Improvement of an online ride-hailing system based on empirical GPS data

Chen，Jinyu1; Li，Wenjing1; Yu，Qing2,3; Shibasaki，Ryosuke1; Zhang，Haoran4

2023

Deep learning Gibbs sampling Greedy algorithm Optimization Ride-hailing system Simulation

英语

Ride-hailing, as a popular shared-transportation method, has been operated in many areas all over the world. Researchers conducted various researches based on global cases. They argued on whether car-hailing is an effective travel mode for emission reduction and drew different conclusions. The detailed emission performance of ride-hailing systems depends on the cases. Therefore, there is an urgent demand to reduce the overall picking-up distance during the dispatch. Moreover, most of the cases only analyze the emission pattern of mature ride-hailing systems. None of them provide a changing pattern of a developing one. Discovering the emission pattern during the development can help understand how the number of users in the system affects the emission performance and furtherly provide a guideline for controlling the number of users to keep the system at a high-performance level. In this study, we answer these two demands by proposing two frameworks. 1. A cross-simulation model combined with Gibbs sampling for a comprehensive computation. Based on the simulation results, we found a strong impact of user scale on emission performance. The mean void distance proportion varies from 3.69% to 31.75% under all situation simulations. Finally, based on this relationship, we provided guidance for the computation of approximate user scale if the emission and efficiency performance of car-hailing is expected to be better than a threshold. 2. An optimization method combined with a prediction model to minimize the global pick-up distance when solving the dispatch problem. We used Didi ride-hailing data for 1day for simulation and found that our method can reduce the picking-up distance by 8.60% compared with the baseline greedy algorithm. The proposed algorithm additionally makes the average waiting time of passengers more than 10min shorter. The statistical results also show that the performance of our method is stable. Almost the metric in all cases can be kept in a low interval. We believe our findings can improve deeper insight into the mechanism of the ride-hailing system and contribute to further studies.

10.1016/B978-0-323-95892-9.00002-4

Handbook of Mobility Data Mining: Volume 3: Mobility Data-Driven Applications

3

23-61

其他

2-s2.0-85152831250

Scopus

1.Center for Spatial Information Science,The University of Tokyo,Kashiwa-shi,Chiba,Japan
2.Department of Computer Science and Engineering,Southern University of Science and Technology,Shenzhen,China
3.The Key Laboratory of Road and Traffic Engineering,Ministry of Education,Tongji University,College of Transportation Engineering,Shanghai,China
4.School of Urban Planning and Design,Peking University,Shenzhen,China

Chen，Jinyu,Li，Wenjing,Yu，Qing,et al. Improvement of an online ride-hailing system based on empirical GPS data. 2023-01-01.