Deep dispatching: A deep reinforcement learning approach for vehicle dispatching on online ride-hailing platform. (May 2022)

• Record Type: Journal Article
• Authors: Liu, Yang; Wu, Fanyou; Lyu, Cheng; Li, Shen; Ye, Jieping; Qu, Xiaobo
• Is Part Of: Transportation research. Volume 161(2022)
• Publication Date: 2022-05
• Subjects: Vehicle dispatching -- Deep reinforcement learning -- Load balancing; Logistics -- Periodicals; Transportation -- Periodicals; 388.011
• DOI: 10.1016/j.tre.2022.102694
• Languages: English
• ISSNs: 1366-5545

Abstract

Highlights: Abstracting the vehicle dispatching problem as a load balancing problem. Solving the challenge of concurrent requests with the help of recommendation system. Designing the DRL method based on the data of a real ride-hailing platform. Abstract: The vehicle dispatching system is one of the most critical problems in online ride-hailing platforms, which requires adapting the operation and management strategy to the dynamics of demand and supply. In this paper, we propose a single-agent deep reinforcement learning approach for the vehicle dispatching problem called deep dispatching, by reallocating vacant vehicles to regions with a large demand gap in advance. The simulator and the vehicle dispatching algorithm are designed based on industrial-scale real-world data and the workflow of online ride-hailing platforms, ensuring the practical value of our approach. Besides, the vehicle dispatching problem is translated in analogy with the load balancing problem in computer networks. Inspired by the recommendation system, the problem of high concurrency of dispatching requests is addressed by sorting the actions as a recommendation list, whereby matching action with requests. Experiments demonstrate that the proposed approach is superior to existing benchmarks. It is also worth noting that the proposed approach won first place in the vehicle dispatching task of KDD Cup 2020.