为有效获取中小城市公交乘客出行信息,挖掘其出行特征,基于公交乘客IC卡数据、稀疏GPS数据和公交站点位置信息等数据,提出一种公交乘客上下车站点推算方法。首先,通过内插法补齐稀疏GPS数据并与公交站点经纬度匹配获得车辆到站时刻,进而匹配乘客的IC卡刷卡时刻,识别乘客上车站点;然后,根据出行链理论划分不同公交出行链,结合乘客出行假设和公交站点特征简化推算过程,获得不同乘客类型公交出行链的下车站点;最后,选取如东县公交系统中沿线站点特征不同的两条线路对推算方法进行验证。结果表明,该方法可分别修正两条线路11.3%和16.9%的乘客上车站点匹配结果并成功推算出84.2%和87.5%的乘客下车站点,有效提高了时间匹配精度,可完整识别出乘客一天的公交出行信息,对于GPS数据稀疏的中小城市公交乘客上下车站点推算具有普适性。

In order to effectively obtain the travel information of bus passengers in small and medium-sized cities and mine their travel characteristics, a calculation method of bus passengers′ boarding and alighting stops was proposed based on IC card data, sparse GPS data and location information of bus station. Firstly, the sparse GPS data was supplemented by interpolation method and matched with the longitude and latitude of bus stops to obtain the bus arrival time, and then matched with the IC card swiping time to identify passengers′ boarding stops. Secondly, according to the travel chain theory, different bus travel chains were divided. The calculation process was simplified by combining passengers′ travel hypothesis and the characteristics of bus stops, and passengers′ alighting stops of different types of bus travel chains were obtained. Finally, two lines with different station characteristics in Rudong county bus system were selected to verify the method. The results show that the method could correct 11.3% and 16.9% of the matching results of the passengers′ boarding stops of the two lines respectively, and successfully calculate 84.2% and 87.5% of the passengers′ alighting stops. In addition, the method could improve the time matching accuracy effectively and identify passengers′ daily bus travel information completely. The method is universal for the calculation of bus passengers' boarding and alighting stops in small and medium-sized cities with sparse GPS data.