Dynamic allocation of regional shared parking spaces based on assignment problem model

To address the issue of inefficient parking space utilization resulting from temporal and spatial disparities in parking demand, shared parking emerges as an innovative traffic management paradigm. The core concept of shared parking lies in the strategic allocation of parking spaces. By analyzing the usage patterns of parking areas, the relationship between supply and demand over time can be represented in a binomial form. An integer programming model for parking allocation is formulated, where the objective at each allocation stage is to minimize walking distance and parking costs while maximizing the degree of temporal alignment with overall parking demand in the region. The weights of these three indices are calculated using the entropy weight method, and subsequently, the parking cost matrix is derived through linear weighting. It is proposed to utilize the Hungarian Algorithm as a method for solving the assignment problem to obtain the allocation scheme that minimizes total parking costs. A Python program is developed to execute the phased optimal allocation of parking spaces. At the conclusion of the operation, the results for overall parking demand satisfaction, parking space utilization, and parking costs are generated. The model is applied to parking space management in both real-world instances and simulation experiments. The results demonstrate that the model not only ensures a higher utilization rate of parking spaces but also effectively reduces overall parking costs and the idle rate of parking time compared to the current parking methods. This model is well-suited for the parking allocation processes of smart parking platforms.