一种考虑乘客等待耐心不确定性的网约车实时派单方法及系统
By extracting data features and constructing parameters for real-time ride-hailing dispatch methods, and combining global optimal matching distribution planning and cyclic probabilistic matching decision-making, the problems of short-sightedness, slow response, and strong data dependence in ride-hailing dispatch systems are solved. This achieves efficient resource allocation and passenger patience management, and improves the robustness and profitability of the system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING JIAOTONG UNIV
- Filing Date
- 2025-12-26
- Publication Date
- 2026-07-17
AI Technical Summary
Existing ride-hailing dispatching technologies suffer from suboptimal greedy matching strategies, slow response times with batch matching strategies, and strong data dependence with reinforcement learning or stochastic optimization strategies. Furthermore, they ignore the uncertainty of passenger patience, leading to misallocation of transportation resources and passenger anxiety while waiting.
By employing data feature extraction and parameter construction, and performing spatiotemporal gridded classification based on historical operational data, a resource allocation model is constructed. Combining global optimal matching distribution planning and cyclic probabilistic matching decision-making, patience period verification and supply and demand adjustment factors are introduced to achieve real-time order response and resource reservation.
It improves the system's robustness and global benefits, takes into account real-time performance, adapts to the randomness of passenger patience, and reduces order loss rate and short-sightedness issues when capacity is scarce.
Smart Images
Figure CN121998301B_ABST