Forecasting postage traffic using principal component analysis and artificial neural network

Abstract

Predicting postage traffic and enhancing factors which affect postage traffic are some of the main concerns in postal services. In this paper, we present a combined approach based on principal component analysis (PCA) and artificial neural networks (ANN) for forecasting postage traffic. Due to the large available data set size, PCA is used to reduce the dimension and the required principal components (PCs) are selected based on the given rules. These PCs are then treated as inputs of an ANN to make forecasts for the postage traffic. Real data of Islamic Republic of Iran Postal Company is used to compare the results of the proposed model with that of the traditional regression model. Key words: Forecasting, artificial neural networks, principal component analysis.