Swarm Neural Network Is Applied Research, Supply Chain Inventory Management

The Supply Chain Management is a hotspot of current research, and it is also a management method business enterprise very favorite. The Stock Management is an important part of The Supply Chain, because the valid stock management can bring huge economic value. Along with the popularization of information technique, more and more enterprise fetch in the Management Information System, but nowadays, traditional enterprise Management Information System has already couldn't satisfied the need of enterprise. The application of intelligence technique in the Management Information System becomes an inevitable trend.The research of Artificial Neural Network which is one of computer intelligence technologies has important academic meaning. The Back Propagation Neural Network is one of the most popular neural networks currently. But the learning arithmetic of Back Propagation is base on the essence of grads descending, so there are inevitably problems of it is easy to get into partial least extremum, slowly constringency speed, long training time and so on. The Particle Swarm Optimization arithmetic is an excellent optimization arithmetic that can solve the non-linear, un-fluxionary and multi-peak value optimizing problems. This arithmetic has some characteristics such as easy calculation, quickly constringency speed, stronger ability of looking for the excellent answer in global space, so it can be the learning arithmetic of Back PropagationIn this thesis, improve the arithmetic for its disadvantages by increase the particles' self-regulating ability of the learning factors and the speed's aberrance factor. Use the improved arithmetic as the learning arithmetic of Back Propagation. Through the simulant experiments, verify the constringency speed and the ability of looking for the excellent answer in global space of the arithmetic has been advanced after the improvement. Then based on the stork theory of the Operational Research and the stock management thinking in the Supply Chain Management, create the system with some factors influencing raw material stock for an enterprise producing edible oil according to its characters of raw material's stock and purchase. According the factors, using the improved PSO neural network as the forecast model, based on the Supply Chain Management Information System, make the development of the accessorial decision module of raw material's stock controlling. Passing that carries on an amount of stock estimate, can the in aid of business enterprise maintains a more reasonable raw material stock quantity in promising producing in a row carrying on of foundation, and measure a suggestion for a homologous raw material purchase, this makes textual research at theories creative of in the meantime have certain realistic meaning. Forecasting the amount of stock passing this module can help the enterprise maintains a more reasonable raw material stock quantity under the continue producing, and give a suggestion of raw material purchase. So the researches of thesis own academic creation and certain realistic meaning.