Open AccessTime series modeling by using exponential smoothing technique for river flow discharge forecasting (case study: Cabenge, Walanae, and Cenranae rivers system)
Author(s) -
Melly Lukman,
Benyamin Tanan
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1088/1/012100
Subject(s) - exponential smoothing , autoregressive integrated moving average , smoothing , exponential function , mathematics , autoregressive model , flow (mathematics) , double exponential function , statistics , time series , mathematical analysis , geometry
The exponential smoothing technique is a time series modelling technique that can be used for forecasting. This technique is very important for use in hydrology to determine trends and predict future river flow conditions. In this study, 4 Exponential Smoothing Techniques were analyzed, namely Single Exponential Smoothing, Double Exponential Smoothing, Double Exponential Smoothing (Winters), and ARIMA to predict the flow of the Cabenge river on SWS Walanae - Cenranae. From the results of the study of the four Exponential Smoothing Techniques, Triple Exponential Smoothing (Winters) Of the 4 ES Techniques analyzed in this study, Triple Exponential Smoothing (TES) provides a more accurate forecasting value than Single Exponential Smoothing (SES), Exponential Smoothing Dual Smoothing (DES), and the Autoregressive Difference Moving Average (ARIMA). Triple Exponential Smoothing (TES) has smaller error values for MAPE, MAD, and MSD than SES and DES, and the comparison of forecasting discharge with measured discharge values shows a KN value of 93%. To predict the flow of the Cabenge river in SWS Walanae - Cenranae it is recommended to use the Triple Exponential Smoothing (Winters) Technique.