Study On Tidal Spatio-Temporal Characteristic And Dispatching Method Of Bike Sharing In Morning Rush Hour

In recent years,as a new type of shared travel mode,bike sharing has brought new vitality to urban transportation with their convenient and fast features.While it also alleviates the pressure on the urban transportation system and meeting the needs of residents for shortdistance travel.However,due to the uneven distribution of bike sharing resources,it is easy to cause "difficulties in borrowing bikes" and "difficulties in returning bikes" in some areas.These supply and demand issues severely restrict the healthy development of bike sharing.Based on the study of the tidal space-time characteristic of bike sharing on Xiamen Island,this study identifies and analyzes hotspots and proposes optimization and deployment plans.The research conclusions can provide decision-making basis for the government and bike sharing operation enterprises,and help to better optimize the allocation of bike sharing resources,with certain theoretical and practical significance.The work is specifically carried out in the following areas:(1)Analysis of the spatio-temporal characteristic of bike sharing in morning rush hour.This study uses data extraction,data integration and reconstruction,data cleaning and other data preprocessing methods to process the required data.Through quantitative and visual analysis,the riding behavior characteristic and spatio-temporal characteristic of bike sharing in morning rush hour during the morning peak period are obtained.It is found that Xiamen’s bike sharing during the morning peak period have the characteristic of short riding time,short distance and fast speed,and have the distribution rules of "internal density,external sparseness" and dense distribution of rail transit stations.(2)Identification of riding tide hotspot areas and analysis of influencing factors.The adaptive DBSCAN clustering algorithm is used to identify the tidal hotspot areas and obtain the tidal hotspot migration pattern of bike-share locking and unlocking by time of day.The migration rules of bike sharing switch tide hotspots are obtained,and the 50 electronic fences with the most serious tidal situation are precisely located by calculating the density of bike sharing and net inflow and outflow of electronic fences.At the same time,a least squares regression model is constructed to explore the impact of the built-up environment on the distribution of bike sharing tidal hotspots.The results show that the built environment,such as residence,office and transportation facilities,has a greater correlation with bike sharing hotspot areas.(3)Developing a regional dispatching model based on morning rush hour demand hotspots.Optimizing bike sharing dispatching through enterprise optimization and user parking guidance strategies based on tidal electronic fencing.In the enterprise optimization scheduling,the Ant System and Max-Min Ant System are used to solve the optimization objective of the enterprise’s minimum scheduling cost and user satisfaction.The comparison of experimental results shows that using the Max-Min Ant System to solve the problem is better.At the same time,user guidance parking strategy is used for assistance,and the KD-Tree algorithm is used to search for alternative parking spots,considering factors such as parking convenience to actively guide users to nearby parking spots,and alleviate parking pressure.