of this project is that both Dormont and Mt. Lebanon, given their history and size, are walking communities. Many school students walk in Dormont to elementary schools, and all students in Mt. Lebanon from K-12 walk to school (no bus service is provided) and most of the schools are on or very close to the corridor. Beyond adding to the “walking community” aspect, this means that any changes to the transportation system in this corridor must ensure that public safety is maintained, i.e., that vehicular speeds are not increased as a result of increasing efficiency.

Given this status quo, we envision this smart mobility challenge project as a first step in a longer-term collaboration between CMU and the two municipalities that will address parking, traffic signals, transit, and other aspects that could have revolutionary effects on managing the Business Rt. 19 corridor and that could be leveraged to other municipal corridors. However in this project we have decided to begin with parking.

A key innovation feature will be shared access to, and visualization of, parking data across the municipal boundaries. This is especially important for Dormont, which lacks the Streetline system and thus cannot easily provide direct information to users. Generally, though, the system would be able to predict or identify available spots, even if the parking exists “across the municipal boundary”. We will also perform detailed analyses of parking supply and demand to estimate whether increases or decreases of rates during peak or off-peak times might lead to better utilization and/or increased revenue to the municipalities. This may also include “jump rates” where the first N minutes are at a certain rate, but increase as vehicles remain in the same spots (to dissuade drivers from staying for long periods).

To accomplish these goals, we will leverage and improve upon an existing web-based parking information system that visualizes parking availability and predicts availability, as we believe this will be the most suitable initial solution for a project aimed to be realized within one year. We will also perform data analytics on historical data for the economic and finance aspects of the project.
Timeline
This project was initiated in September 2017 as proposed, but significant data challenges were identified in acquiring the data on parking from the two communities at the needed resolution.  The Mount Lebanon data, while robust, was only available at the weekly level and by block (not by space).  While not ideal, we could have worked around this problem.  The Dormont challenge was more significant. In short, we have been unable to make connections with MeterFeeder to acquire the Dormont data, and compounding this challenge is that MeterFeeder owes Dormont about 6 months of toll revenue, which has somewhat soured the relationship between the two entities.

In early 2018, we decided to attempt an alternative project, and are in the midst of assessing feasibility of that project.  If we do change scopes, we will re-write all of the sections above, but in the meantime a brief summary is provided here.

Rt 19, the corridor which forms the backbone of this project (and of these two communities) is a state-owned road and the municipalities thus have only limited control of its design, upkeep, signalization, etc. Several years ago Mount Lebanon's part of Rt 19 was upgraded by the State with new signals, timings, crossings, etc.  PennDOT has just announced that a similar project will occur in the Dormont portion of Rt 19 in about 2 years.  The municipalities are both "walking and pedestrian" communities, and have business districts along Rt 19. They are thus both concerned that PennDOT will prioritize throughput of traffic over citizen and pedestrian safety. We seek to do a before/after study of the Mount Lebanon Rt 19 project, to then build a model to help Dormont assess the likely affects of a similar project in their area.  We are currently trying to communicate with PennDOT to get "before and after" data, as well as project signalization design and timing data for both projects, to be able to do this. We hope to build a model that is able to predict the effects of PennDOTs design in Dormont, and also to create several alternative designs that could ensure safety of pedestrians, bikers, etc.
Strategic Description / RD&T

    
Deployment Plan
(This will be updated when scope changes are assessed)

We expect to analyze existing historical data, as well as physical parking facilities, from September-November 2017. We will in parallel begin to develop new web or mobile phone based solutions to provide information to prospective parking customers for the two municipalities in late fall 2017. We will begin testing to users in spring 2018. Two PhD students will separately provide a portion of their effort for these projects.
Expected Outcomes/Impacts
(This will be updated when scope changes are assessed)

We expect to complete an analysis of parking rates for increasing revenue by the end of 2017, and the integrated web-based parking data provision solution in spring 2018.
Expected Outputs

    
TRID


    

Individuals Involved

Email Name Affiliation Role Position
hsm@cmu.edu Matthews, H. Scott Carnegie Mellon University Co-PI Faculty - Tenured
seanqian@cmu.edu Qian, Sean Carnegie Mellon University PI Faculty - Untenured, Tenure Track

Budget

Amount of UTC Funds Awarded
$55000.00
Total Project Budget (from all funding sources)
$55000.00

Documents

Type Name Uploaded
Progress Report 58_Progress_Report_2018-09-30 Sept. 30, 2018, 6:39 a.m.
Final Report 58_-_Final_Report.pdf Jan. 27, 2020, 8:20 a.m.
Publication Turning meter transactions data into occupancy and payment behavioral information for on-street parking. Dec. 2, 2020, 9:05 a.m.
Publication A deep learning approach to real-time parking occupancy prediction in transportation networks incorporating multiple spatio-temporal data sources. Dec. 2, 2020, 9:16 a.m.
Publication A general formulation for multi-modal dynamic traffic assignment considering multi-class vehicles, public transit and parking. Dec. 2, 2020, 9:18 a.m.
Publication Predicting Occupancy of Parking Spaces in Transportation Networks: A Deep Learning Approach with Multi-Source Spatio-Temporal Data Dec. 2, 2020, 9:25 a.m.

Match Sources

No match sources!

Partners

No partners!