Research On Reservation Control Strategy Of Shared Parking Platform

Parking problem becomes more and more serious with the increase of the motor vehicles.Shared parking platform emerged recently with the support of the intelligent technology.The shared parking platform can reuse the idle resources and provide temporary supplies.Shared parking spaces alleviates the shortage of the parking spaces.It uses existing spaces intended for parking cars when the owner is not using it,so the randomness of the parking time and the capacity of shared parking spaces has led to significant challenges for reservation control.In order to improve the utilization rate of parking spaces and alleviate parking difficulties,it is necessary to protect the benefits of parking sharing platform which promotes the sustainable development of the platform.Therefore,this research studies the reservation control policy of the platform with the dynamic nature of the supply and the demand information consideration.The purpose of the reservation control is to maximize the profit of the platform.The main contents are as follows:(1)Research on reservation control strategy of shared parking platform with contract supply consideration.Firstly,in view of the platform faces the reservation customer,this thesis constructed a model and analyzed the nature of the substitute mechanism.Based on the substitute mechanism,the period-based and product-based decomposition algorithms have been proposed to solve the reservation control problem.The result of the research is the algorithms are effective.The sensitivity of the reservation periods and reservation prices to the service rate of the platform and the utilization rate of the parking spaces are studied under the period-based decomposition algorithm.Secondly,the reservation control model is constructed for the two customers: reservation customers and walk-in customers.In view of the high dimension of the state variables,the research has proposed the period-based and product-based decomposition model,and has proved the existence of the optimal strategy.Based on the properties of the decomposition model,three algorithms are constructed.Finally,we conducted numerical simulation to test the effectiveness.The applicability of the three algorithms under different parameters are given.(2)Research on the reservation control strategy of shared parking platform with random supply consideration.Considering that some suppliers may provide the supply information to the platform at any time,the reservation control policy with the uncertain supply and demand consideration has been researched.Because of the randomness of the supply,the product-based decomposition method failed.For dimension reduction of the state variables,firstly,we used opportunity constraints to change random constraints into certain constraints,and then constructed the asymptotically boundary conditions to make the product-based decomposition method valid.Secondly,this thesis researched the properties of the decomposition models and the existence of the optimal solution.Finally,the effectiveness of the algorithms under different parameters has been verified,and the impact of randomness of the supply on reservation control strategy have been analyzed by numerical simulation.The results show that the reservation control strategy based on product-based decomposition is applicable to the situation where the arrival rate of the supply is large,while the reservation control strategy based on the period-based decomposition is applicable to the situation where the arrival rate of the supply is small.(3)Research on the reservation control strategy of shared parking platform with the stay-over requests consideration.The reservation customers may be unable to leave on time due to emergencies.First of all,we have proposed a reservation control model with stay-over request consideration.Then,designed the reservation control strategies in the two cases where the platform can reject the stay-over demand or not.Finally,testified the effectiveness of the two strategies under different parameters.Furthermore,the sensitivity of the reservation prices,reservation periods and the stay-over probability to opportunity cost,the service rate of the parking space and the utilization rate of the parking spaces is analyzed.The research has pointed out the relationship between the probability of the stay-over request and the reservation control policies.When the platform can decide whether the stay-over request is accepted or not,the increase of the stay-over probability will increase the opportunity cost and the willingness of accepting a reservation.When the platform can only passively accept stay-over demand,opportunity cost will increase with the increase of the stay-over probability,and the probability of accepting a reservation will decrease with the increase of the stay-over probability.(4)Research on reservation control strategy of shared parking platform with the heterogeneous of the parking spaces consideration.Because shared parking spaces are provided by different suppliers,parking spaces are more different.To satisfy customers with different shared parking spaces,customers have different experiences.So,the shared parking spaces are heterogenetic.In view of the heterogeneity of parking spaces,this thesis constructed a random dynamic programming model of cross-regional allocation,and given the necessary conditions for cross-regional allocation.A single period allocation algorithm is designed to solve the problem.Finally,the effectiveness of the approximation algorithm and the sensitivity of related parameters have been verified by numerical simulation.The researches show that the heterogeneity of the shared parking space has significant influence on the allocation mechanism.Cross-regional allocation will effectively improve the profit of the sharing platform and the utilization rate of parking space.In summary,the reservation control problem extends from deterministic domain to stochastic domain because of the random of the shared parking supply.Then,the reservation control problems with stay-over behavior and the heterogeneity of parking space consideration is studied.The models constructed in this thesis have generality and can be extended to the other sharing platforms,such as household appliances sharing platform,car sharing platform,house sharing platform,and so on.The algorithms constructed in this thesis is effective and practical.The platform can use the proposed algorithms for reservation control.Therefore,the researches on the reservation control of sharing platform have a very important significance for ensuring the profit of the sharing platform and promoting the healthy and sustainable development of the sharing economy.