Amirkabir University of Technology Amirkabir Journal of Civil Engineering 2588-297X 43 1 2011 08 23 Analyzing and Forecasting Railway Passenger Demand with Time Series Algorithm Analyzing and Forecasting Railway Passenger Demand with Time Series Algorithm 9 17 116 10.22060/ceej.2011.116 FA Shahriar Afandi Zade Amir Masoud Rahimi Ali Asghar Talaei Ehram Safari Journal Article 2014 03 16 Regarding the importance of the supply management for existing transportation facilities and allocating these resources in the rail transportation, travel demand forecasting has a very important role. In this paper the time series models are used to forecast passenger demand in Iranian railway network. <br />Before estimation, model selection and forecasting, the stationary and non-stationary time series models of railway passenger demand are analyzed with the tests of unit root and seasonal unit root. In the modeling part the Box-Jenkins method are used that the main reason for using them was the strong correlation between the data in several months and seasons and repeating exact trends in the fixed basis of time. The Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) criteria are used in order to evaluate the performance of models. The final fitted models are in conformity with family of seasonally ARIMA, and have at least 92 percent accuracy in the forecasting. Regarding the importance of the supply management for existing transportation facilities and allocating these resources in the rail transportation, travel demand forecasting has a very important role. In this paper the time series models are used to forecast passenger demand in Iranian railway network. <br />Before estimation, model selection and forecasting, the stationary and non-stationary time series models of railway passenger demand are analyzed with the tests of unit root and seasonal unit root. In the modeling part the Box-Jenkins method are used that the main reason for using them was the strong correlation between the data in several months and seasons and repeating exact trends in the fixed basis of time. The Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) criteria are used in order to evaluate the performance of models. The final fitted models are in conformity with family of seasonally ARIMA, and have at least 92 percent accuracy in the forecasting. https://ceej.aut.ac.ir/article_116_c45147dee729311ef5b5c3003946c48f.pdf