交通运输系统工程与信息 ›› 2022, Vol. 22 ›› Issue (1): 171-181.DOI: 10.16097/j.cnki.1009-6744.2022.01.019

• 系统工程理论与方法 • 上一篇    下一篇

考虑临时和预约需求的共享泊位动态分配方法

刘永红,罗霞*,胡剑鹏   

  1. 西南交通大学,交通运输与物流学院,成都 611756
  • 收稿日期:2021-09-08 修回日期:2021-11-23 接受日期:2021-12-07 出版日期:2022-02-25 发布日期:2022-02-23
  • 作者简介:刘永红(1992- ),男,重庆开州人,博士生。
  • 基金资助:
    四川省科技厅科技计划项目

Dynamic Allocation Method for Shared Berth Considering Immediate and Reservation Demands

LIU Yong-hong, LUO Xia* , HU Jian-peng   

  1. School of Transportation and Logistics, Southwest Jiaotong University, Chengdu 611756, China
  • Received:2021-09-08 Revised:2021-11-23 Accepted:2021-12-07 Online:2022-02-25 Published:2022-02-23
  • Supported by:
    The Science & Technology Project of Sichuan Province(2020YJ0255)

摘要: 为解决多居住区共享停车场同时存在临时共享停车需求和提前预约共享停车需求的动态泊位分配问题,本文提出将未来几个时段的共享停车供需同时考虑到当前时段泊位分配中的单时段泊位分配方法,并将本时段泊位匹配结果延伸到后续时段,从而提出共享泊位动态分配流程。对临时停车需求,以用户成本最小为目标确立优先泊位匹配的原则;对预约停车需求,以系统收益最大为目标确立在当前时段结束时统一分配泊位的原则。定义4个指标评价模型效果。 选择来源于成都市某共享停车APP的3个居住小区一天的数据,对模型进行实证研究以及灵敏度分析。研究结果表明:临时共享停车需求和提前预约共享停车需求所占比例在0.4~0.6时,泊位分配后总目标表现最好;将未来几个时段的供需同时考虑到当前时段中,泊位分配效果更好,且考虑的未来时段越多,泊位分配后总目标越好;在泊位分配时,为了较好地保证临时停车需求能够优先分配到泊位,往往会造成总目标变差。

关键词: 交通工程, 动态泊位分配, 双目标规划, 共享停车, 居住区

Abstract: This paper focuses on the dynamic parking spaces allocation of multi- residential shared parking lots with both immediate and pre-booking shared parking demands. The paper proposes a single period parking spaces allocation method, which includes the supply and demand of shared parking in the future. The results of berth matching in the current period are extended to the next periods, and then the shared parking spaces dynamic allocation process is proposed. The principle of priority berth matching for immediate shared parking demand is established with the goal of minimizing user costs. For pre-booking parking demand, the principle of uniformly allocating berths at the end of the current period is established with the goal of maximizing system revenue. The study defined four indicators to evaluate the effect of the model. Data from one day shared parking APP containing three residential communities in Chengdu were used to conduct empirical study and sensitivity analysis. The results indicate that: (a) when the proportion of immediate shared parking demand and pre-booking shared parking demand is between 0.4 and 0.6, the total objective performance reaches the desired value; (b) the berth allocation performance is better when the supply and demand of the future time intervals are considered at the current time interval. The more future periods are considered, the better of the overall target; (c) it often leads to the deterioration of the overall target in order to ensure the immediate parking demand can be allocated to parking spaces in priority.

Key words: traffic engineering, dynamic berth allocation, bi-objective programming, shared parking, residential areas

中图分类号: