Sales forecasting based on ERP system through BP neural networks

Abstract

For the characteristics of manufacturing industry, with the application of business process analysis and optimization methods, the enterprise resources are optimized and a ERP system is formed. For the deficiencies of ERP in decision-making, BP neural network is used to mine the rules. To carry out the nonlinear prediction, the BP neural network defines the inherent relation between the inputs and outputs and the solution to the involved problem by adjusting the weight of the neural network. By extracting the data of the ERP and constructing the neural network prediction model, sales management functions of decision supporting module are to expanded so the sales data of ERP is transformed into the information of decision. In the end, the model is validated by the application. © 2010 Springer-Verlag Berlin Heidelberg.