Station Segmentation with an Improved K-Means Algorithm for Hangzhou Public Bicycle System

摘要

In China, Hangzhou is the first city to establishthe Public Bicycle System. Now, the system has been thelargest bike- sharing program in the world. The software ofHangzhou Public Bicycle System was developed by our team.There are many and many technology problems in thedecision of intelligent dispatch. Among of these problems,determining how to segment the stations into severalsections to give different care is very important. In thispaper, an improved k-means algorithm based on optimizedsimulated annealing is used to segment the stations ofHangzhou Public Bicycle System. The optimized simulatedannealing(SA) algorithm is used to assign k-means initialcluster centers. Practice examples and comparison with thetraditional k-means algorithm are made. The results showthat the proposed algorithm is efficient and robust. Theresearch result has been applied in Hangzhou.