Abstract
The rapid growth of new transportation options in the United States has been met with increasing use of these expanded modes of transportation in conjunction with traditional mass transit. However, current engineering models of mass transit systems do not reflect these trends. Existing models tend to 1) only consider traditional public transit (e.g., fix-route buses), and 2) model choices and accessibility for a representative user, with inadequate attention paid to the choices and trade-offs faced by vulnerable populations. The proposed work will advance fundamental knowledge in transportation network modeling by incorporating emerging mobility options, such as shared mobility services, micro-transit, and micro-mobility. This project will develop models, algorithms, and systems used to identify affordable, reliable, and efficient routes that leverage diverse mobility options coupled with conventional mass transit, and ultimately optimize their provisions to low-income populations to improve their accessibility to necessities and essential services.
Description
The rapid growth of new transportation options in the United States has been met with increasing use of these expanded modes of transportation in conjunction with traditional mass transit. However, current engineering models of mass transit systems do not reflect these trends. Existing models tend to 1) only consider traditional public transit (mostly fix-route buses), and 2) model choices and accessibility for a representative user, with inadequate attention paid to the choices and trade-offs faced by vulnerable populations.
To fully understand the potential implications of these deficiencies, consider, for example, the low labor force participation rate in the United States among the poorly educated, which is less than 60% among those with a high school degree and less than 50% among those without a high school degree. One factor driving these low participation rates is the high transportation costs, relative to income, that low-income residents incur when seeking employment outside of their immediate neighborhoods. Relative costs are high because low-income urban residents tend to live far from the city districts where job creation is concentrated, and do not have access to affordable mobility options.
The proposed work will advance fundamental knowledge in transportation network modeling by incorporating emerging mobility options, such as shared mobility services, micro-transit, and micro-mobility. This project will develop models, algorithms, and systems used to identify affordable, reliable, and efficient routes that leverage diverse mobility options coupled with conventional mass transit, and ultimately optimize their provisions to low-income populations to improve their accessibility to necessities and essential services. Modeling mass transit systems and their interaction with a range of complementary transit technologies will support multi-modal transportation strategies that can further expand mass transit ridership and resident mobility. In particular, this research will focus on the current mobility options among public transit and shared mobility services, including buses, shared bikes, CommuteInfo, Uber/Lyft, RideACTA, Spin, among planned on-demand feeder mobility services. Port Authority’s APC-AVL, Department of Human Services’ customer data, and large-scale GPS data from mobility service providers will be used to understand those mobility options and demand patterns. This data will then be integrated into dynamic network models for a large-scale regional network to derive the accessibility, equity, and potential barriers for low-income populations seeking opportunity employment. These models will provide public agencies with the tools needed to identify and optimize changes in multi-modal transportation networks, thereby increasing ridership without sacrificing access to current riders, increasing revenue for the transit system, and maximizing mobility/accessibility for a targeted population (e.g., population with high-school education looking for opportunity employment, and low-income mothers).
Task 1: Develop multi-modal transportation network models
As previously described, no single transportation option is likely to constitute a “magic bullet” solution given the plethora of transportation challenges facing the target demographic group. Consequently, the general approach will comprise the development of a multi-modal transportation network model that integrates all available mobility options with their respective coverages over time and space. Based on an investigation of current mobility options among public transit and shared mobility services in the Pittsburgh area, data sources, which are being provided through Carnegie Mellon University’s Mobility Data Analytics Center, will include Port Authority of Allegheny County ridership and route structure data in the form of Automatic Passenger Counting (APC) and Automatic Vehicle Location (AVL) data; CommuteInfo, which is a nonprofit that helps establish carpools and vanpools throughout Southwestern Pennsylvania; MovePGH, which is a public-private partnership that connects the City of Pittsburgh to a consortium of innovative transportation services providers such as Waze Carpooling, Spin (electric scooter ridesharing company), and Healthy Ride (City of Pittsburgh’s public bike sharing program); Uber, which is the region’s leading provider of ridehailing services (will need to request data); and Ride ACTA, which is a micro transit agency serving Pittsburgh’s airport corridor, a region with large numbers of vibrant businesses that could offer attractive employment opportunities for many city residents. For the target population (e.g., low-income mothers), this data will be integrated into a dynamic network model for large-scale regional networks in order to quantify the accessibility, equity, and potential barriers for members of this population seeking opportunity and employment.
