Research On Double Auction Mechanism And Parking Management Strategy For Shared Parking In Residential Areas

With the gradual development of sharing economy,shared parking is considered as an effective solution to alleviate the parking difficulty problem nowadays.Based on mechanism design theory and focusing on the practical problems of parking space sharing,the parking space allocation mechanism and parking management mechanism of shared parking to provide a certain degree of theoretical basis for the implementation of shared parking were studied in this thesis.First,a shared parking space allocation model based on a double auction mechanism were proposed,based on which the allocation effects of the model were analyzed with respect to the prerequisites of reservation mode and realtime mode,respectively.Subsequently,a shared parking reward and punishment management model based on game theory was proposed to analyze the parking behavior of users when they perform shared parking after parking space allocation,and the management strategy of the platform was studied from the perspective of the platform regarding the different parking behaviors of users.The specific research contents are as follows.(1)Based on the double auction mechanism,the double auction mechanism of "reservation + real-time" was designed,taking into account the price willingness of both the supply and demand sides and the temporary nature of the demand for parking spaces.The platform processes user reservation requests at a preset moment for a future time period,and requests sent after the preset moment are processed in real-time mode.On the basis of considering the rejection rate of users,a car space sharing auction model with the objective of maximizing the benefits of the platform was constructed by setting the fairness constraint of price order,time window constraint,bidding price constraint,walking distance constraint,compatibility constraint and one-to-many constraint,and an immune algorithm was designed to realize the solution of the model.(2)The space allocation model and the shared space allocation model based on double auction mechanism were solved and analyzed by example.The allocation results of three models,reservation auctions with price order,reservation auctions without price order and real-time auctions,were solved and compared separately for different user sizes.The results show that the double auction mechanism of "reservation + real-time" can effectively allocate the supply and demand sides,and the mechanism is better than the single reservation auction and the single real-time auction mechanism,and the allocation results of the mechanism considering the price order condition are better than those without the price order condition in terms of user rejection rate,space utilization,and average price of bidding.(3)Based on game theory in the shared parking reward and punishment management model,the time cost function of user parking,the utility function of platform reward and punishment were built by game analysis of the strategies of platform and user.Firstly,the parking management model under the reward and punishment mechanism was constructed to find out the equilibrium point of the game;secondly,the evolutionary game model of parking management under the reward and punishment mechanism was constructed to analyze the equilibrium point and stability of the game under the static and dynamic reward and punishment conditions.The results show that the equilibrium point of both strategies is stable under dynamic punishment and dynamic reward and punishment conditions.(4)The game theory-based shared parking reward and punishment management model is simulated through an arithmetic example.Under a single reward and penalty mechanism,the platform can adjust the reward and penalty intensity according to the parking space vacancy rate and correct location identification rate to guide users to choose regulated parking.Under the reward and punishment mechanism,it was further verified that the strategies of both parties are convergent and stable when dynamic reward and punishment conditions are used.It was concluded that the probability of users regulating parking is higher under dynamic rewards and penalties,and that increasing penalties will lead to faster stabilization of both strategies.