数据仓库技术在电信经营决策中的应用
摘要
数据仓库不同于传统数据库。传统数据库在联机事物处理(OLTP)中获得了较大的成功，但对于管理人员的决策分析却无法很好地满足，而数据仓库可以很好地做到这一点。目前，国内关于电信企业经营决策支持的数据仓库系统还没有十分成熟的产品，更没有一个完全的标准。
近年来，电信行业信息化进程得到快速发展，运营网络系统、综合业务系统、计费系统、办公自动化等计算机应用系统的相继使用，积累了大量的历史数据。但在很多情况下，这些海量数据在原有的作业系统中是无法提炼并升华为有用的管理决策信息的。一方面，联机作业系统因为需要保留足够的详细数据以备查询而变得笨重不堪，系统资源的投资跟不上业务扩展的需求；另一方面，管理者和决策者只能根据固定的、定时的报表系统获得有限的经营与业务信息，无法适应激烈的市场竞争。数据仓库成为其进一步信息化的必然需求。
本文对数据仓库的发展和应用、数据组织及关键技术进行了分析，并结合SAS数据仓库系统特点，给出了数据仓库开发的一般方法，然后结合为驻马店市移动通讯公司开发的“电信经营决策系统”这一实例，具体讨论了系统的需求分析、设计与实现。特别地，本文对电信行业的需求做出了较为详尽的分析。
“电信经营决策系统”系统的实现，解决了驻马店移动通讯公司因缺乏数据仓库系统而产生的经营决策数据支持问题，可以比较方便地完成对多维数据模型的转轴、上卷、下钻、切片及切块操作，进行多种分析与预测，还具有许多符合电信行业特色的新功能，包括：直接利用切片数据进行基本的统计运算，如时间序列、品牌分析等，并将结果直接反映在界面上或生成位图输出；将OLAP结果直接生成文本、HTML等格式文件，方便决策领导者书写报告。
There is very difference between the Traditional DB and Data Warehouse. The Traditional DB succeeds in OLTP but can' t satisfy the managers' decision-making and analysis. By contrast, Data Warehouse can do it quite well. At present, domestic telecom industry has no perfect system of Data Warehouse, let alone its entire standard.
In recent year, information technology has been applied diffusely in telecom industry. Lots of history data has been accumulated according to the use of computer application systems, such as Operating Network System, Integrated Service System, Statistic Fee System, OA System, etc. In traditional conditions, however, large amount of data can' t be turned into useful information for decision-making. OLTP system becomes inflexible because of the lack of detailed data and systematic resource can' t meet the need of business. On the other hand, managers and decision-makers can' t adapt to violent competition because they only get limit information according to the OLTP system. Therefore, Data Warehouse meets the trend.
First of all, this article analyzed development and application of Data Warehouse, organization of data, and pivotal techniques, moreover, offered the general construction method of Data Warehouse application system in relation to SAS Data Warehouse. At last, it supplied an example of application to Data Warehouse that named A System of Decision Sustainer and Market Analyzer, and discussed the requirements, design, and implement of the system. Especially, it analyzed detailedly the needs of telecom industry, too.
The System of Decision Sustainer and Market Analyzer constructed a fine Data Warehouse application. It can accomplish Rotate, Slice, Drill up/down operation in Data Cube, and all kinds of analysis or predict. At the same time, the system also has some new functions, including to do some basic statistics by using
Slice-Data straightly, to create the results to txt or HTML document expediently which is convenient for managers' reports, and so on.
