Flood Prediction Using Seasonal Autoregressive Integrated Moving Average (SARIMA) Model
Abdulrazak Yahya Saleh1, Roselind Tei2
1Abdulrazak Yahya Saleh, FSKPM Faculty, University Malaysia Sarawak (UNIMAS), Kota Samarahan, 94300 Sarawak, Malaysia. 2Roselind Tei, FSKPM Faculty, University Malaysia Sarawak (UNIMAS), Kota Samarahan, 94300 Sarawak, Malaysia.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 1037-1042 | Volume-8 Issue-8, June 2019 | Retrieval Number: H65560068819/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper aims to evaluate the performance of the Seasonal Autoregressive Integrated Moving Average (SARIMA) model for the purpose of flood forecasting. Seven datasets are provided by the Drainage and Irrigation Department (DID) for Sungai Bedup, Serian, Sarawak, Malaysia; and these loads of valuable information are used to evaluate the performance of the SARIMA algorithm. A distinctive network was trained and tested using the daily data obtained from the DID from the years 2014 to 2017. The performance of the algorithm was evaluated based on the technique of Root Mean Square Error (RMSE) by comparing with the Long Short Term Memory Network (LSTM) and Backpropagation Network (BP). Among the seven datasets, the Sungai Bedup set shows a small testing error rate, which is (0.008), followed by Sungai Meringgu (0.011), Semuja Nonok (0.023), Bukit Matuh and Sungai Busit with the same value (0.025); and lastly the value of Sungai Merang is (0.029). The results prove that the SARIMA model can be employed reliably to forecast the water level of Sungai Bedup with the lowest RMSE value, which is 0.008. Meanwhile, LSTM has a RMSE value of 0.08 and Backpropagation has an RMSE value of 0.711. More discussions will be provided to demonstrate the effectiveness of the model in flood prediction.
Keyword: Artificial Neural Seasonal Autoregressive Integrated Moving Average (SARIMA), Long Short-Term Memory (LSTM), Backpropagation (BP).
Scope of the Article: Regression and Prediction.