Algorithm research for supply chain demand prediction - Taking fresh agricultural product enterprises as example

Supply chain demand prediction plays a very important role for enterprises to realize sales and markets management target effectively, especially for fresh agricultural product enterprises. A new model for supply chain demand prediction for fresh agricultural product enterprises is presented based on improved BP neural network. First the advantages and disadvantages of BP neural network algorithm are analyzed when it is used in supply chain demand prediction; Second, Legendre wavelets algorithm is used to speed up the convergence and improve the prediction accuracy of original BP neural network algorithm and based on this the paper advances a new supply chain demand prediction model for fresh agricultural product enterprises. Finally, certain fresh agricultural product enterprise is taken for example to verify the validity and feasibility of the model and the experimental results show that the model can improve prediction accuracy and decrease the calculation time and can be used for supply chain demand prediction practically.