How do travel characteristics of ridesplitting affect its benefits in emission reduction? evidence from Chengdu
Zhe Zhang,Kun Gao,Hong-Di He,Jin-Ming Yang,Ruo Jia,Zhong-Ren Peng
DOI: https://doi.org/10.1016/j.trd.2023.103912
IF: 7.041
2023-09-16
Transportation Research Part D Transport and Environment
Abstract:Ridesplitting, a shared mobility service, has the potential to reduce traffic-related air pollution. This study evaluates the impacts of ridesplitting on reducing different types of emissions and investigate how travel characteristics of ridesplitting affect emission reduction based on a ridesourcing dataset in Chengdu, China. First, this study quantifies the influence of ridesplitting on emissions reduction compared to single ride (i.e., non-ridesplitting) for each trip. The results indicate that a ridesplitting trip averagely reduce CO 2 , CO, NOx, and HC emissions by 34.52%, 5.98%, 33.10%, and 13.42%, respectively. Subsequently, using explainable machine learning, we quantitatively analyze how the travel characteristics of ridesplitting affect emission reduction at two levels. At the trip level, shared travel distance, shared travel time, delay, and detour are important factors for emission reduction. At the grid level, the number of orders that match co-riders within the same spatial community is more important than the total number of orders.
transportation,transportation science & technology,environmental studies