December 7, 2022

Pre-engineering is the time to define the goals of the application, to make the initial analysis of the system and make other decisions. This includes deciding whether the data warehouse should be build at all. Each phase of pre-engineering process should be put to better understand the system from a data warehousing point of view.

The first step in identifying the right application is to secure executive buy-in. Once the executive backing has been granted, the next step would be to determine what the company’s business objectives are and how the data warehousing application will help to achieve them. We may ask the following questions about the goal of building the data warehouse.

  • Is the company is in great need to handle & manage bulk of enterprise information?
  • Is the company trying to build up a system that supports intelligent decision making & trend analysis from the bulk of distributed enterprise data?
  • Is the company preparing for BPR by IT?
  • Is the company attempting to combat Business competition? Is the company preparing for deregulation of prices?
  • Is the company trying to increase revenue by building up total quality information management setup?
  • Is company trying to understand its customer base better by analyzing the bulk of current and historic data?
  • Is the company trying to increase the market share and reduce cost by implementing the data warehouse?

Now if company is upto any of the above items and if it has sufficient executive sponsorship and if it is sure to get the ROI faster, it may sure go for a data warehouse project to attain its goals, otherwise it shouldn’t.

Once the broad parameters of the application are identified, (e.g. deciding that it should build up a system that supports intelligent decision-making & trend analysis from the bulk of distributed enterprise data) the “how to” must be addressed.

As the next step, a data warehouse project team is assembled and tasks such as Analysis, Feasibility study, Design & Implementation are as The next step is to study and analyse the system to set up the base for starting up the data warehouse design. For eg. we may ask the following questions.

  • What the application should do?
  • What are the performance criteria?
  • What type of system architecture may be chosen?
  • What are the system integration details? How much historical data is to be retained?
  • What type of database (Relational or object oriented etc) is required?
  • What are the distributed application concerns?
  • What are data entities, attributes, relations, data sources, data flows and work flows involved in the system that are important while designing the data warehouse?
  • How the Data warehouse database may be designed and implemented?
  • How the baseline OLTP systems may be designed, implemented and integrated with the Data warehouse?
  • How the distributed Data warehousing system may be implemented and integrated?
  • How the Data warehouse operations such as loading, refreshing etc be designed & implemented
  • What type of analysis queries and reports are needed at the front end?

Answers to these questions gives us the basic skills and supplies needed to build up the Data warehousing system. At the end of the pre-engineering stage, fiscal budgets & timelines and the baseline system models can be created. An early budget is an essential guide that will assist in determining the choices for the application. One method for identifying the budget is by performing Cost/benefit analysis of the system. Here we understand the value of the information for the enterprise and understand the benefit it provides. From this we can implement a plan and budget and calculate ROI.