International Journal of Engineering and Information Systems (IJEAIS)   Year: 2022 | Volume: 6 | Issue: 9 | Page No.: 27-38 Forecating Water Demand with ARFIMA models: A case study of Dhofar Governorate, Oman Download PDF Oladipupo Bello Abstract: Water demand models are indispensable tools in the planning, design and operations of water supply systems. These models are developed based on historical water consumption data which usually have trends and seasonality. The autoregressive fractionally integrated moving average (ARFIMA) process is used in this study to evaluate the performance of univariate water demand models. Several ARFIMA models were fitted to monthly water consumption data of Dhofar Governorate, Oman from January 2002 to October 2020. The best fitted ARFIMA model was selected based on Akaike Information Criterion, and compared with autoregressive integrated moving average (ARIMA) model. The simulation study showed that ARFIMA model performed better than the ARIMA model. The water demand forecast values for two (2) years was obtained for the Dhofar Governorate using the best fitted ARFIMA model. The decision makers in the water supply industry will find the results of the study useful to evaluate investment projects to expand their facilities and determine optimum water tariff for their customers.