基于K-means的城市轨道交通社区接驳共享单车停靠点规划

研究基于K-means的城市轨道交通社区接驳共享单车停靠点规划问题。首先,在分析现有共享单车停靠存在问题基础上,对其骑行数据挖掘,采用K-means算法聚类后进行二次划分,得到候选停靠点;然后建立在可供选择和容量限制的共享单车停靠点双层规划模型,上层为政府追求出行者广义出行成本最小化、共享单车出行量最大化模型,下层为出行者不同接驳方式和站点选择的交通分配模型;最后采用遗传算法优化求解,通过实例予以验证。结果表明:该方法确定共享单车停靠点与规模,方便共享单车出行,增加其出行量,同时解决了停放混乱问题,提高了接驳服务水平。

Planning of shared Bicycle stop for urban rail transit community connection based on K-means

The problem of planning of shared bicycle stop for urban rail transit community connection based on K-means was studied. Through the analysis of the existing parking problems of shared bicycle, the riding data were mined. After clustering with K-means algorithm, the candidate stops were obtained by the second partition. A bi-level programming model of determining the shared bicycle stops and their capacities was established, the upper level was the generalized travel cost minimization and shared bicycle travel volume maximization model which was supported by the government; the lower level was the traffic allocation model for travelers with different connection modes and stop selection. Genetic algorithm was used to optimize the solution and verified by a practical example. The results show that the proposed method can determine the parking stops and scale of shared bicycles, facilitate the travelling and increase travelling frequency by shared bicycles. Meanwhile, the problem of parking disorder can be addressed with the improvement of the service level.