Operational Impacts of On-Demand Ride-Pooling Service Options in Birmingham, AL

Furat Salman, Virginia P. Sisiopiku, Jalal Khalil, Wencui Yang, Da YanFuture 2023 · DOI

TL;DR

This study uses a sophisticated simulation model (MATSim) to analyze the impact of ride-pooling services (like Uber Pool and Lyft Line) on traffic operations in Birmingham, AL. It finds that ride-pooling can reduce Vehicle Kilometers Traveled (VKT) and improve service efficiency, suggesting that a fleet of 200 vehicles is optimal for this medium-sized city.

What Problem Does This Paper Address?

Transportation Network Companies (TNCs) have revolutionized urban mobility, but their impact on traffic congestion and overall network efficiency, especially ride-pooling services, remains a complex area. While ride-pooling promises to reduce the number of cars on the road by allowing multiple passengers to share a ride, its actual effect on traffic operations, including Vehicle Kilometers Traveled (VKT), detour distances, and customer service metrics, is not fully understood. This is partly due to the difficulty in obtaining real-world data and the limitations of existing simulation tools. This paper aims to fill this knowledge gap by simulating ride-pooling services in a medium-sized city, Birmingham, AL, to quantify their operational impacts.

Methods

The core of this research lies in the use of the Multi-Agent Transport Simulation (MATSim) platform. MATSim is an agent-based, activity-based microsimulation framework that models individual transportation users (agents) and their daily travel plans. The researchers adapted an existing MATSim model of the Birmingham metro area.

Key methodological aspects include:

  • Simulation of Ride-Pooling Services: The study utilized MATSim's Demand Responsive Transit (DRT) extension, which is specifically designed to simulate on-demand ride-pooling services. This extension allows for the simulation of multiple trip requests being served by a single TNC vehicle.
  • Scenario Design: Six ride-pooling scenarios were developed, varying two key attributes:
  • Service Type: Door-to-door (d2d) and stop-based (sB) ride-pooling.
  • Fleet Size: Three fleet sizes were simulated: 200, 400, and 800 vehicles.
  • Baseline Comparison: A baseline scenario with no TNC services was also simulated to provide a point of comparison for evaluating the impacts of ride-pooling.
  • Performance Metrics: The study analyzed several performance measures, including total daily VKT, ride-pooling VKT, detour distances, vehicle empty ratios, customer wait times, in-vehicle travel times, and ride request rejection rates.

Datasets

The study did not utilize new real-world datasets for training or evaluation. Instead, it relied on:

  • Existing MATSim Model of Birmingham: The research built upon a pre-existing MATSim model of the Birmingham transportation network.
  • Synthetic Population: The study used a synthetic population technique to generate travel plans for agents within the MATSim model.
  • Simulated Data: All results were generated through the simulation of various scenarios within the MATSim environment. The paper notes that obtaining direct TNC trip data from Uber and Lyft was challenging, hence the reliance on simulation.

Key Insights and Ideas

  • Ride-pooling's Potential for VKT Reduction: The most significant insight is that ride-pooling, particularly the door-to-door (d2d) service, can indeed reduce overall Vehicle Kilometers Traveled (VKT) compared to a baseline with no TNCs. This directly addresses the promise of ride-pooling for easing urban congestion.
  • Fleet Size Optimization is Crucial: The study highlights a trade-off with fleet size. While larger fleets reduce rejection rates and improve driver efficiency (lower empty ratios), they also increase total ride-pooling VKT and detour distances. This suggests an "optimal" fleet size is necessary to balance user convenience, driver economics, and network impact.
  • D2D vs. Stop-Based Trade-offs: Door-to-door (d2d) services offer greater convenience to users, leading to higher demand and potentially greater VKT reduction. However, stop-based (sB) services are more efficient from an operator's perspective, resulting in lower VKT and detour distances, though they may be less convenient for users.
  • Medium-Sized City Context Matters: The study's focus on Birmingham, a medium-sized city, is important. Findings from larger cities might not directly translate, and this research provides valuable insights for similar urban areas.

