Abstract: The rapid development of information and communication technologies has not only brought about great changes in social production and lifestyle, but also boosted the emergence of a new type of activity space and built environment. The activity space of residents based on a bike-sharing system is a case in point. Using Mobike data, this paper conducts a spatiotemporal analysis of a bike-sharing system and identifies the factors influencing the travel OD  (origin-destination) density of Guangzhou old town. The research shows that residential and bus-station density affects the travel OD density both on weekdays and weekends. Subway stations' density only affects travel OD density on weekdays and recreation density mainly affects travel OD density on weekends, while working density has a negative effect on weekend evening peak travel OD density. In order to guide and monitor the relevant enterprises in launching and deploying bicycles, the study shows that relevant government departments and enterprises need to jointly build and share the dynamic information about bicycles so as to perform real-time monitoring and provide a service platform.

Key words: old town, bike-sharing, travel behaviors , Influencing factors

摘要： 移动信息通讯技术的快速发展在推动社会生产和生活方式发生重大变革的同时，也促进新的活动空间类型不断浮现。基于共享单车出行的居民活动空间即为代表。使用摩拜单车骑行数据，分析广州老城区共享单车出行OD 点的时空间分布特征，重点探究高峰期OD 点密度的影响因素。研究发现，共享单车工作日的出行具有明显的早晚高峰特征，在高峰时段集中分布在地铁站点和主干道周边；休息日的出行高峰时段为午间和傍晚，主要分布在各游憩场所。居住、公交站密度均对工作日和休息日出行高峰OD 点密度具备影响，地铁站、游憩密度则分别仅对工作日、休息日高峰OD 点密度具备影响。研究认为政府相关部门和企业需要共建共享单车动态实时监测与服务平台，以实现对相关企业投放、调配单车工作的指引和监控。

关键词: 老城区, 共享单车, 出行, 时空特征, 影响因素