Electricity

Syariza Abdul Rahman, NorhafizaMohd Noor And Kamal Khalid(2005). Comparison Of Time Series Methods For Electricity Forecasting: A Case Study In Perlis. Icoqsia 2005,6 - 8 December, Penang, Malaysia.

Electricity is one of the main sources of energy and it is an important need in our daily lives. Problems will occur if the energy supply to the customers is not sufficient or exceeds the demand. Thus this journal attempts to forecast the monthly electricity demand in Perlis. Three time series methods namely Box-Jenkins ARIMA, Multiplicative Holt-Winter Exponential Smoothing and Time Series Regression were used by this study. The objectives of this study are to build a model of short-term forecasting of electricity demand using Box-Jenkins, Exponential Smoothing and Regression Method, to produce forecast value, and to compare the forecast error among the method.

The data was obtained from the review of sales report of monthly electricity consumption in Perlis. The data from September 1996 until February 2004 was considered in this study. This study produced six month period of electricity consumption forecasting starting from September 2003 until February 2004 and another data were used to fit the models. The comparison is based on the forecast error for six month period which was evaluated from September 2003 until February 2004.

ARIMA Box-Jenkins method is widely used in forecasting load of electricity. For example, studies by Guerrero and Berumen (1998), Darbellay and Slama (2000) and Ekwue and Short (1990) used non-seasonal ARIMA to forecast electricity demand. This method required four steps to complete the forecast process which are model identification, model estimation, model testing and forecast. Exponential Smoothing is also the most popular technique for short-term forecasting method. Alfares and Nazeeruddin (2002) said that this is one of the classical methods that are used for load forecasting. Mbamalu and El-Hawary (1993) used the multiple...