摘要
以共享单车订单数据、空间地理数据以及公共交通刷卡数据等多源数据为基础,构建基于地理加权回归的需求影响模型.结果表明:共享单车订单的空间分布具有空间集聚效应;影响因素对共享单车需求的影响程度具有空间非稳态的特征,在大多数交通小区,居住用地、休闲娱乐、购物服务、科教服务、公交登降量、轨道登降量等因素与共享单车需求为正相关关系;办公用地、风景名胜与共享单车需求为负相关关系.本研究可为共享单车的投放规模确定和日常运营调度提供支撑.
In order to study how factors influence the dockless bicycle share demand,using dockless bicycle share trip data,spatial geographic data and smart card data,this study develops the demand impact model based on the geographically weighted regression.The results show that the spatial distribution of dockless bicycle share demand has a spatial agglomeration effect,and the influence degree of the variables has a spatial nonstationary characteristic.In most traffic analysis zones,residence,leisure,shopping and education services and bus ridership,subway ridership factors are positively related to the demand for dockless bicycle share,office and scenic spots services are negatively related to demand.This research can provide support for the determination of the scale of the sharing of bicycles and the daily operation and scheduling.
作者
林鹏飞
翁剑成
胡松
梁泉
尹宝才
LIN Pengfei;WENG Jiancheng;HU Song;LIANG Quan;YIN Baocai(Key Lab of Traffic Engineering,Beijing University of Technology,Beijing 100124,China)
出处
《交通工程》
2020年第2期65-72,共8页
Journal of Transportation Engineering
基金
国家自然科学基金国际合作项目资助项目(No.61420106005)
北京市“科技新星”计划项目(No.Z171100001117100).
关键词
共享单车
需求分布
建成环境
地理加权回归
交通小区
dockless bicycle sharing
demand distribution
built environment
geographically weighted regression
traffic analysis zone
