Font Size: a A A

Research And Application Of Data Mining Technology On Quality Management In Photonics Enterprises

Posted on:2010-12-05Degree:MasterType:Thesis
Country:ChinaCandidate:J GuoFull Text:PDF
GTID:2178360275453221Subject:Mechanical and electrical engineering
Abstract/Summary:PDF Full Text Request
With increasing automatization level of photonics enterprises,management information system is widely used in photonics enterprises;they have established ERP database system which covers with the process of production in photonics enterprises.In order to make full use of mass data accumulated during the running of ERP and discover available knowledge containing in the data,and elevate the level of the quality in photonics enterprise,so the research work is based on data of quality which exist in ERP of FoSan X Group Co.,Ltd,the point is to research the data mining system based on the data warehouse of quality,for making full use of the function of the data mining in quality in Photonics enterprise,which can direct the production.The synthetic application principles of data warehouse technology and data mining technology was discussed in this paper on the basis of analyzing the current situation of data mining.Firstly,in this paper,establish data warehouse of quality based on database system of former ERP in photonics enterprise,and then assume to exploit the data mining system based on the data warehouse,and use the system to analyze quality with data mining.This paper has achieved the forecast of quality according to quality with the neural network BP arithmetic to train corresponding model,and the result attains prospective purpose,so the means has partly practical meaning.The exploitation of the data mining system based on the warehouse of quality,has established groundwork for informational course of photonics enterprise and establishment decision-making backup system.The works done by the author in this thesis are as follows:(1) Analyzed the present condition and existent problem of enterprise quality management,put forward scooping out the data warehouse and data mining a technique to lead to go into to the quality management to resolve a counterplan in conduct and actions.(2) This research studied the basic technology of data-warehouse and data-mining deeply,and improved them to suit the photonics enterprise' characters. The new system frame based on data warehouse and data mining is proposed and analyzed.(3) The developing process and method of quality data warehouse was analyzed in this paper and conceptual model,logical model,physical model were also designed, and applied and exhibited the data in data warehouse.(4) In the foundation of learning the neural network technology principle,the BP neural network structure,sample data and the forecast model of the quality forecast determined.The data mining prototype developed using SQL language and C++ Builder 5 in a SQL Server 2000 environment.The prototype system was applied in to the some enterprise actually and obtained preferable effect.These confirmed the feasibility of theory study and implement method.So this method and system can be applied to enterprise widely and effectively.
Keywords/Search Tags:data warehouse, data mining, BP neural network, quality forecast
PDF Full Text Request
Related items