Owens Corning Case 1. Data Provisioning OCF conceptualizes the principal IS production activities that produce value for the company as transaction processing, data provisioning, and information delivery. Data provisioning manages the inventory of data and information, using relational database management systems and a data dictionary and data catalog. Transaction processing stores only the current data necessary to provide the status of current operations. Data provisioning receives, stores, and manages all validated transaction data from transaction processing. It also provides data back to transaction processing and information-to-information delivery when requested.
Data dictionary and data catalog keep way of what data exists, where it is stored, what it means, and who has authority to access it. That helps data provisioning to protect the data and to provide access to data and information to those who need it. Data provisioning stores the organizational structure regulations in tables in its database, so these regulations can be changed without any difficulties when management decides to modify the organization. OCF created two relational databases to support data provisioning one to support transaction processing and another to support information delivery. They used for transaction processing technology IBMs DB 2 running on the mainframe, designed to process high transaction volumes fast.
For the information delivery technology they used a TERADATA database machine - a combination of hardware and software that is optimized for answering information questions. 2. Data warehouse The data warehouse is part of the data provisioning function. It could be described as a big depository.
Data warehouse must provide interfaces that accept transaction data from different types of transaction processing systems and move them into the warehouse environment. During thi process that data are tested and validated to assure that only high quality data are accepted. On the output side of the data warehouse, there must be interfaces that allow a user to ask for data from the devices (PCs or terminals) that he typically use. The information in the warehouse must be structured, so that users can easily obtain answers to their questions. OCF built data warehouse in 1987. They had selected a combination of IBMs DB 2 relational DBMS, SQL query language, and TERADATA database machine to support the data warehouse.
Few months later they built a data warehouse for the Residential and Commercial Insulation Products division that contained product data, sales data, and customer data. After some time they purchased a larger TERADATA machine and related software, which enabled them to expand this data warehouse to serve the other OCF divisions. 3. The Retail Outlet The concept of a retail outlet came from common experience. OCF used the analogy that a warehouse is really accessed by retailers, not by customers. The retail outlet is a set of data that is selected and packaged to serve the needs of a group of users.
There are some important rules when the set of data is packaged. First and most important is that only the data that are relevant to the business unit is selected. The data could be summarized according to the organizational categories and by time period. Packaging can also combine different types of data.
The data into the packages are small and focused, so that the user can simply and fast specify what is wanted. That decreases the delay in retrieving the data. Ocf first retail outlet was a profitability database for one of its divisions. They wanted to be able to look at the profitability of a product, sales office, or a specific customer. The profitability retail outlet was a successful, and they developed similar retail outlets for the other OCF divisions. The retail outlet enabled OCF to simplify its data warehouse - OCF has moved the summarized data and the organizational tables into the retail outlets.
Some other important aspect was that the retail outlet made easier the way by which OCF managers can obtain the data that they need to manage the business. 4. Executive Information System The Executive Information System (EIS) must provide all the information that managers need to identify trends and provide understanding of what exactly is happening in the business. They also need the information as fast as possible, because they have to react fast to the changes.
The retail outlets do not contain the information that managers need. Rather, the outlets contain the data from which this information can be created by using some analytical and statistical techniques. The OCF reached the same conclusion when they have been working on a project to develop an Executive Information System (EIS) for one of its senior managers. The collected information is only available to those for whom it was created. The information and the process by which it was created leave the system. The OCF started thinking about some new kind of concept which could be more effective in a way that to return the information to the system to be available for use by other OCF managers, or to be used again to create similar information in the future.
Thus they created the information warehouse concept. The main idea of the concept was to be another set of databases designed to supply simple and fast access to the management information and the processes used to create it. Moreover, the information warehouse has to be able to provide the tools to turn this information into presentations and also to store these presentations. The OCF people considered that the necessary tools needed to support an information warehouse (such as information warehouse store compound documents that may include image, voice, and graphics information) do not exist yet but that they are being developed.