马尔可夫决策过程
计算机科学
水准点(测量)
共享单车
动态规划
集合(抽象数据类型)
数学优化
经济短缺
运筹学
随机规划
过程(计算)
马尔可夫过程
模拟
实时计算
运输工程
工程类
算法
操作系统
程序设计语言
地理
政府(语言学)
哲学
统计
语言学
数学
大地测量学
作者
Xue Luo,Li Li,Lei Zhao,Jianfeng Lin
出处
期刊:Transportation Science
[Institute for Operations Research and the Management Sciences]
日期:2022-07-01
卷期号:56 (4): 799-826
被引量:14
标识
DOI:10.1287/trsc.2021.1122
摘要
In bike-sharing systems, the spatiotemporal imbalance of bike flows leads to shortages of bikes at some locations and overages at some others, depending on the time of the day, resulting in user dissatisfaction. Repositioning needs to be performed timely to deal with the spatiotemporal imbalance and to meet user demand in time. In this paper, we study the dynamic intra-cell repositioning of bikes by a single mover in free-floating bike-sharing systems. Considering that users can drop off bikes almost anywhere in free-floating systems, we study the simultaneous reposition of bikes among gathering points and collection of bikes scattered along the paths between gathering points under stochastic demands at both the gathering points and along the paths. We formulate the problem as a Markov decision process (MDP), design a policy function approximation (PFA) algorithm, and apply the optimal computing budget allocation method (OCBA) to search for the optimal policy parameters. We perform a comprehensive numerical study using test instances constructed based on the real data set of a major free-floating bike-sharing company in China, which demonstrates the outperformance of the proposed PFA policy against the benchmark policies and the practical implications on the value of repositioning and the impact of bike scatteredness.
科研通智能强力驱动
Strongly Powered by AbleSci AI