Study On The Influence Of Weather And Built Environment On The Demand For Shared Bicycles

Under the development concept of green ecological city and low-carbon city construction that is actively promoting,the shared bicycle economy,as an emerging green and environmental protection shared economy,has gradually penetrated into ordinary Citizens live in daily life and act as an important member of daily travel tools,playing an indispensable role in leisure entertainment and commuting work.However,it is precisely because shared bicycles are non-motorized public transportation with the nature of short-distance self-service travel,which are directly affected by external objective environmental factors such as local climate and built environment far more than other factors.Therefore,it is necessary to explore the changes in the travel demand of shared bicycles from the perspective of weather and built environment,so as to improve people’s travel experience,so as to better realize the positive impact of shared bicycles on urban environmental protection and urban public transportation.Therefore,this study is based on the Mobike shared bicycle order data in the central urban area of Shanghai,combined with GIS spatial analysis method,basic statistical analysis method,and model analysis method,to deeply explore the influence of weather and built environment on the demand for shared bicycles.The main work contents and research conclusions are as follows:（1）Analysis of the basic characteristics of shared bicycle riding.This paper determines the distribution law of the riding time and riding distance of shared bicycles,as well as the morning and evening peak distribution periods of bicycle travel.The study found that the travel time by bicycle is usually less than 15 minutes,and the travel distance is usually less than 1.5kilometers.It is the main tool for residents to travel for short distances;there are obvious morning and evening travel peaks on weekdays,which appear at 7:00-8:00 and 17:00-18:00respectively,the rest days are relatively flat,but the average daily riding volume on working days and rest days is basically the same.Using the kernel density tool and multi-distance spatial clustering analysis to identify the spatial distribution characteristics of shared bicycles,the study found that bicycles are distributed differently in hotspot areas such as working days and rest days,among which commercial centers,employment centers and subway stations such areas are hotspots where bicycles gather.On the whole,cycling activities are mostly clustered in space.（2）Analysis of the impact of weather on shared bicycles.Firstly,the thermal map of the correlation coefficient of each weather influencing factor was drawn,and it was found that there was a high positive correlation between the AQI index and the concentrations of PM2.5,PM10,SO₂,NO₂,CO,and O₃.Then through preliminary statistical analysis,it is found that the temperature,atmospheric pressure,relative humidity and wind speed are too high or too low,which will inhibit the travel demand of shared bicycles.Travel demand during overcast,moderate rain and thunderstorms is lower than that of clear and cloudy.Finally,the multiple linear regression model is used to comprehensive analysis of the influence direction and degree of weather elements on the demand for shared bicycles,and it is found that moderate rain,showers,thunderstorms and air quality indicators have a significant impact on travel demand.On this basis,the shared bicycle trip data with good rainfall and air quality are selected,and the k-means clustering method is used to analyze the affected change types and spatial distribution in each region.The research shows that:The use demand and return demand of shared bicycles under rainfall conditions are most in the areas of“significantly reduced morning and evening peak hours,and no significant changes in idle time”;When the air quality is good,there are the largest number of areas in the category of“no significant change in each time period”for use demand and“significant increase in morning and evening peaks and no significant change in idle time”for return demand.（3）Based on the overall research scope of the city,a multiple linear regression model（MLR）was constructed to explore the interactive effects of weather and built environment on the demand for shared bicycles,and the weather was divided into two categories:rainfall and good air quality.The results show that the number of built environments such as science and education and cultural services,commercial residences,medical care services,and accommodation services have a positive correlation with the changes in the demand for shared bicycles during rainfall and air quality,while the number of bus stops only affects the shared bicycles during rainfall.There is a significant positive correlation with changes in usage demand.It shows that the factors of the built environment are one of the main reasons for the differences in travel demand changes between regions under the same weather conditions.（4）Based on the geospatial heterogeneity of local areas,the geographic weighted regression model（GWR）was used to deeply explore the specific impact of various built-up environmental factors on the demand for shared bicycles and the demand for return in different regions when rainfall and air quality were good.The visualization shows and compares the spatial differences of the built environment impact coefficients in different models,it is found that the influence directions and degrees of various built environment elements in different grid areas are different,further illustrating that changes in travel demand are not only disturbed by the weather,but also significantly affected by the built environment attributes in the region.At the same time,the fitting accuracy of geographically weighted regression model has been greatly improved.