| As fixed assets management is a very important part of Enterprise Resource Management, how to make use of existing database of enterprises'fixed assets management and increase the utilization of fixed assets is an important task of the financial departments. Making full use of enterprise property, striving to raise economic benefit and making exertive biggest effect of business enterprise property are the purport of this thesis. There is a wide variety of capital assets with different functions in China's enterprises, and renewing speeds and functions are also dramatically different. There is no pertinence in the existing managerial system, therefore how to properly categorize the fixed assets of enterprises, adopt different management methods, bring into play the maximal utilization of fixed assets and lower the impropriation of unused funds, and realize the optimum allocation of resources become more and more important.My thesis chooses more than 10,000 data from the fixed assets database of a state-owned enterprise (SOE) as its object and classifies them into four kinds according to the uses and functions of resources. Meanwhile, corresponding management should be applied to them thereby the whole process of purchase, use and disuse of the fixed assets could become more scientific and reasonable. Furthermore, the waste of resources caused by ill management, unused resources and repeated purchase could be avoided, providing effectively assistant decision-making knowledge for further improvement of fixed assets management.Firstly, this thesis introduces the background, purpose and meaning of my research. The reasonable allocation of enterprises'fixed assets and efficient management are important aspects which strengthen competition ability of enterprises. Beginning with the theory related with data mining and clustering analysis, and combining the characteristics of management of the enterprises'fixed assets, this thesis tries to use clustering analysis in fixed assets management and achieve the categorization with the use of statistical analysis software.Secondly, this thesis will elaborate the related theories of clustering analysis, including domestic and overseas research status of data mining technology, the technical foundation and the basic process of data mining, and common arithmetic in clustering analysis. It also carries on improvement to the k-means arithmetic according to the demand of this paper.Thirdly, this thesis discusses the possibility which applies clustering analysis technique to enterprises'fixed assets management. According to a state-owned enterprises'fixed assets management database and on many foundations with analytical experiments assurance, the study chooses effective index signs. According to the improved k-means arithmetic, the thesis adopts statistical analysis software SPSS to carry out substantial evidence analysis. In order to verify the possibility of conclusion, the thesis adopts another layered clustering method. The results of the two analyses are basically identical, which proves the possibility of the method.At last, this paper discusses targeted management to fixed assets with different characteristics. Analyzing respectively the four kinds of fixed assets caused by clustering and adopting effective measures provide powerful support for the practical management of enterprises. The problem which exists currently is the information of fixed assets database is still not enough, and the factors affecting the maintenance and function of fixed assets cannot be found. However, with increasing perfection of database, the application of clustering analysis in the fixed assets will be more and more extensive and it will also provide a more and more accurate result .All in all, the research is based on the clustering analysis algorithm and combines qualitative analysis with quantitative analysis. On a certain scale it provides theoretical basis to make the most of fixed assets management and has certain value in practice. |
PDF Full Text Request