Key Results

  • VKT Reduction: Ride-pooling services led to a reduction in total daily VKT by up to 5.78% for d2d services and up to 2.71% for sB services compared to the baseline.
  • Optimal Fleet Size: A fleet size of 200 ride-pooling vehicles was identified as optimal for Birmingham, capable of meeting current demand while minimizing negative impacts.
  • Fleet Size Impacts:
    • Increasing fleet size decreased the ride request rejection rate (e.g., from 55% to 33% for d2d with 200 vs. 800 vehicles).
  • Increasing fleet size decreased the vehicle empty ratio.
  • Increasing fleet size increased total ride-pooling VKT and detour distance.
  • Service Type Impacts:
    • D2D services generally resulted in higher average customer wait times and in-vehicle travel times compared to sB services.
  • Slightly higher VKT reductions were observed with d2d services.
  • Vehicle Occupancy: The simulations showed that most ride-pooling trips involved single passengers, and vehicles accommodating up to three passengers were sufficient for the Birmingham area's needs.

Limitations and Future Work

  • Generalizability: The study acknowledges that findings are generalizable to settings similar to Birmingham. Further studies in other medium-sized cities are recommended to expand the understanding of ride-pooling impacts.
  • Data Limitations: The reliance on simulated data due to the difficulty in obtaining direct TNC trip data from providers like Uber and Lyft is a limitation. The use of a synthetic population also introduces potential inaccuracies in reflecting real-world travel behaviors.
  • Mode Shift Nuances: While ride-pooling scenarios showed a shift towards transit/walking, larger fleet sizes seemed to reduce this shift, suggesting a complex interplay.
  • Future Work Suggestions:
    • Conducting studies in other medium-sized metro areas.
  • Exploring the impact of higher acceptable wait times on rejection rates.
  • Investigating the dynamic evolution of TNC services and user adoption practices.
  • Encouraging data sharing between TNC providers and public agencies for more robust modeling.
  • Considering the impact of vehicle electrification on TNC operations.

Summary updated

Abstract

Transportation Network Companies (TNCs) use online-enabled apps to provide on-demand transportation services. TNCs facilitate travelers to connect with drivers that can offer them rides for compensation using driver-owned vehicles. The ride requests can be for (a) individual or (b) shared rides. The latter, also known as ride-pooling services, accommodates requests of unrelated parties with origins and destinations along the same route who agree to share the same vehicle, usually at a discounted fare. Uber and Lyft offer ride-pooling services in select markets. Compared to individual ride requests, ride-pooling services hold better promise toward easing urban congestion by reducing the number of automobiles on the road. However, their impact on traffic operations is still not fully understood. Using Birmingham, AL as a case study, this research evaluated the impact that ride-pooling services have on traffic operations using a Multi-Agent Transport Simulation (MATSim) model of the Birmingham metro area. Scenarios were developed to simulate baseline conditions (no TNC service) and ride-pooling availability with two types of ride-pooling services, namely door-to-door (d2d) and stop-based (sB) service and three fleet sizes (200, 400, and 800 vehicles). The results indicate that when TNC vehicles are added to the network, the Vehicle Kilometers Traveled (VKT) decrease by up to 5.78% for the door-to-door (d2d) service, and up to 2.71% for stop-based (sB) services, as compared to the baseline scenario (no TNC service). The findings also suggest that an increase in the size of the ride-pooling fleet results in a rise in total ride-pooling service VKT, network-wide total VKT, and detour distance. However, increasing the size of the ride-pooling fleet also results in a decrease in the ride request rejection rates, thus benefiting the customers and decreasing the vehicle empty ratio which, in turn, benefits the TNC drivers. The results further suggest that a fleet of 200 ride-pooling vehicles can meet the current demand for service in the Birmingham region at all times, thus it is the optimal ride-pooling TNC fleet size for a medium-sized city such as Birmingham.
BibTeX
@inproceedings{salman2023operational,
  author = {Furat Salman and Virginia P. Sisiopiku and Jalal Khalil and Wencui Yang and Da Yan},
  title = {{Operational Impacts of On-Demand Ride-Pooling Service Options in Birmingham, AL}},
  booktitle = {Future},
  year = {2023},
  doi = {10.3390/futuretransp3020030},
}
Figure