Reinforcement Learning for Digital Twin Empowered Ride-Sharing System Optimization
IEEE Network · 2025-03 · 39 (2) · 184-193
DOI: 10.1109/mnet.2024.3499953 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Network, 2025-03. Verfasst von Kai Jiang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 15 Literaturangaben.
Abstract (Originalsprache)
Recently, as the popularity of ride-sharing services continues to rise, a crucial technical requirement is to optimize operational efficiency for ride-sharing systems. Digital Twin (DT) and Reinforcement Learning (RL) have emerged as promising solutions for this requirement. By creating virtual replicas of environments, DT enables RL to learn from data and facilitate more informed decisions that benefit all ride-sharing stakeholders. However, the integration of DT and RL is still in its infancy due to its dependence on processing power, memory resources, and data quality. Thus, this article provides a comprehensive overview of this novel field from a system-level perspective. First, we introduce the technical basis of RL, DT, and ride-sharing architecture. Then, we explore the integration of RL and DT for ride-sharing system optimization. The logic framework under this integration is elaborated holistically. Subsequently, a specific embodiment is elaborated under the integration. Especially, simulations in embodiment show that the proposed method achieves 13.6% improvement at maximum and 7.3% on average compared with other baselines, when the capacity constraint is 6. Finally, we highlight potential challenges, which may facilitate the transformation of this topic from theory to practice.
Zitieren
APA
Jiang, K., Cao, Y., Wang, Z., Zhou, H., Zhu, H., Liu, Z., & Xu, L. (2025). Reinforcement Learning for Digital Twin Empowered Ride-Sharing System Optimization. IEEE Network, 39(2), 184-193. https://doi.org/10.1109/mnet.2024.3499953
Vancouver
Jiang K, Cao Y, Wang Z, Zhou H, Zhu H, Liu Z, et al. Reinforcement Learning for Digital Twin Empowered Ride-Sharing System Optimization. IEEE Network. 2025;39(2):184-193. doi:10.1109/mnet.2024.3499953
BibTeX
@article{Jiang2025,
author = {Jiang, Kai and Cao, Yue and Wang, Zhenning and Zhou, Huan and Zhu, Hong and Liu, Zhi and Xu, Lexi},
title = {Reinforcement Learning for Digital Twin Empowered Ride-Sharing System Optimization},
journal = {IEEE Network},
year = {2025},
volume = {39},
number = {2},
pages = {184--193},
doi = {10.1109/mnet.2024.3499953}
}
EUVIMED