Optimal reservation control strategies in shared parking systems considering two types of customers. (July 2023)

• Record Type: Journal Article
• Authors: Zhang, Lifeng; Mu, Yinping; Chao, Xiangrui; Jing, Fuying
• Is Part Of: Computers & operations research. Volume 155(2023)
• Publication Date: 2023-07
• Subjects: Shared parking -- Reservation control strategy -- Dynamic programming -- Decomposition model; Operations research -- Periodicals; Electronic digital computers -- Periodicals; 004.05
• DOI: 10.1016/j.cor.2023.106235
• Languages: English
• ISSNs: 0305-0548

Abstract

Highlights: DP models are constructed to solve reservation control problems with two types of customers. The supermodularity and concavity of two decomposition models are constructed. A general framework to balance the efficiency and accuracy for similar problems is proposed. Effective reservation control strategies are proposed to deal with the heterogeneity of shared time. Abstract: The operational challenges of shared parking platforms include heterogeneous sharing time intervals and two types of customers randomly providing demand information. Thus, the operation of shared parking platforms necessitates a more intricate reservation strategy than those of the hotels and leasing industries. To address the problem, this study proposes reservation control strategies in a shared parking system with two types of customers and then allocates capacity among customers. Firstly, we propose a dynamic programming ( DP ) model with dynamic characteristics of demand information and prove that the boundary condition of the model is NP-hard. Secondly, period- and product-based decomposition models have been proposed to approach the DP model. The objective functions of the period- and product-based decomposition models are concave and supermodular, respectively. We also propose three approximation algorithms to obtain reservation strategies based on decomposition models. Finally, several numerical experiments are conducted to verify the effectiveness of the proposed models and algorithms. Extended experiments are conducted to test the robustness of the method for real-world applications. The results support reservation control in shared parking systems.