近年来，电信行业信息化进程得到快速发展，运营网络系统、综合业务系统、计费系统、办公自动化等计算机应用系统的相继使用，积累了大量的历史数据。但在很多情况下，这些海量数据在原有的作业系统中是无法提炼并升华为有用的管理决策信息的。一方面，联机作业系统因为需要保留足够的详细数据以备查询而变得笨重不堪，系统资源的投资跟不上业务扩展的需求；另一方面，管理者和决策者只能根据固定的、定时的报表系统获得有限的经营与业务信息，无法适应激烈的市场竞争。数据仓库成为其进一步信息化的必然需求。
本文对数据仓库的发展和应用、数据组织及关键技术进行了分析，并结合SAS数据仓库系统特点，给出了数据仓库开发的一般方法，然后结合为驻马店市移动通讯公司开发的“电信经营决策系统”这一实例，具体讨论了系统的需求分析、设计与实现。特别地，本文对电信行业的需求做出了较为详尽的分析。
“电信经营决策系统”系统的实现，解决了驻马店移动通讯公司因缺乏数据仓库系统而产生的经营决策数据支持问题，可以比较方便地完成对多维数据模型的转轴、上卷、下钻、切片及切块操作，进行多种分析与预测，还具有许多符合电信行业特色的新功能，包括：直接利用切片数据进行基本的统计运算，如时间序列、品牌分析等，并将结果直接反映在界面上或生成位图输出；将OLAP结果直接生成文本、HTML等格式文件，方便决策领导者书写报告。
There is very difference between the Traditional DB and Data Warehouse. The Traditional DB succeeds in OLTP but can' t satisfy the managers' decision-making and analysis. By contrast, Data Warehouse can do it quite well. At present, domestic telecom industry has no perfect system of Data Warehouse, let alone its entire standard.
In recent year, information technology has been applied diffusely in telecom industry. Lots of history data has been accumulated according to the use of computer application systems, such as Operating Network System, Integrated Service System, Statistic Fee System, OA System, etc. In traditional conditions, however, large amount of data can' t be turned into useful information for decision-making. OLTP system becomes inflexible because of the lack of detailed data and systematic resource can' t meet the need of business. On the other hand, managers and decision-makers can' t adapt to violent competition because they only get limit information according to the OLTP system. Therefore, Data Warehouse meets the trend.
First of all, this article analyzed development and application of Data Warehouse, organization of data, and pivotal techniques, moreover, offered the general construction method of Data Warehouse application system in relation to SAS Data Warehouse. At last, it supplied an example of application to Data Warehouse that named A System of Decision Sustainer and Market Analyzer, and discussed the requirements, design, and implement of the system. Especially, it analyzed detailedly the needs of telecom industry, too.
The System of Decision Sustainer and Market Analyzer constructed a fine Data Warehouse application. It can accomplish Rotate, Slice, Drill up/down operation in Data Cube, and all kinds of analysis or predict. At the same time, the system also has some new functions, including to do some basic statistics by using
Slice-Data straightly, to create the results to txt or HTML document expediently which is convenient for managers' reports, and so on.
引文
1．周君毅等．数据仓库和自动挖掘在CRM中的实现．计算机科学，第29卷8月增刊(NDBC 2002)
2．戴超凡．数据仓库中元数据技术研究．计算机工程与应用，2001-14
3．石磊．OLAP与数据挖掘一体化模型的研究与发展．计算机科学 2000，Vol．27，No．5：45～49
4. Inmon W H. BuildingtheDataWarehouse. Second Edition. New York:John Wiley&Sons, 1996.
5. Jose Samos, Felix Saltor, Jaume Sistac and Agusti Bardes. Database architecture for data warehousing: an evolutionary approach. In DEXA 1998: 746-756.
