Annual energy demand estimation of Iran industrial sector by Fuzzy regression and ARIMA

This research presents a fuzzy regression model to efficiently estimate long term energy consumption in industry sector of Iran from 1982 to 2006. Four independent variables such as energy price, energy intensity, gross domestic production, and employment are introduced as model inputs. The presented model better estimates energy consumption than the conventional technique, auto regressive integrated moving average (ARIMA), based on the preprocessed provided data in terms of mean absolute percentage error (MAPE), root mean square error (RMSE), and mean absolute error (MAE). In addition, the applicability and efficiency of the provided Fuzzy regression model is tested through analysis of variance (ANOVA) and proved its remarkable performance.