基于时间标签的公交运营时段划分方法  

Division Method of Bus Operation Time Period Based on Time Label

作　　者：张瑾[1] 向一擘 ZHANG Jin;XIANG Yibo(School of Transportation and Engineering,Kunming University of Science and Technology,Kunming 650000,China)

机构地区：[1]昆明理工大学交通工程学院,云南昆明650000

出　　处：《物流科技》2020年第5期93-96,共4页Logistics Sci-Tech

摘　　要：公交车辆在不同时段其运营特征会产生较大变化,也影响着公交运营调度水平的提高,因此,研究公交运营特征需要有效的公交运营时段划分方法。文章参考K-means聚类,提出以公交车辆站间运行时间为单位的公交运营时段划分方法,并根据公交运行数据分布的特征,以数据的时间标签为基础从初始簇中心生成、簇中心与元素距离计算及元素的归属上制定了优化运算规则。最后以昆明市5路公交的案例数据验证了文章的方法可行,并进一步讨论了本方法生成簇中心与边界点的分布特征以及运算复杂度,旨在为GPS数据的应用提供新的思路。The operating characteristics of bus vehicles at different time periods will change greatly,which will also affect the improvement of the level of bus operation dispatching.Therefore,studying the characteristics of bus operation requires an effective method of dividing bus operation hours.This paper refers to the K-means clustering,and proposes a method of dividing the bus operation period based on the running time between bus stations.Based on the characteristics of the bus operation data distribution and time label,optimized operation rules have been formulated for the initial cluster center,cluster center and element distance calculation and the elements classify.Finally,the case data of 5 buses in Kunming city are used to verify the feasibility of this method.The distribution characteristics and computational complexity of cluster centers and boundary points generated by this method are further discussed.The aim is to provide new ideas for the application of GPS data.

关 键 词：城市交通 公交运营时段 K-means聚类分析 时间标签 

分 类 号：F570[经济管理—产业经济]