Research On Evaluation Model Of Supply Chain Information Sharing Based On BP Neural Network

With the development of enterprise information technology,the market environment in which enterprises depend for survival has undergone major changes.It has become very difficult to gain a place in the fierce market by relying on its own strength.Enterprises need to make up their own resources by establishing long-term cooperative relationships.Insufficient,thus achieving effective management of all aspects of the supply chain.In this process,information sharing between enterprises is extremely important,and reasonable information sharing can make information flow smooth.The so-called supply chain information sharing means that upstream and downstream enterprises in the supply chain coordinate and integrate the entire supply chain by sharing information between them to optimize operational efficiency.This paper finds out that the BP neural network method can deal with nonlinear problems and uncertainties very well.The advantage of BP neural network method is that it can realize arbitrary nonlinear mapping,which can make up for the deficiencies of analytic hierarchy process and regression analysis to some extent.It can get the most through repeated learning and generalization of a large number of training sample data.Excellent model.However,the BP neural network method also has certain limitations.At present,the choice of BP neural network structure is still a difficult problem.Based on the relevant literature,this study summarizes the research status of supply chain information sharing evaluation and artificial neural network,analyzes the characteristics and deficiencies of existing research,and builds evaluation indicators of supply chain information sharing.system.At the same time,in view of the problems existing in the current supply chain information sharing evaluation method,the BP neural network method is innovatively used to construct the model of the index system,and the MATLAB 2018 a is used for network design and calculation,and the sample is continuously Training and testing are performed so that the error of the model reaches a predetermined range.In order to verify the accuracy of the model,different iterative schemes are used for analysis,and the results of different models are compared.The model is simulated and analyzed to obtain the optimal model.The research shows that the BP neural network evaluation method is very feasible for the research of supply chain information sharing evaluation.It has opened up a new method for supply chain information sharing evaluation and enriched the existing supply chain information sharing evaluation research.system.