Forecasting the Monthly Electricity Demand of Georgia using Competitive Models and Advices for the Strategic Planning

Abstract

Electricity is a kind of energy that consumption increases rapidly. However, Electricity is used from the residences to the industry everywhere. This non-storable energy needs to be predicted precisely and supplied sufficiently. For this reason in this research electricity demand of Georgia is predicted using comparative models as Box Jenkins and Neural Network. By this way, an accurate model was developed to perform a strategic plan. For the calculation of significance level, MAPE is utilized to see error. It was seen that artificial neural network multilayer perceptron model provided better predictions rather than SARIMA model. Finally, some suggestions are expressed to the government of Georgia for future plans.