Forecasting Model of Amount of Water in Huai Saneng Reservoir of Surin Province

ตัวแบบการพยากรณ์ปริมาณน้ำในอ่างเก็บน้ำห้วยเสนง ของจังหวัดสุรินทร์

Abstract

     The aims of this research were: 1) to forecast the amount of water in Huai Saneng Reservoir of Surin province using statistical forecasting methods including a regression analysis method with an additive seasonal model, a decomposition method with a multiplication pattern, the Box-Jenkins method with a multiplicative seasonal model, and the Winter method with a multiplicative model; 2) to compare the errors derived from four statistical forecasting methods through root mean square error (RMSE) and mean absolute percent error (MAPE). The data used in the analysis was the monthly information of the amount of water in Huai Saneng Reservoir of Surin province between January and December in the last 15 years during 2005-2019. The results revealed that the Box-Jenkins’s forecasting model is SARIMA(1,0,0)(0,1,1)₁₂. The RMSE and the MAPE are equal to 2.159 and 15.539, respectively. The study shows that the Box-Jenkins’s forecasting model has the least error. It might be said that the forecasting model using Box-Jenkins method with a multiplicative seasonal model is the best model to forecast the amount of water followed by the forecasting model using Winter method with a multiplicative model, decomposition method with a multiplication pattern, and regression analysis method with an additive seasonal model, respectively.