6. Chaudhuri S, Dayal U. An Overview of Data Warehousing and OLAP Technology. SigmodR, 26(1):65～74
7. Lou Agosta. The Essential Guide to Data Warehousing. Prentice-Hall, 2000
8. William A Giovinazzo. Object-Oriented Data Warehouse Design. 2000
9. A. Berson, J. Smith. Data Warehousing, Data Mining and OLAP. McGraw-Hill, 1997
10．陈京民．数据仓库与数据挖掘技术．北京：电子工业出版社，2002
11．彭木根．数据仓库技术与实现．北京：电子工业出版社，2002
12．Jiawei Han，等 著．数据挖掘概念与技术(Data Mining Concepts and Techniques)．范明，等 译．北京：机械工业出版社，2001
13．王珊等．数据仓库技术与联机分析处理．北京：科学出版社，2001
14．Tom Hammergren 著．数据仓库技术(Data Wharehousing：Building the Corporate Knowledge Base)．曹增强 等译．北京：中国水利水电出版社，1999
15．Harjinder S．GILL 等著．数据仓库—客户／服务器计算指南．王仲谋，刘书舟 译．北京：清华大学出版社 西蒙舒斯特国际出版公司，1999
16．高人伯、陈文伟著．数据仓库和数据开采相结合的决策支持新技术．计算机世界专题综述，1997
17．陈文伟，等．数据开采和数据仓库论文集．信息与决策系统，1998，3(1)：24～29
18．王红兵．数据仓库中的元数据[J]．微机发展，1999-05
19．杨文川．数据仓库技术在统计行业应用的研究与实现．计算机科学，第29卷8月增刊(NDBC 2002)
20. SAS Institute Inc. Data Warehousing Overview Theory and Business Concepts. SAS Institute Inc., 1998
21. SAS Institute Inc. Introduction to Programming Concepts Using SAS Software. SAS Institute Inc., 2001
22. SAS Institute. Inc. SAS/ACCESS Software for Relational Databases: Reference. Version 8. SAS Institute Inc., 2002
2．戴超凡．数据仓库中元数据技术研究．计算机工程与应用，2001-14
3．石磊．OLAP与数据挖掘一体化模型的研究与发展．计算机科学 2000，Vol．27，No．5：45～49
4. Inmon W H. BuildingtheDataWarehouse. Second Edition. New York:John Wiley&Sons, 1996.
5. Jose Samos, Felix Saltor, Jaume Sistac and Agusti Bardes. Database architecture for data warehousing: an evolutionary approach. In DEXA 1998: 746-756.
6. Chaudhuri S, Dayal U. An Overview of Data Warehousing and OLAP Technology. SigmodR, 26(1):65～74
7. Lou Agosta. The Essential Guide to Data Warehousing. Prentice-Hall, 2000
8. William A Giovinazzo. Object-Oriented Data Warehouse Design. 2000
9. A. Berson, J. Smith. Data Warehousing, Data Mining and OLAP. McGraw-Hill, 1997
10．陈京民．数据仓库与数据挖掘技术．北京：电子工业出版社，2002
11．彭木根．数据仓库技术与实现．北京：电子工业出版社，2002
12．Jiawei Han，等 著．数据挖掘概念与技术(Data Mining Concepts and Techniques)．范明，等 译．北京：机械工业出版社，2001
13．王珊等．数据仓库技术与联机分析处理．北京：科学出版社，2001
14．Tom Hammergren 著．数据仓库技术(Data Wharehousing：Building the Corporate Knowledge Base)．曹增强 等译．北京：中国水利水电出版社，1999
15．Harjinder S．GILL 等著．数据仓库—客户／服务器计算指南．王仲谋，刘书舟 译．北京：清华大学出版社 西蒙舒斯特国际出版公司，1999
16．高人伯、陈文伟著．数据仓库和数据开采相结合的决策支持新技术．计算机世界专题综述，1997
17．陈文伟，等．数据开采和数据仓库论文集．信息与决策系统，1998，3(1)：24～29
18．王红兵．数据仓库中的元数据[J]．微机发展，1999-05
19．杨文川．数据仓库技术在统计行业应用的研究与实现．计算机科学，第29卷8月增刊(NDBC 2002)
20. SAS Institute Inc. Data Warehousing Overview Theory and Business Concepts. SAS Institute Inc., 1998
21. SAS Institute Inc. Introduction to Programming Concepts Using SAS Software. SAS Institute Inc., 2001
22. SAS Institute. Inc. SAS/ACCESS Software for Relational Databases: Reference. Version 8. SAS Institute Inc., 2002