New product forecasting demand by using neural networks and similar product analysis

Abstract

This research presents a new product forecasting methodology that combines the forecast of analogous products. The quantitative part of the method uses an artificial neural network to calculate the forecast of each analogous product. These individual forecasts are combined using a qualitative approach based on a factor that measures the similarity between the analogous products and the new product. A case study of two major multinational companies in the food sector is presented to illustrate the methodology. Results from this study showed more accurate forecasts using the proposed approach in 86 percent of the cases analyzed.