Sharing Behavior in Ride-hailing Trips: A Machine Learning Inference Approach (2201.12696v1)

Abstract: Ride-hailing is rapidly changing urban and personal transportation. Ride sharing or pooling is important to mitigate negative externalities of ride-hailing such as increased congestion and environmental impacts. However, there lacks empirical evidence on what affect trip-level sharing behavior in ride-hailing. Using a novel dataset from all ride-hailing trips in Chicago in 2019, we show that the willingness of riders to request a shared ride has monotonically decreased from 27.0% to 12.8% throughout the year, while the trip volume and mileage have remained statistically unchanged. We find that the decline in sharing preference is due to an increased per-mile costs of shared trips and shifting shorter trips to solo. Using ensemble machine learning models, we find that the travel impedance variables (trip cost, distance, and duration) collectively contribute to 95% and 91% of the predictive power in determining whether a trip is requested to share and whether it is successfully shared, respectively. Spatial and temporal attributes, sociodemographic, built environment, and transit supply variables do not entail predictive power at the trip level in presence of these travel impedance variables. This implies that pricing signals are most effective to encourage riders to share their rides. Our findings shed light on sharing behavior in ride-hailing trips and can help devise strategies that increase shared ride-hailing, especially as the demand recovers from pandemic.

• The paper finds that shared ride requests declined from 27.0% to 12.8% in 2019 while overall trip volumes remained stable.
• By employing ensemble machine learning, the study shows that travel impedance factors like cost, distance, and duration provide over 90% of the predictive power in ride-sharing behavior.
• The findings suggest that adjusting per-mile costs and pricing strategies could enhance shared ride adoption and inform urban transportation planning.

The paper "Sharing Behavior in Ride-hailing Trips: A Machine Learning Inference Approach" provides an in-depth analysis of the factors influencing the willingness of riders to opt for shared rides in ride-hailing services. The paper utilizes a comprehensive dataset encompassing all ride-hailing trips in Chicago in 2019, revealing significant trends and insights into ride-sharing dynamics.

Key Findings:

1. Decline in Sharing Requests:
   • Over the course of 2019, the willingness of riders to request shared rides decreased significantly from 27.0% to 12.8%.
   • This decline was observed despite the trip volume and mileage remaining statistically stable.
2. Influence of Travel Impedance Variables:
   • The paper identifies that travel impedance variables—specifically trip cost, distance, and duration—are the primary factors driving ride-sharing behavior.
   • Through the application of ensemble machine learning models, it was found that these variables contribute 95% and 91% of the predictive power in determining the likelihood of requesting a shared ride and the successful completion of sharing, respectively.
3. Impact of Per-mile Costs:
   • An increased per-mile cost for shared trips and a shift towards shorter solo trips are cited as primary reasons for the decline in ride-sharing preference.
4. Limited Influence of Other Variables:
   • Other potential factors, including spatial and temporal attributes, sociodemographic characteristics, built environment, and transit supply variables, were found to have negligible predictive power at the trip level when travel impedance variables are accounted for.

Implications:

• Pricing Strategies:
  • The findings suggest that pricing mechanisms are most effective in encouraging riders to share their trips. This insight is critical for devising strategies that could enhance ride-sharing adoption, particularly important as the demand for ride-hailing services rebounds post-pandemic.
• Urban Transportation Planning:
  • The paper provides empirical evidence that can influence urban transportation planning and policy-making focused on reducing congestion and environmental impacts through increased ride-sharing.
• Machine Learning in Transportation Research:
  • By employing ensemble machine learning models, the research demonstrates the utility of advanced analytical techniques in uncovering complex behavioral patterns in urban transportation.

Overall, this paper adds valuable empirical insights and methodological advancements to the paper of ride-sharing behavior in ride-hailing services, emphasizing the central role of cost-related variables in shaping rider preferences.