A Two-Stage Dispatching Approach for One-to-many Ride-Sharing with Sliding Time Windows

Yongwu Liu,Binglei Xie,Gangyan Xu,Jinqiu Zhao,Tianyu Li

DOI: https://doi.org/10.1007/s00521-024-09631-z

2024-01-01

Abstract:Ride-sharing has transformed people’s travel habits with the development of various ride-sharing platforms, which can enhance the utilization of transportation resources, alleviate traffic congestion, and reduce carbon emissions. However, the development of a general and efficient matching framework is challenging due to the dynamic real-time conditions and uncertainty of ride-sharing problems in the real world. Additionally, previous research has identified limitations in terms of model practicability and algorithmic solution speed. To address these issues, a two-stage dispatching approach for one-to-many ride-sharing with sliding time windows is proposed. The dynamic ride-sharing problem is formally defined, and an integer programming model is constructed to solve it. A multi-rider distance and time constraint algorithm uses a distance matrix and sliding time windows to preprocess data before matching is proposed, thereby optimizing data quality and improving computational efficiency. The ride-sharing process is divided into a reservation order matching stage based on path similarity and a real-time order matching stage based on path distance degree. A two-stage collaborative mechanism is designed to guide the collaboration of the two stages. Furthermore, numerical experiments are conducted using two real-world datasets from developing and developed country regions to verify the efficiency and practicability of the proposed approach.