next up previous
Next: Preprocessing the input Up: A NEURAL NETWORK APPROACH Previous: Introduction

Sale forecast by neural networks

For our project we use the sale information of 53 articles of the same product group in a supermarket. The information about the number of sold articles and the sales revenues in DM are given weekly starting September 1994. In addition there are advertising campaigns for articles often combined with temporary price reduction. Such a campaign lasts about two weeks and has a significant influence on the demand on this article. Sale, advertising and price for two different articles are shown in figures 1 and 2.

Figure 1:  article 362900 (without advertising)

Figure 2:  article 372138 (with advertising)

Figure 3:  Feedforward multilayer perceptron for time series prediction

We use feedforward multilayer perceptron networks with one hidden layer together with the back-propagation training method [2], [3]. For prediction the past information of n recent weeks is given to the input layer. The only result in the output layer is the sale of one article for the next week. So there is a window of n weeks in the past and one week in the future (see figure 3).

Mon Jun 12 14:12:53 MET DST 1995