Task 2: Develop algorithms to find optimal multi-modal paths for disadvantaged population
One of the ways better transportation can promote significant socioeconomic mobility is by providing disadvantaged citizens with a pathway to higher levels of skill. Given the dynamic system model, when identifying a best route on the multi-modal transportation network for a target population, two problems will be addressed. First, a utility function for the target population will be defined to make route choices among all options, given an origin, a destination, and a departure time of day. The utility includes factors such as time, cost, reliability, job-related position availability, and training-related facility availability. Second, additional constraints that are specific to the target population will need to be considered. For example, time constraints and childcare for the case of low-income mothers.
Task 3: Estimate the impact of expanded mobility options on accessibility outcomes
Once a multi-modal transportation network considering expanded mobility options is established, models will be developed to estimate the change and improvement in accessibility when some mobility options are expanded (e.g., Ride ACTA). We will identify several communities (with the input from the Port Authority of Allegheny County) who are in the greatest need of accessibility improvement to expand mobility options hypothetically. Change in travel time, reliability, cost and accessibility will be used to assess and derive insights on the effectiveness of those expanded options.
Task 4: Optimize the provisions of multiple mobility options to complement the public transit
Using the insights derived from Task 3, an optimization model will be developed to optimally add additional mobility services to maximize the social benefits, given limited financial resources.
Upon the completion of this project, the research team plans to actively seek both industry and federal funding based on this initial development. The developed framework will be applicable to any city networks supporting expanded mobility services. For communities that do not currently support such technological transportation infrastructure, the ability to estimate the impact of expanded mobility options on accessibility outcomes is a powerful mechanism for initiating capital investments in these modes of transportation. Additionally, while the proposed work can (as is emphasized by the application space considered by the proposal) enhance access to work and training, it can also be used to measurably enhance access to other essential services such as fresh food, parks and recreation, and social infrastructure. The team will assess the accessibility to those additional resources in the Pittsburgh regional network as well.
Timeline
Task 1: Develop multi-modal transportation network models (4 months)
Task 2: Develop algorithms to find optimal multi-modal paths for disadvantaged population (4 months)
Task 3: Estimate the impact of expanded mobility options on accessibility outcomes (2 months)
Task 4: Optimize the provisions of multiple mobility options to complement the public transit (2 months)
Strategic Description / RD&T
Deployment Plan
This project will develop models, algorithms, and systems used to identify affordable, reliable, and efficient routes that leverage diverse mobility options coupled with conventional mass transit, and ultimately optimize their provisions to low-income populations to improve their accessibility to necessities and essential services. This project will focus on low-income mothers in the Pittsburgh region who are seeking pathways to higher levels of skill as a testbed. Once a multi-modal transportation network considering expanded mobility options is established, models will be developed to estimate the change and improvement in accessibility when some mobility options are expanded (e.g., Ride ACTA). We will identify several communities (with the input from the Port Authority of Allegheny County) who are in the greatest need of accessibility improvement to expand mobility options hypothetically. Change in travel time, reliability, cost and accessibility will be used to assess to derive insights on the effectiveness of those expanded options. An optimization model will then be developed to optimally add additional mobility services to maximize the benefits, given limited financial resources. While this project will consider Pittsburgh as a test case, the developed framework will be applicable to any city networks supporting expanded mobility services. For communities that do not currently support such technological transportation infrastructure, the ability to estimate the impact of expanded mobility options on accessibility outcomes is a powerful mechanism for initiating capital investments in these modes of transportation. Additionally, while the proposed work can (as is emphasized by the application space considered by the proposal) enhance access to work and training, it can also be used to measurably enhance access to other essential services such as fresh food, parks and recreation, and social infrastructure. This generality will attract the attention of both industry and federal organizations.
