Moss L.T., Atre S. Business intelligence roadmap: The complete project lifecycle for decision-support applications
Подождите немного. Документ загружается.
[ Team LiB ]
Roles Involved in These Activities
Application lead developer
The application lead developer should be prepared to discuss access and
analysis issues on the review agenda. He or she may also actively participate in
discussions regarding database design, tool selection, and technical
infrastructure. Topics such as OLAP tools, ease of use, reports, and the
extraction of data into private data sets require input from the application lead
developer as well.
BI infrastructure architect
The BI infrastructure architect should be prepared to discuss the technical
infrastructure components, such as servers, network, the database
management system (DBMS), and the various tools. In addition, the BI
infrastructure architect should be able to address the scalability of the current
platform and the plans for expanding the platform for future BI application
releases.
Business representative
The business representative who was involved in the BI project provides his or
her opinions—from a business person's perspective—about the development
process. He or she may comment on budgetary issues, the project plan, the
overall management of the project, the effectiveness of his or her own
contribution to the project, testing activities, or on any other topic he or she
feels could be improved.
Business sponsor
The business sponsor spearheads the BI project from its inception to its
implementation and initiates the post-implementation review. He or she sends
the invitation to the attendees, prepares opening remarks for the review, and
closes the session.
Data administrator
The data administrator must be prepared to discuss the data requirements and
the business decisions made during the requirements and analysis activities.
The data administrator should review the procedure used for resolving data
disputes and make recommendations for improving it. He or she can also
contribute to discussions about the nontechnical infrastructure.
Data mining expert
The data mining expert should be prepared to suggest any necessary
improvements to the BI application (or the BI decision-support environment as
a whole) that could enhance the use of the data mining tool. These suggestions
could involve discussions about the extract/transform/load (ETL) process, the
cleanliness of the data, and the completeness of the data, as well as any
limitations the current technologies impose on data mining activities.
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
461/631
Data quality analyst
The data quality analyst should be a very active participant on the topics of
source data analysis and data cleansing. He or she must be prepared to present
a summary of data quality problems. This summary should include the impact
of the bad data on the BI application as well as on the operational systems. The
data quality analyst must also explain what type of cleansing is being
performed in the ETL process. He or she should send out a document before the
meeting that lists the data elements not being cleansed, indicating whether
they are being rejected or moved into the BI target databases as is. He or she
should also be able to discuss the current triage procedure for prioritizing the
source data for cleansing.
Database administrator
The database administrator must be prepared to discuss the design and content
of the BI target databases and how to navigate through them. If one of the
session topics is database design or database performance, it is the database
administrator's responsibility to explain the database design decisions made
during the project.
Developers
All developers, whether ETL developers, application developers, meta data
repository developers, or Web developers, should be encouraged to share their
experiences with the development process. This could include discussions about
methodology, status reporting, technical infrastructure components, testing
procedures, and any other topics directly related to their development
activities.
ETL lead developer
The ETL lead developer should be prepared to discuss the data transformations
being performed in the ETL process. He or she must be able to explain how the
data from the source systems is being reconciled to the BI target databases and
where the reconciliation totals can be viewed. On the technical side, the ETL
lead developer may actively participate in platform scalability and tool
discussions.
Facilitator (not on the BI project team)
The facilitator must be someone who was not directly involved with the project.
This person must have training in facilitation. His or her responsibilities include
time-boxing questions and discussion topics, halting runaway discussions,
keeping on schedule with the agenda, giving turns to attendees to speak, and in
general "directing traffic" at the review meeting.
Meta data administrator
The meta data administrator must be prepared to discuss what meta data is
available and how to access it. He or she should send out a document with
meta data examples prior to the meeting. The meta data administrator should
also walk through the ETL load metrics that are being captured in the meta data
repository, such as ETL reconciliation totals, trapped data error totals, and data
quality (reliability) statistics.
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
462/631
Project manager
The project manager is primarily responsible for organizing the review session.
This includes finding a venue, creating the agenda, shipping documents to be
reviewed during the session, scheduling the session, arranging for a facilitator
and a scribe, and following up on action items.
Scribe (not on the BI project team)
The scribe's main responsibility is to document the review discussions. The
scribe writes and distributes the minutes of the meeting, prepares and
distributes the action item list, and helps with other administrative activities.
Stakeholders (including data owners)
Occasionally other stakeholders may want to participate in the post-
implementation review. Stakeholders could be business people from other
departments, the data owners of the operational source data, or IT staff from
other BI applications who want to benefit from the lessons learned.
Stakeholders could also be staff from operations, technical support, or the help
desk. In general, stakeholders (with the exception of the data owners) do not
actively participate in the review discussions.
Subject matter expert
The subject matter expert is an active participant who represents the business
view during the review discussions. Topics of data quality, meta data, ease of
use, and problem resolution procedures should be of particular interest to this
person. The subject matter expert may also contribute to the discussions
regarding cost justification, the measurement of ROI, the impact on operational
systems, and potential improvements to business processes.
Web master
The Web master must be prepared to review the Web application issues and to
answer questions regarding data access capabilities through the Web, as well
as data security on the Web server. If some data is accessible globally through
the Internet, the Web master should invite the security officer to the meeting to
answer global security and privacy questions raised during the review session.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
463/631
[ Team LiB ]
Risks of Not Performing Step 16
As George Santayana once said, "Those who cannot remember the past are
condemned to repeat it."
