Final Project (Project 2) Requirements - ISDS 556 BI/ DW System Design Project See due dates on last page of syllabus This group project is intended to give you practical experience in identifying and specifying requirements for a business intelligence/ data warehousing solution. This project will be done in groups of about 4 students. These groups will be assigned alphabetically by the instructor before class commences (see posted Teams sheet on Canvas). For this project, read the provided case study, and imaging you are consultants that help the leaders generate a BA plan that caters to their needs. The deliverable is a plan or a consultant report- not a BI solution (i.e., there is no coding needed). The project write-up should build on concepts from the textbook and PPTs, and should cover the following topics (items 1-3 are also included in the proposal): 1. Overview Description of the principle and the idea. A few sentences that describe what the project is and why it is needed, What are the gaps between current analytical capabilities and desired needs? Is the organization ready? High level enterprise data warehouse bus matrix (Chapter 3, e.g., pg. 90) 2. Description of the functionality and features of the required/desired BI system. Which process/ KPIs will you focus on? Types of reports and analyses, frequency, types of users, etc. (Build on Chapter 3 & Chapters 11&12) - what are the reports you expect to generate? Who will use them? What are the applications that will extract data from the data warehouse for presentation and/or analyses? 3. What are the data models that need to be implemented for the system to meet the business requirements? Your data model/s should have at least one fact table + several dimension tables, each containing multiple fields. Justify the selection of model type (see my PPTs), snow-flaking (needed or not?), the treatment of slowly changing dimensions, fast changing dimensions, degenerate dimensions, etc. (Build on Chapters 6 &7 & the extra material on Titanium) 4. Data Analysis - what data are available? At what grain? Are the data from primary (e.g., an operational system) or secondary (e.g., census) sources? Are these data sources appropriate? Need major cleansing? Etc. If needed, generate a matrix of data source mapped to the measures of interest 5. Extraction, Transformation, and Loading issues - which are the crucial issues and how will you mitigate them?

2 Briefly discuss- What are the transformations that need to be implemented for the system to work? Validity checks? Aggregations (granularity changes)? Integration of data from multiple sources? Partitioning? De-normalization? Etc.(Build on Chapters 4,5,6, 7 & 9) Scheduling issues? When would it be best to extract and/or load data? Why? How frequently? Do you need real-time or near real-time loading? 6. The recommended architecture Draw a diagram of the different products that you will use to meet the business needs. Include all backroom and front-room technologies. Justify the use of specific solutions (we do mention some vendors throughout the course) when appropriate. Build on Chapters 4&5. You may specify particular products and vendors (but you do not have to), 7. Summary, conclusion & recommendation for action. - What did you describe in the report? What are the next steps? Please remember that this project is a proposal that is limited to only 15 pages count. Append important material to these pages as needed. Following are the metrics for grading your project: Coverage of topics as outlined in the project description Clear problem statement & need Accuracy and optimality of the data model Quality of proposed architecture To what extent does the solution solve the reporting/analytical needs of the organization? Overall quality of the proposed solution Overall write-up quality & flow