The application of BP neural network optimized by genetic algorithm in logistics forecasts
Huilin Yuan1, 2, Jia Fu2, Wei Hong3, Jinbo Cao4, Jing Li5
1Beijing University of Aeronautics and Astronautics Mechanical Engineering Post-doctoral Mobile Station, Beijing, China
2Northeastern University at Qinhuangdao, Hebei, China
3China petroleum & chemical corp., Hebei oil products co., Hebei, China
4Yanshan University, Qinhuangdao, Hebei, China
5TBEA Shenyang Transformer Group Co., high-voltage switch Institute, Shenyang, Liaoning, China
This paper points out disadvantages of traditional forecast methods and elaborates the advantages of the method based on BP neural network. On this basis, the paper puts forward a logistics forecasting model of BP neural network optimized by genetic algorithm. The new method uses historical data to establish and train BP neural network and thus obtain logistics forecasting model. The results implemented by MATLAB show that, neural network possesses memorizing and learning capability, and can forecast logistics development trend perfectly, which is proved by a large amount of actual forecast results. Compared with BP neural network model, the model has the advantages of less number of iterations, convergence speed and strong generalization ability.