基于改进K-means算法的公共自行车站点区域划分

为了解决公共自行车调度过程中调度路径过长的问题,文章提出了一种改进的K-means聚类算法。该算法通过数据分析估算出k个中心点作为初始中心点,在一次K-means算法聚类划分后,引进调度需求量参数,将边缘站点做二次K-means算法,得到新的区域划分结果。实例分析表明,该算法有良好的全局收敛性,能有效地改善调度路径过长,调度效率低下等问题。

In order to solve the problem that the scheduling path is too long during the process of public bicycle scheduling, this paper proposes an improved K-means clustering algorithm. The K-means algorithm is used to estimate the k-center points as the initial center point. After the K-means algorithm clustering, the scheduling requirements parametears are introduced, and the K- means algorithm is obtained by the edge site to obtain Hie new regional partition result .The example shows that the algorithm has good global convergence, which can effectively improve the problem that scheduling path is too long and the scheduling efficiency is low.