交通运输系统工程与信息 ›› 2015, Vol. 15 ›› Issue (5): 202-207.

• 系统工程理论与方法 • 上一篇    下一篇

考虑节假日效应的交通枢纽客流量预测模型

成诚1,杜豫川*1,刘新2   

  1. 1. 同济大学道路与交通工程教育部重点实验室, 上海201804; 2. 青岛海信网络科技股份有限公司, 山东青岛266071
  • 收稿日期:2015-05-07 修回日期:2015-07-20 出版日期:2015-10-25 发布日期:2015-10-28
  • 作者简介:成诚(1989-),男,广西桂林人,博士生.
  • 基金资助:

    工信部电子发展基金项目(201406).

A Passenger Volume Prediction Model of Transportation Hub Considering Holiday Effects

CHENG Cheng1,DU Yu-chuan1,LIU Xin2   

  1. 1. Key Laboratory of Road and Traffic Engineering of the Ministry of Education, Tongji University, Shanghai 201804, China; 2. Qingdao Hisense Network Polytron Technologies Inc, Qingdao 266071, Shandong, China
  • Received:2015-05-07 Revised:2015-07-20 Online:2015-10-25 Published:2015-10-28

摘要:

客流量预测是城市交通枢纽管理的基础,准确的客流量估计为交通枢纽的运力调整,管理预案的设计提供基础.目前对客流量预测的研究较多,但现有模型并未考虑节假日效应对枢纽客流量的影响.因此,本文基于多元季节性时间序列(SARIMAX) 原理,建立考虑节假日效应的城市交通枢纽客流量预测模型,并以上海虹桥2 号航站楼站轨道交通客流量数据为基础,对该模型进行了标定和预测.标定结果显示,在春节期间,该站点客流量将有明显的下降,而在其他法定节假日期间流量均有一定程度的提升.对模型预测值和真实值比对结果显示,该模型的平均误差在5%以内,表明该模型具有较强的实用性.

关键词: 城市交通, 客流量预测, 多元季节性时间序列模型, 节假日效应, 交通枢纽

Abstract:

Passenger prediction model is one of the fundamental process in transportation hub management. The precise estimation of passenger volume provides instructions for transit scheduling and transportation hub management solution planning. At present, some of the studies are proposed in forecasting passenger and traffic volume. However, most of these studies fail to consider the holiday effect on passenger volume variability. Therefore, a passenger volume prediction model of transportation hub is proposed which take the holiday effects into consideration based on the seasonal ARIMA model that considers explanatory variables (SARIMAX) method. The metro passenger volume of Shanghai Hongqiao International Airport Terminal 2 Station is used for calibration and prediction. The calibration results indicate that during spring festival, the passenger volume witnesses a relative decrease while increases would occur in other legal holiday periods. The mean absolute percent error of the prediction results is less than 5% . The accuracy suggested its advantage in passenger volume evaluation and on site application.

Key words: urban traffic, passenger volume prediction, SARIMAX model, holiday effects, transportation hub

中图分类号: