Forecasting problems entail the construction of a model, and using the information available, estimating the parameters of the model to optimize the prediction performance. This requires the application of complex mathematical functions to represent highly nonlinear problems. In this paper, the accumulative back propagation artificial neural network has been used to solve forecasting problems. As a case study, neural network and regression technique have been used to make a forecast for the future electric loads. A comparison with regression technique has been made. The neural network architecture proved to give better results especially for nonlinear loads