In our project we use the sale information of 53 articles of a certain product group in a supermarket. The information about the number of sold articles and the sales revenues in DM (German currency unit) are given weekly starting September 1994. In addition there are advertising campaigns for articles often combined with temporary price reductions. Such a campaign lasts about two weeks and has a significant influence on the demand on this article. Sale, average price and advertising campaigns for a specific article are shown in figure 2.
Figure 1: feedforward MLP for time series prediction
We use feedforward multilayer perceptron (MLP) networks with one hidden layer together with the back-propagation training method [3]. In order to predict the future sale the past information of n recent weeks is given in the input layer. The only result in the output layer is the sale for the next week. So there is a window of n weeks in the past and one in the future (see figure 1).
Figure 2: sale and prediction for an article with advertising