Track: Decision Sciences

Abstract

Being able to predict the future demand of electricity constitutes part of the issues utilities companies, policy makers and private investors willing to invest in developing countries are facing. The use of efficient, reliable electricity demand predictor would improve significantly the infrastructure planning and the expansion of power transmission. In this paper, the national demand for electricity at a macro level, based on data relating to macro level economic and demographic indicators was predicted using Artificial Neural Network (ANN). Forecasted values for five electricity sectors namely agricultural, transport, mining, domestic and commerce/manufacturing sectors were obtained using ANN. Four growth scenarios have been considered for the forecasting namely low, moderate, high (less energy intensive) and high (same sectors) scenarios. These inputs values for the period of 2014 to 2050, from the Council for Scientific and Industrial Research (CSIR), were used to test data and validate the use of this new approach for the prediction of electricity demand. The deviations between the predicted values using ANN and the recommended values by CSIR were well within an acceptable range. This study demonstrates that the use of ANN would improve significantly the decision making within the energy sector.