Energy efficiency assessment using artificial neural network combined with data envelopment analysis

A management technique that uses energy information as a basis to eliminate waste, reduce and control current level of energy use is important to every sector. This paper uses artificial neural network (ANN) and data envelopment analysis (DEA) to examine the effectiveness of energy management. First, the artificial neural network using economic activity, the gross domestic product (GDP), was used to calculate the predicted energy consumption in the industrial sector of South Africa between 1993 and 2000. Data envelopment analysis was then employed to calculate the overall energy efficiency, using the predicted energy consumption as output and the observed energy consumption as input. The results showed that most of the years during this period efficient energy management were successful in achieving energy saving.