Modeling And Optimization Of Shared Parking Slots Allocation Problem In Residential Communities

In recent years,with the continuous improvement of people’s living standards,the number of motor vehicles in the society has been increasing continuously,and the contradiction between supply and demand of parking slots has become more and more prominent.In the contexts of the great convenient life-style provided by the sharing economy,due to problem of parking difficulties in the city,sharing the sharable private parking slots of the residential communities through the park-sharing allocation platform can alleviate or even solve the problem.Considering the realization form of the sharing economy and the unique characteristics of the shared parking slots,reasonable and effective allocation rules and the key factors which affect the allocation results should be seriously addressed.Therefore,in the view of the parking slots allocation management platform,this thesis considers the time windows between the supply and demand sides of parking slots,the demander satisfaction,as well as the uncertainty of the demander arrival time under the soft time windows,and solve these problems by constructing mathematical models and designing corresponding algorithms.The main research contents are as follows:(1)The relevant researches about the development of the sharing economy and optimization of shared parking slots are summarized at first,based on consulting a large number of domestic and international literature.Then,the design of matching mechanism for parking slots allocation,the berth allocation problem,and the uncertainty of arrival times are reviewed with respect to the related algorithms and the current research states of related fields.(2)From the perspective of slots allocation management platform,the auction theory pricing model is used to determine the trade price of the supply and demand sides.Then,the allocation problem of shared parking slots is described and formulated as an integer programming model which takes the time windows of the suppliers and demanders into account.In the experimental part,a large number of random instances are generated to verify the correctness of the model and test the solution efficiency.Moreover,compared with First Come First Serve(FCFS)allocation method,the contrast experiments show that the method proposed in this thesis has obvious advantages in utilizing of shared parking slots.(3)According to the real situations,we consider the preference satisfaction of the parking slots demanders,introduce the satisfaction evaluation function to describe the changing trend of the demanders’ satisfaction,as well as the time windows and parking slots cost.A multi-objective optimization model is established.An improved Adaptive Genetic algorithm is designed to solve the problem.The correctness of the model and the performance of the algorithm are confirmed by solving a large number of random generated instances.The sensitivities of key parameters in the model is analyzed.(4)For the problem that the arrival time of the parking slots demanders possesses uncertainty,a penalty function of soft time windows is introduced to punish the parking activities that violate a given time window.An integer linear programming model is established for the problem.Furthermore,a Genetic Simulated Annealing algorithm is designed to solve large-scale instances.The correctness of the model is validated.The sensitivity of relevant parameters is analyzed.Finally,the effectiveness of Genetic Simulated Annealing algorithm is confirmed by comparison experiments.