[1]
This statement applies to BI projects as much as it does
to life and politics. In order to know how to improve the next project, you have to
learn from the mistakes made on the last project. The post-implementation review is
the vehicle for discovering the mistakes and correcting them. Excluding this step
would result in repeating the same mistakes in a rapidly growing environment that
affects more people with each release. We have learned from experience that
correcting mistakes on small systems is much easier than correcting mistakes on
large systems. The BI decision-support environment can quickly become a very large
system!
[1]
Santayana, George. Life of Reason, Reason in Common Sense. Scribner's, 1905, p. 284.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
464/631
[ Team LiB ]
Bibliography and Additional Reading
Adelman, Sid, and Larissa Terpeluk Moss. Data Warehouse Project Management.
Boston, MA: Addison-Wesley, 2000.
Beck, Kent. Extreme Programming Explained: Embrace Change. Boston, MA:
Addison-Wesley, 2000.
Cockburn, Alistair. Agile Software Development. Boston, MA: Addison-Wesley, 2002.
DeMarco, Tom. Slack: Getting Past Burnout, Busywork, and the Myth of Total
Efficiency. New York: Broadway Books, 2001.
English, Larry P. Improving Data Warehouse and Business Information Quality:
Methods for Reducing Costs and Increasing Profits. New York: John Wiley & Sons,
1999.
Humphrey, Watts S. Winning with Software: An Executive Strategy. Boston, MA:
Addison-Wesley, 2002.
Inmon, William H., J. D. Welch, and Katherine L. Glassey. Managing the Data
Warehouse: Practical Techniques for Monitoring Operations and Performance,
Administering Data and Tools and Managing Change and Growth. New York: John
Wiley & Sons, 1997.
Kuan-Tsae, Huang Yang W. Lee, and Richard Y. Wang. Quality Information and
Knowledge. Upper Saddle River, NJ: Prentice Hall, 1999.
Yourdon, Edward. Death March. Upper Saddle River, NJ: Prentice Hall, 1997.
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
465/631
[ Team LiB ]
Part II: At a Glance
Chapter Seventeen. Human Resource Allocation Matrix
Chapter Eighteen. Entry & Exit Criteria and Deliverables Matrix
Chapter Nineteen. Activity Dependency Matrix
Chapter Twenty. Task/Subtask Matrix
Chapter Twenty-One. Practical Guidelines Matrix
Appendix Work Breakdown Structure
[ Team LiB ]
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
466/631
[ Team LiB ]
Chapter Seventeen. Human
Resource Allocation Matrix
Listed in this matrix are roles, not people! Multiple
roles can and probably will be assigned to one individual;
occasionally, one role may be assigned to multiple
individuals.
Development Step
[a]
Vital Roles Involved in the Step
[b]
1. Business Case Assessment
Business representative
Business sponsor
Data quality analyst
Project manager
Subject matter expert
2. Enterprise Infrastructure
Evaluation
Section A. Technical Infrastructure
Evaluation
BI infrastructure architect
Database administrator
Section B. Nontechnical
Infrastructure Evaluation
BI infrastructure architect
Data administrator
Data quality analyst
Meta data administrator
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
467/631
3. Project Planning
Application lead developer
Business representative
Data administrator
Data quality analyst
Database administrator
ETL lead developer
Meta data administrator
Project manager
Subject matter expert
4. Project Requirements Definition
Application lead developer
Business representative
Data administrator
Data quality analyst
Meta data administrator
Subject matter expert
5. Data Analysis
Business representative
Data administrator
Data quality analyst
ETL lead developer
Meta data administrator
Stakeholders (including data owners)
Subject matter expert
6. Application Prototyping
Application lead developer
Business representative
Database administrator
Stakeholders
Subject matter expert
Web master
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
468/631
7. Meta Data Repository Analysis
Data administrator
Meta data administrator
Subject matter expert
8. Database Design
Application lead developer
Data administrator
Database administrator
ETL lead developer
9. Extract/Transform/Load Design
[c]
Data quality analyst
Database administrator
ETL lead developer
Subject matter expert
10. Meta Data Repository Design
BI infrastructure architect
Data administrator
Meta data administrator
11. Extract/Transform/Load
Development
Business representative
Database administrator
ETL developers
ETL lead developer
Subject matter expert
Testers
12. Application Development
Application developers
Application lead developer
Business representative
Database administrator
Subject matter expert
Testers
Web developers
Web master
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
469/631
13. Data Mining
Business representative
Data mining expert
Database administrator
Subject matter expert
14. Meta Data Repository
Development
Business representative
Database administrator
Meta data administrator
Meta data repository developers
Testers
15. Implementation
Application developers
Application lead developer
Data mining expert
Database administrator
ETL developers
ETL lead developer
Meta data administrator
Meta data repository developers
Web developers
Web master
16. Release Evaluation
Application lead developer
BI infrastructure architect
Business representative
Business sponsor
Data administrator
Data mining expert
Data quality analyst
Database administrator
Developers (ETL, application, meta data
repository, and Web)
北斗成功社区 BeiDouWeb.com 教育音视频/电子书/实用资料文档/励志音乐影视 仅供免费试用/版权原著所有
470/631