In this study, United States' energy demand in industrial sector
is forecast. ANN is chosen as the forecasting tool. The reasons
behind this choice are the ability of ANN to forecast the future
values of many variables simultaneously and to model the
nonlinear relation in the data structure. To show these reasons, a
comparison between the performance of ANN models and multiple
linear regression (MLR) models is done. The organization of the rest
of the paper is as follows. The second part includes a brief review of
the background of this problem. Third part reviews the previous
studies which contributed to this topic. In part four, background
information of ANN and MLR modeling is briefly summarized. In
part five, data are analyzed, models are compared, and results are
discussed. The final section reviews the conclusions of this paper