期刊
TRANSPORTATION RESEARCH PART B-METHODOLOGICAL
卷 87, 期 -, 页码 44-63出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.trb.2016.02.006
关键词
Parking; Stochastic Poisson game; Equilibrium; Distributed algorithm
类别
资金
- NSF [CMMI 1436786]
This paper proposes a decentralized and coordinated online parking mechanism (DCPM), which seeks to reduce parking congestion at multiple parking facilities in a central business district (CBD) through guiding the parking decisions of a parking coordination group. To establish this DCPM, this study develops a stochastic Poisson game to model the competitions among parking vehicles en route at multiple parking facilities. The equilibrium condition for the proposed stochastic Poisson game is formulated through involving travelers' parking choice behavior described by multinomial logit model. Furthermore, we prove that the stochastic Poisson game is a potential game with a unique equilibrium. A simultaneously updating distributed algorithm is developed to search the equilibrium solution of the DCPM. Its convergence is proved by both mathematical analysis and numerical experiments. The numerical experiments are conducted to test the efficiency of the DCPM, based on a real-world CBD covering Guicheng Community, Nanhai District at Foshan in China. The performance of the DCPM is compared to three greedy strategies following the nearest first, cheapest first, and least cruise first policies, respectively. The experimental results demonstrate that the DCPM significantly reduces cruise vehicles and average cruise distance per vehicle from all other three greedy strategies; the least cruise first strategy, which takes advantage of the real-time open spots information at parking facilities, performs better than the nearest first and the cheapest first strategies without the access to real-time information. The DCPM can further improve the benefit of the real-time information. Additionally, in terms of walking distance and parking cost, the DCPM provide a trade-off solution between the nearest first and the cheapest first strategies. (C) 2016 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据