摘要
针对上海地铁车站客流统计系统中的客流数据预测问题,首先,以城市轨道交通的自动售票系统(AFC系统)作为数据基础;其次,基于随机森林与BP神经网络2种经典预测模型进行客流数据预测;最后,利用MAE、RMSE和R 2等3种不同的评价指标,将不同预测值与真实值进行对比分析。结果表明,随机森林算法的预测结果更优,更适合用于上海地铁车站客流预测,且预测精度均可达到90%以上。
Regarding the passenger flow data prediction issue in the Shanghai Metro station,this paper firstly takes the automatic ticketing system(AFC system)of urban rail transit as the data foundation.Secondly,two classic prediction methods,random forest and BP neural network,are used for passenger flow data prediction.Finally,three different evaluation indicators,MAE/RMSE/R2,are used to compare and analyze the different predicted values with the actual values.The results show that the random forest algorithm is more suitable for the passenger flow prediction of Shanghai Metro station,which can reach above 90%.
作者
孙承洁
SUN Chengjie(Shanghai Shentong Metro Group Co.,Ltd.Rail Transit Training Center,Shanghai 201102,China)
出处
《交通与运输》
2023年第5期36-40,共5页
Traffic & Transportation
