GENERATING MULTIPLE TIME SERIES FORECASTS WITH ARTIFICIAL NEURAL NETWORKS IN A TELECOMMUNICATIONS COMPANY

In this work, the generation of multiple period demand forecasts was approached through the use of artificial neural networks (ANNs) for time series. The study was carried out for a telecommunications company and called for the use of a method to fine-tune the ANNs by using design of experiments (DOE) and optimization techniques. The results give evidence of better forecasting performance of the A…