Revealing the CO2 emission reduction of ridesplitting and its determinants based on real-world data (2204.00777v2)

Abstract: Ridesplitting, which is a form of pooled ridesourcing service, has great potential to alleviate the negative impacts of ridesourcing on the environment. However, most existing studies only explored its theoretical environmental benefits based on optimization models and simulations. By contrast, this study aims to reveal the real-world emission reduction of ridesplitting and its determinants based on the observed data of ridesourcing in Chengdu, China. Integrating the trip data with the COPERT model, this study calculates the CO2 emissions of shared rides (ridesplitting) and their substituted single rides (regular ridesourcing) to estimate the CO2 emission reduction of each ridesplitting trip. The results show that not all ridesplitting trips reduce emissions from ridesourcing in the real world. The CO2 emission reduction rate of ridesplitting varies from trip to trip, averaging at 43.15g/km. Then, interpretable machine learning models, gradient boosting machines, are applied to explore the relationship between the CO2 emission reduction rate of ridesplitting and its determinants. Based on the SHapley Additive exPlanations (SHAP) method, the overlap rate and detour rate of shared rides are identified to be the most important factors that determine the CO2 emission reduction rate of ridesplitting. Increasing the overlap rate, the number of shared rides, average speed, and ride distance ratio while decreasing the detour rate, actual trip distance, and ride distance gap can increase the CO2 emission reduction rate of ridesplitting. In addition, nonlinear effects and interactions of the determinants are examined through the partial dependence plots. To sum up, this study provides a scientific method for the government and ridesourcing companies to better assess and optimize the environmental benefits of ridesplitting.

• The paper calculates CO2 emission reductions by 43.15 g/km using Chengdu's trip data and the COPERT model.
• The paper highlights that emission reductions vary with trip characteristics, underlining non-uniform environmental benefits.
• The paper utilizes gradient boosting machines and SHAP to identify determinants and suggest strategies for optimizing ridesplitting services.

This paper explores the environmental impact of ridesplitting, a form of pooled ridesourcing service, by using real-world data from Chengdu, China. Unlike prior studies that have focused on theoretical benefits using optimization models and simulations, this research leverages observed trip data to calculate the actual CO2 emissions resulting from ridesplitting.

Key Findings:

1. CO2 Emission Reduction Calculation: The paper employs the COPERT model to estimate and compare the CO2 emissions of ridesplitting trips with those of traditional single rides. On average, ridesplitting trips were found to reduce CO2 emissions by 43.15 grams per kilometer.
2. Variation in Emission Reduction: It's noteworthy that not all ridesplitting trips result in CO2 reductions. The emission reduction rate fluctuates based on various trip characteristics, indicating that the benefits of ridesplitting are not uniform.
3. Determinants of CO2 Emission Reduction: Using gradient boosting machines and SHapley Additive exPlanations (SHAP), the paper identifies several key factors that influence the CO2 emission reduction rate of ridesplitting:
   • Positive Factors: High overlap rate (the extent to which shared rides coincide with each other), a greater number of shared rides, higher average speed, and a higher ride distance ratio (ratio of trip distance to the straight-line distance).
   • Negative Factors: High detour rate (extent of deviation from the direct route), longer actual trip distance, and a larger ride distance gap (difference between shared and standalone trip distances).

Methodology:

The paper integrates trip data from Chengdu with the COPERT emission model, enabling precise calculation of emissions for both shared and individual ridesourcing trips. Interpretable machine learning techniques, specifically gradient boosting machines, are employed to analyze the dataset. The SHAP method is used to uncover the relative importance and interaction of various factors affecting emission reductions. Additionally, partial dependence plots are utilized to explore the nonlinear effects of these determinants.

Implications:

The findings suggest that in order to maximize the environmental benefits of ridesplitting, policy-makers and ridesourcing companies should focus on:

• Increasing the overlap rate of shared rides.
• Promoting higher average speeds and appropriate ride distances.
• Minimizing detours and unnecessary deviations during shared trips.

By reflecting on the nonlinear interactions among these determinants, the paper provides a structured approach for optimizing ridesplitting services to achieve significant environmental benefits. This research offers valuable insights that can guide ongoing efforts to enhance the sustainability of urban transportation systems.