摘要
出租车作为城市公共交通的重要组成部分,在人们日常出行中发挥着重要的作用。不同司机在服务策略上具有差异性,这些差异性将会导致司机收入的不同。本文基于出租车GPS数据对出租车司机收入问题进行分析,在数据清洗和地图匹配的基础上,提取出租车的OD和单笔订单收入,制定合理的规则将出租车司机分为三类:高、中、低收入司机,并采用多元有序Logistic回归模型研究出租车司机收入与空载寻客距离、载客距离和载客速度之间的关系。最后采用DBSCAN聚类算法挖掘高收入司机在各个高峰时间段的载客热点区域,提升司机寻客效率。
As an important part of urban public transportation,taxis play an important role in people's daily travelDifferent drivers have differences in service strategies,and these differences will lead to differences in driver incomeBased on the taxi GPS data,the income problem of taxi drivers is analyzed,and on the basis of data cleaning and map matching,the OD and single order income of taxis are extracted,and reasonable rules are formulated to divide taxi drivers into three categories:high,medium and low income driversA multivariate ordinal Logistic regression model was used to study the relationship between taxi driver income and empty passenger seeking distance,passenger carrying distance and passenger loading speedThe DBSCAN clustering algorithm is used to mine the passenger hotspot areas of high-income drivers in various peak hours to improve the efficiency of driver search.
作者
齐宝德
刘涵
Qi Baode;Liu Han(Gansu Provincial Transportation Science and Technology Communication Center,Lanzhou 730000,China;School of Transportation,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《青海交通科技》
2022年第6期57-63,91,共8页
Qinghai Transportation Science and Technology
