Research On Key Technology Of Urban Public Bicycle Scheduling

With the continuous progress of urbanization,the problems of traffic congestion and environmental deterioration become worse day by day,seriously restricting the sustainable development of cities.Under the background of people’s increasing health consciousness and needs for low-carbon life,public bicycle,a green and healthy way to get around has been promoted and become an essential part of urban public transportation system.Although the public bicycle system is widely adopted nowadays,the dispatching of public bicycles has just started to develop.This paper discusses around the two key technologies of public bicycle dispatching — dispatching area division and dispatch path planning,and the discussion is followed by corresponding methods for improvements.For dispatching area dividing,based on intensive research on dispatching and the quantification method of dispatching in the real scenario of public bicycle,this paper proposes a community discovery algorithm that is based on multi-objective optimization to realize the dividing of dispatching area for public bicycles.This algorithm accords with the law of renting and returning of public bicycles.At the meantime,it keeps the dispatching workload between regions as balanced as possible.As to the problem of scheduling path planning,based on the combination of genetic algorithm,simulated annealing algorithm,and adaptive adjustment strategy,this paper proposed an improved simulated annealing genetic algorithm,which makes expanding the search space and enhance the global searching ability of genetic algorithm.Simulation experiments of dispatching regional planning and path planning were separately carried out on actual operational data.In the dispatching simulation experiment,the effectiveness of the community discovery algorithm based on multi-objective optimization is verified by comparing with the results of community discovery algorithm and clustering algorithm.The result of the division is consistent with the law of lease,and it can satisfy the balance of regional workload as much as possible.In the simulation experiment of the scheduling area partition,the improved simulated annealing genetic algorithm can search better solution stably.