基于多源数据的网约车与地铁接驳特征分析

Analysis of Connection Characteristics between Online Car-hailing and Metro Based on Multi-source Data

网约车出行作为一种新兴的出行模式,在人们日常出行中发挥着重要作用。基于网约车订单数据,引入手机信令、地图导航等多源数据,首先,分析网约车与地铁接驳的时空特征,并与手机用户的地铁接驳出行数据进行对比;其次,基于XGBoost(extreme gradient boosting,极端梯度增强算法)构造充分考虑接驳距离、地铁乘距、接驳需求、公交便利程度、通勤出行占比等因素的回归模型;最后,从到达地铁站与离开地铁站两个方面,分析各因素对网约车与地铁接驳的影响。结果表明:使用网约车到达地铁站的订单量远大于离开地铁站的订单量,网约车与地铁接驳距离在1.5~3.5 km的接驳量最大;网约车与地铁接驳存在明显的早高峰和晚高峰时段,高峰时段使用网约车与地铁接驳的出行占比与手机用户使用地铁出行的占比一致。

As an emerging travel mode,online car-hailing plays an important role in people’s daily travel.The research is based on the online car-hailing order data,and multi-source data such as mobile phone signaling and map navigation.Firstly,the temporal and spatial characteristics of online car-hailing and metro connection are analyzed and compared with the metro connection travel of mobile phone users.Secondly,a regression model is constructed based on XGBoost,which fully considers the factors such as connection distance,ride distance,connection demand,bus convenience and commuter travel proportion.Finally,from the two aspects of arriving and leaving the metro station,the impact of various factors on the connection between online car-hailing and metro is analyzed.The results show that the number of people who use online car-hailing to arrive at the metro stop is much higher than that who use online car-hailing to leave the metro station.The connection between online car-hailing and metro is the highest in the range of 1.5~3.5 km.There is an obvious morning peak and evening peak in the connection between online car-hailing and metro station,and the proportion of trips using online car-hailing and metro connection during peak hours is consistent with the proportion of mobile phone users using metro transportation.