基于SARIMA模型的铁路站点客流量预测研究

Study of Railway Passenger Volume Forecast Based on SARIMA Model

以我国某铁路站点作为研究对象,选取为期435天的某铁路局全部列车旅客乘车数据进行分析建模.在建立模型时首先将数据分为节假日与非节假日两种类型.对于非节假日数据选用包含周期性的SARIMA模型,对于节假日数据选用波动系数模型,通过两种模型组合对铁路站点客流量进行预测,得到了较好的预测效果.运用方法所得到的短期内铁路客流量变化的准确预测,能够为铁路部门合理安排调度、充分利用人力物力提供参考,有效避免了资源的浪费或因准备不足而造成的车站拥挤混乱.

In this paper,a railway station of China is chosen as the research object.The data of all passengers of a railway administration is used to build models and forecast.Before building the model,the data are divided into two types:non-holidays and holidays.SARIMA model is used to predict non-holidays data,and for holiday’s data,fluctuation coefficient model is used to predict.The accurate forecasting of the change of the railway traffic in the short term can provide a reference for the railway department to arrange the scheduling and make full use of the manpower and material resources,thus effectively avoiding the waste of resources or the chaos caused by the lack of preparation.