Research On Application Of Data Mining For SCM

Data Mining which researches how to discovery valuable knowledge and rules from large scale database, is one of the active sub area of Artificial Intelligence with a broad application in many business and scientific area. Supplies chain management is a novel management method which contains the contents of management science, information science and system engineering. The aim of SCM is to obtain the least cost under the circumstance of most profits, and the biggest obstacle is the re-distribution of profits. This thesis focused on how to mining some useful information which may be a few help for the managers out the tons of enterprises' data.First, the background of supplies chain management is introduced, including the basic concept, the basic idea, key technique and the development trend of SCM, followed by the basic idea about data mining and knowledge discovery, as well as the basic method and technique of data mining. In chapter 2 we introduce the basic concept and methods for financial statement analysis, explicate ratio formulae and the meanings of ratios which are mentioned in the financial ratio analysis.Because finance statement contains the information about the refund ability, profit and loss and the development potency, finance statement analysis is an important way of managers to get decision information and other valuable knowledge.In chapter 4, we research the technique of data mining, which can be applied into the domain of SCM and financial statement analysis. The main method include: DBSCAN algorithm to mining the cluster information in some enterprises, the main elements analysis to reduce the number of financial ratios, statistics on the basis of analysis to time-series databases, analysis of relations between financial ratio. These methods improve the ability of system.This thesis also presents a novel method for managers to select the most possible and the most stable enterprise partner. And then, after giving the basic concepts of rough set theory, we put forward that how to reduce the attribution of information data according to the algorithms in rough set theory. Finally a system is designed to fulfill basic function of SCM, which mainly introduces abstraction design and detailed design.