Research On Automatic Distribution System Of Delicatessen Chain Industry Based On Data Mining

Recent years we have experienced several major food safety accidents,the government and the national attach more importance on food safety than ever before.As one member of food FMCG industry,delicatessen chain industry face greater risks of food safety because of the short product shelf life.And the food safety supervision is quite strict.At the same time as the promotion of informatization,the revolution of information technology has also set off in the traditional fresh and delicatessen industry,and constantly promote the delicatessen industry for meticulous management.How to ensure that only three days to complete the shelf-life of product,while also reducing the inventory of the chain stores,so accurate distribution,the chain will be the urgent need to solve the problem of delicatessen enterprises.Through the mining of historical sales data,establish data model,and then use the model to computer system automatic calculation of sales forecast,combined with the existing inventory,generate the corresponding order data,to achieve the purpose of automatic distribution,which will may solve these problems faced by large delicatessen chain enterprises.In order to solve this program,specific studies are as follows:(1)Based on the analysis of historical sales data and business model,the multivariate data analysis model is determined.(2)Using multiple regression and time series data mining method,carries on the analysis to the model of correlation,finally determined the expression of multiple regression model,reliability of the data model is verified through testing and practical.(3)Combined with the current distribution system and data mining algorithm,the existing business processes and systems to do the corresponding optimization,so that it can meet the requirements of automatic distribution system.(4)Design and implement automatic distribution system.Combined with the new business process and system flow,design system.Using Java batch processing technology,database script task,to transform the existing distribution system,implement the system.The testing of the automatic distribution system,it found that the distribution data of data mining algorithm with automatic distribution system requirements.In the practical application of the automatic distribution system,not only to maximize the sales demand,but also can effectively reduce the store inventory.