Expected Outcomes/Impacts
The expected outcome of this research is a dynamic network model for large-scale regional networks that advances fundamental knowledge in transportation network modeling by incorporating emerging mobility options, such as shared mobility services, micro-transit, and micro-mobility. For a target population, the model will be used to 1) identify communities that are in the greatest need of accessibility improvement; 2) quantify the accessibility, equity, and potential barriers for members of this population seeking opportunity and employment; 3) identify affordable, reliable, and efficient routes by leveraging diverse mobility options coupled with conventional mass transit to optimize provisions; 4) estimate the change and improvement in accessibility when some mobility options are expanded, where change in travel time, reliability, cost and accessibility will be used to assess and derive insights on the effectiveness of those expanded options; and 5) optimally add additional mobility services to maximize the benefits, given limited financial resources. As a case study, we will specifically consider low-income mothers in the Pittsburgh region who are seeking pathways to higher levels of skill. For the test region, focus will be placed on the current mobility options among public transit and shared mobility services, including buses, shared bikes, CommuteInfo, Uber/Lyft, RideACTA, Spin, among planned on-demand feeder mobility services. Port Authority’s APC-AVL, Department of Human Services’ customer data, and large-scale GPS data from mobility service providers will be used to understand those mobility options and demand patterns.
The research team will communicate all intellectual contributions with the scientific community through journal publications, conference presentations/papers, seminars, and UTC/Mobility21 partner/advisory council meetings.
Expected Outputs
TRID
Individuals Involved
| Email |
Name |
Affiliation |
Role |
Position |
| sbenicky@andrew.cmu.edu |
Benicky, Sheryl |
Carnegie Mellon University |
Other |
Other |
| kaflanig@cmu.edu |
Flanigan, Katherine |
Carnegie Mellon University |
PI |
Faculty - Untenured, Tenure Track |
| lgraff@andrew.cmu.edu |
Graff, Lindsay |
Carnegie Mellon University |
Other |
Student - PhD |
| kh3m@andrew.cmu.edu |
Lightman, Karen |
Carnegie Mellon University |
Other |
Other |
Budget
Amount of UTC Funds Awarded
$100001.00
Total Project Budget (from all funding sources)
$200000.00
Documents
| Type |
Name |
Uploaded |
| Data Management Plan |
DataManagementPlan_K5wHKJa.pdf |
Dec. 7, 2022, 11:39 p.m. |
| Presentation |
Project_Overview_Slides.pptx |
Dec. 7, 2022, 11:55 p.m. |
| Publication |
Measuring time-dependent accessibility with emerging mobility options: a generic multi-modal network modeling framework |
March 30, 2023, 8:15 p.m. |
| Presentation |
Measuring time-dependent accessibility with emerging mobility options: a generic multi-modal network modeling framework |
March 30, 2023, 8:15 p.m. |
| Presentation |
A multimodal network modeling framework to evaluate time-dependent accessibility under generalized travel costs |
March 30, 2023, 8:15 p.m. |
| Progress Report |
394_Progress_Report_2023-03-30 |
March 30, 2023, 8:15 p.m. |
| Publication |
Functional Requirements Enabling Levels of Predictive Maintenance Automation and Autonomy |
April 10, 2023, 8:54 p.m. |
| Publication |
Optimal event-based policy for remote parameter estimation in wireless sensing architectures under resource constraints |
April 10, 2023, 8:55 p.m. |
| Publication |
Smart and equitable parks: quantifying returns on investments based on probabilistic mobility-dependent correlates of park usage using cyber-physical system technologies |
April 10, 2023, 8:55 p.m. |
| Publication |
Community engagement using urban sensing: technology development and deployment studies |
April 10, 2023, 8:56 p.m. |
| Final Report |
Final_Report_-_Flanigan_394.pdf |
Sept. 15, 2023, 8:04 a.m. |
Match Sources
No match sources!
Partners
| Name |
Type |
| Pittsburgh Parks Conservancy |
Deployment Partner Deployment Partner |
| Metro21 |
Deployment Partner Deployment Partner |
