Research And Application Onthe Big Data Mining Of Product Information For Mass Customization

With the rapid development of information technology and manufacturing technology and the fierce competition in the market,the development direction of manufacturing industry is being transformed from the traditional mass production mode to the mass customization modewhich takes the customer as the center.Manufacturing industry is more and more dependent on information technology to remain competitive.In the meantime,a mass of product information are generated in the various stages of product life cycle,such as:product requirement information,product design information,manufacturing data information,supply chain data and operational data information and externalindustry information,etc.In the future,the big data environment of product information will be the underlying data support to build the intelligent plant and the basis for the implementation of mass customization.When the product data of MC reaches a certain scale,data mining can be used to perceived product information in a new perspective,gaining valuable knowledge and experience.By analyzing the data mining demand of MC,this article researches the acquisition,storage,classification of customer requirements and the process of creating product configuration repository,then applies data mining technology to customer requirements process and product configuration design.The research content of this paper mainly includes the following aspects:(1)By Expounding the trends and issues of MC,this paper did the feasibility analysis of applying Hadoop and data mining technology to MC.Analyzing some related issues of customer requirements process and product configuration design which appear in MC,this thesis designed the architecture of data mining system based on Hadoopand elaborated its design ideas and each module.(2)Data mining ofcustomer requirementsinformation.Structured model of general customer requirement information was built in order to design the acquisition and storage solutions of customer requirement.MapReduce-based Bayesian classification algorithm,which was used to train classifier by historical customer requirement information,was designed.The classifier normalized and classified the customer information and then marked the processing level of each customer requirement,which was the preparation forproductconfiguration tasks.(3)Data mining of product configuration information.Product configuration unit model and product configuration template model was designed to build product configuration database,which could storage information of product configuration instances.MapReduce-based C4.5 decision tree algorithm was designed to mine potential configuration rules from a large number of product configuration instancesin product configuration database,whichwas eventually used to build product configuration rules database.(4)According to above theory,data mining system of product information for MC was developed.In view of the business requirements of MC,the workflow of Hadoop-based Big Data mining system was elaborated and running condition of data mining algorithms was shown.Data mining of product information for MC by constructing the big data mining system is a useful attemptin theory and practicalapplication,which is the basis for intelligent product configuration.