An official website of the United States government
Ride-sourcing services play an increasingly important role in meeting mobility needs in many metropolitan areas. Yet, aside from delivering passengers from their origins to destinations, ride-sourcing vehicles generate a significant number of vacant trips from the end of one customer delivery trip to the start of the next. These vacant trips create additional traffic demand and may worsen traffic conditions in urban networks. Capturing the congestion effect of these vacant trips poses a great challenge to the modeling practice of transportation planning agencies. With ride-sourcing services, vehicular trips are the outcome of the interactions between service providers and passengers, a missing ingredient in the current traffic assignment methodology. In this paper, we enhance the methodology by explicitly modeling those vacant trips, which include cruising for customers and deadheading for picking up them. Because of the similarity between taxi and ride-sourcing services, we first extend previous taxi network models to construct a base model, which assumes intranode matching between customers and idle ride-sourcing vehicles and thus, only considers cruising vacant trips. Considering spatial matching among multiple zones commonly practiced by ride-sourcing platforms, we further enhance the base model by encapsulating internode matching and considering both the cruising and deadheading vacant trips. A large set of empirical data from Didi Chuxing is applied to validate the proposed enhancement for internode matching. The extended model describes the equilibrium state that results from the interactions between background regular traffic and occupied, idle, and deadheading ride-sourcing vehicles. A solution algorithm is further proposed to solve the enhanced model effectively. Numerical examples are presented to demonstrate the model and solution algorithm. Although this study focuses on ride-sourcing services, the proposed modeling framework can be adapted to model other types of shared use mobility services.
more »
« less
Gravity Model of Passenger and Mobility Fleet Origin–Destination Patterns with Partially Observed Service Data
Mobility-as-a-service systems are becoming increasingly important in the context of smart cities, with challenges arising for public agencies to obtain data from private operators. Only limited mobility data are typically provided to city agencies, which are not enough to support their decision-making. This study proposed an entropy-maximizing gravity model to predict origin–destination patterns of both passenger and mobility fleets with only partial operator data. An iterative balancing algorithm was proposed to efficiently reach the entropy maximization state. With different trip length distributions data available, two calibration applications were discussed and validated with a small-scale numerical example. Tests were also conducted to verify the applicability of the proposed model and algorithm to large-scale real data from Chicago transportation network companies. Both shared-ride and single-ride trips were forecast based on the calibrated model, and the prediction of single-ride has a higher level of accuracy. The proposed solution and calibration algorithms are also efficient to handle large scenarios. Additional analyses were conducted for north and south sub-areas of Chicago and revealed different travel patterns in these two sub-areas.
more »
« less
- Award ID(s):
- 1652735
- PAR ID:
- 10281680
- Publisher / Repository:
- SAGE Publications
- Date Published:
- Journal Name:
- Transportation Research Record: Journal of the Transportation Research Board
- Volume:
- 2675
- Issue:
- 6
- ISSN:
- 0361-1981
- Format(s):
- Medium: X Size: p. 235-253
- Size(s):
- p. 235-253
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Urban dispersal events occur when an unexpectedly large number of people leave an area in a relatively short period of time. It is beneficial for the city authorities, such as law enforcement and city management, to have an advance knowledge of such events, as it can help them mitigate the safety risks and handle important challenges such as managing traffic, and so forth. Predicting dispersal events is also beneficial to Taxi drivers and/or ride-sharing services, as it will help them respond to an unexpected demand and gain competitive advantage. Large urban datasets such as detailed trip records and point of interest ( POI ) data make such predictions achievable. The related literature mainly focused on taxi demand prediction. The pattern of the demand was assumed to be repetitive and proposed methods aimed at capturing those patterns. However, dispersal events are, by definition, violations of those patterns and are, understandably, missed by the methods in the literature. We proposed a different approach in our prior work [32]. We showed that dispersal events can be predicted by learning the complex patterns of arrival and other features that precede them in time. We proposed a survival analysis formulation of this problem and proposed a two-stage framework (DILSA), where a deep learning model predicted the survival function at each point in time in the future. We used that prediction to determine the time of the dispersal event in the future, or its non-occurrence. However, DILSA is subject to a few limitations. First, based on evidence from the data, mobility patterns can vary through time at a given location. DILSA does not distinguish between different mobility patterns through time. Second, mobility patterns are also different for different locations. DILSA does not have the capability to directly distinguish between different locations based on their mobility patterns. In this article, we address these limitations by proposing a method to capture the interaction between POIs and mobility patterns and we create vector representations of locations based on their mobility patterns. We call our new method DILSA+. We conduct extensive case studies and experiments on the NYC Yellow taxi dataset from 2014 to 2016. Results show that DILSA+ can predict events in the next 5 hours with an F1-score of 0.66. It is significantly better than DILSA and the state-of-the-art deep learning approaches for taxi demand prediction.more » « less
-
Public transit in the U.S. has an unsettled future. The onset of the COVID-19 pandemic saw a dramatic decline in transit ridership, with agency operations, and user perceptions of safety changing significantly. However, one new factor beyond the control of agencies is playing an outsized role in transit ridership: the shifting employment patterns in the hybrid work era. Indeed, a lasting and widespread adoption of telework has emerged as a key determinant of individual transit behaviors. This study investigates the impact of teleworking on public transit ridership changes across the different transit services in the Chicago area during the pandemic, employing a random forest machine learning approach applied to large-scale survey data (n = 5637). The use of ensemble machine learning enables a data-driven investigation that is tailored for each of the three main transit service operators in Chicago (Chicago Transit Authority, Metra, and Pace). The analysis reveals that the number of teleworking days per week is a highly significant predictor of lapsed ridership. As a result, commuter-centric transit modes—such as Metra—saw the greatest declines in ridership during the pandemic. The study's findings highlight the need for transit agencies to adapt to the enduring trend of teleworking, considering its implications for future ridership and transportation equity. Policy recommendations include promoting non-commute transit use and addressing the needs of demographic groups less likely to telework. The study contributes to the understanding of how telework trends influence public transit usage and offers insights for transit agencies navigating the post-pandemic world.more » « less
-
During COVID-19 lockdowns, transit agencies need to respond to the decline in travel but also maintain the essential mobility of transit-dependent people. However, there are a few lessons that scholars and practitioners can learn from. Using highway traffic data in the Twin Cities, this study applies a generalized additive model to explore the relationships among the share of low-income population, transit service, and highway traffic during the week that occurred right after the 2020 stay-at-home order. Our results substantiate that transportation impacts are spread unevenly across different income groups and low-income people are less able to reduce travel, leading to equity concerns. Moreover, transit supply influences highway traffic differently in areas with different shares of low-income people. Our study suggests that transportation agencies should provide more affordable travel options for areas with concentrated poverty during lockdowns. In addition, transit agencies should manage transit supply strategically depending on the share of low-income people to better meet people’s mobility needs.more » « less
-
Electrifying the ride-hailing services has the potential to significantly reduce greenhouse gas emissions in the shared mobility sector. However, these emission reduction benefits depend on the utilization of EVs to serve trip requests, especially during the fleet electrification process. In this paper, we evaluated the performance and emission impacts of ride-hailing service with three dispatching policies and various EV penetration levels in the ride-hailing fleet. A large-scale simulation platform was developed for the city of San Francisco in SUMO to enable the application of ride-hailing, electric vehicle charging, and idle vehicle repositioning. Simulation results indicate that with a 60% EVs in the simulated fleet, the off-peak EV priority policy and off-peak EV only policy can reduce CO2 emissions by 32% - 40% while preserving the mobility performance in terms of deadheading, total travel distance, and average rider pick-up time. It is important for ride-hailing platforms to increase the zero-emission rides and encourage ride pooling to comply with California’s Clean Miles Standard.more » « less
