The Application Of Data Mining Technology In The Retail Inventory Control

Because of the increasingly fierce market competition, modern enterprise need to continuously improve their ability to meet the clients'need rapidly if they want to succeed in such a competitive environment. Retail enterprises, in the end of the supply chain, directly to the consumers, are the hub to connect consumers and suppliers. Retail inventory is hard to control because it is impacted both by the changes of consumers and the suppliers which are uncertain factors. Nowadays competition in the retail business is mainly based on the competition of time and customers'demand, and a good inventory control is the key to improve customer satisfaction and to control inventory cost.Data mining is a production of database technology and artificial intelligence technology which can find potential useful information in large nosy data. Retail inventory system stores a lot of history data, however most of which is just used for querying and processing, and it is failed to take advantage of these huge amounts data from which to explore the helpful information for decision-making, so resulting a phenomenon that there is a large amount of data but with no use.This paper mainly studies the application of data mining method in inventory control, for inventory control involves many aspects, the author chose only inventory classification and inventory forecasting to study. At first, author describe the application process of data mining in inventory control, including data collection, data collation, data integration and model selection and result evolution. Then author analyze the various modes of data warehouse and its building process. The most important part of this paper is modeling of the two aspect of inventory control which is inventory classification and inventory forecasting, in this paper, author use MATLAB software as platform, and use BP neural network set up inventory classification model and inventory forecasting model, and also do simulation analysis of which. The model involves three aspects:the determination of sample data, the conversion of sample data and the determination of network structure. Finally the paper design a inventory control system which is based on data mining, and analyses the system demand, including target demand and user needs, and build the system architecture which has four layers:data acquisition, storage layer, presentation layer and application layer, and design the system function module structure, including database management, inventory analysis, supplier management and system security and maintenance of four modules, and achieve the ability of inventory control analysis, mainly the ability of inventory classification and inventory forecasting. At last, author prospect the application of data mining in inventory control.