Research On Scale And Layout Of Shared Bike Facilities For Rail Transit Stations In Residential Historic Urban Area

In the rapid urbanization process of our country,the old city area has gradually evolved into a city center with rich historical and cultural resources and unique characteristics.In the historical residential area,it is convenient to use shared bikes to connect rail transit.However,it is difficult to find a shared bicycle.It is difficult to find or return the bike resulting in unbalanced distribution of time and space.At the same time,the random parking of shared bikes also affects the environment and features of the old city.In response to this problem,this study analyzes the travel characteristics of residents in the historical residential areas,proposes a method for selecting the location of shared bike parking spots,analyzes its reasonable scale,and establishes a dynamic scheduling model for shared bike to connect rail transit.The main research contents include:(1)The willingness to pay in the consumer surplus theory establishes a traveler benefit calculation model based on Logsum differences to quantify traveler's willingness to pay for travel costs corresponding to different travel modes.The travel characteristics of residents are analyzed,and traveler benefits under the five typical working conditions are calculated,that is shared bike search and parking time,road traffic congestion,road tolls,subway fare,and private car parking fee changes.The results show that if the total time for finding and parking shared bicycles is shortened by 20%,the total traveler benefit of shared bicycle connecting rail transit is increased by 15.7%,and the traveler benefit of shared bicycle connection rail transit travel is maximized;due to the shortage of parking resources and traffic management measures in the old city residential areas,private car travel has the lowest total traveler's benefit in five working conditions,which is a less ideal way of travel.(2)Based on the analysis of the travel characteristics of the residents in the old city residential area,combined with the analysis of the historical travel data of shared bicycles,the growth coefficient method is used to establish the prediction model of the shared bicycle travel demand in the old city residential areas,and the model solving algorithm is designed.The prediction and analysis results of the demand for bike sharing and borrowing are obtained.(3)Based on the prediction and analysis results of the shared bicycle trip demand,to meet the goal of the shortest total walking distance from all trip generation points to the shared bike parking locations,a pre-selected model is established.According to the location and scale of the pre-selected parking spots,a two-stage final site selection model is established.First,based on the analysis of the factors affecting the location of shared bicycle parking spots,the weight of the shared bicycle parking evaluation index is calculated by the analytic hierarchy process.Second,based on the indices,a MOORA multi-objective optimization method is establish,the judgment value of the multiobjective optimization method for each pre-selected parking point is calculate,and the ranking determines the layout and scale of shared bike parking locations.(4)In order to solve the problem of uneven spatial and temporal distribution of shared bicycles,the current scheduling methods and existing problems by shared bicycle operators are combed and analyzed.The scheduling of shared bicycles is divided into static scheduling during low peak and dynamic scheduling during peak periods.Scheduling strategies based on customers' rides and scheduling strategies based on employees of shared bike operators are given.Taking the judgment of the dispatching status of the shared bicycle dispatching point and the allocation of dispatched vehicles among the dispatching points as the starting conditions,a dynamic bicycle sharing dispatching model based on employees was established.A software program based on genetic algorithm is designed to solve the dynamic scheduling path and scheduling quantity.(5)Take Tianqiao Residential Area in the old city of Beijing as an example of analysis.First,analyze the historical cycling data of shared bicycles in the area,analyze and compare the peak hours,locations and numbers of working and non-working days.Second,on the basis of the current shared bicycle parking locations,the planned parking locations are divided into 5 categories: maintain,newly-added,cancel,increase or reduce the size,and finally select 79 shared bicycle parking spots,and based on historical data analysis and recent shared bicycle travel demand forecast,determine 6 scheduling periods: 7: 00-8:00,8:00-9:00,9:00-10:00,16:30-17:00,17:00-18:00,18:00-19:00.Among the 79 designated parking locations,12 parking spots that need to be dynamically scheduled are clustered into 6 dispatch areas(including 3 subway stations in the area)through the judgment of dispatch points,and the dynamic dispatch during peak hours is performed.Finally,the genetic algorithm is used to solve the dispatching quantity and dispatching route of each dispatching point of the dynamic dispatching plan during peak hours,and the dispatching plan for reference in the case area is proposed.