The Design And Implementation Of Purchasing Module Based On SVM

Today economy is increasingly globalized and information technology is evolving,the traditional purchasing model,that is,according to the inventory to determine the purchasing needs and work out purchasing plans can not meet the needs of enterprises in the purchasing.Purchasing department needs to work out appropriate purchasing plans based on knowing the needs of other departments in the enterprise,and according to the production plan and sales.Reasonable choice of suppliers not only in the purchasing process for enterprises to reduce their procurement costs,get a higher price,but also bring a series of other benefits.The work of this paper includes the following points: According to the process of enterprise purchasing business,the paper analyzes the specific requirements of purchasing module,divides the purchasing module into five functional modules: purchasing,quality,receiving,purchase order maintenance and supplier data,analyzes the functions and dataflow in each module,and clear the data exchange relationship between every modules of the entire system.On the basis of the demand analysis,the function module of the purchasing module is designed and realized,and the key functions in the function module are introduced in detail,including database design,detailed functional design and implementation.In the completion of the basic functions of the purchasing module,to add the supplier recommending function.Because the number of suppliers is large,the evaluation of all suppliers is not significant,so the support vector machine model is used to classify the suppliers,and then evaluate and sort the suppliers that are classified as optional suppliers.This paper focuses on the design and implementation of the recommended functions of the supplier,and gives the pseudo code implementation of Sequential Minimal Optimization algorithm,Logistic regression model and Vlsekriterijumska Optimizacija I Komprois-no Resenie method.