[an error occurred while processing this directive]



Book Contents

Ch. 7
Building Data-Driven Decision Support Systems

Chapter Contents
Previous Page
Next Page

Online Analytical Processing

OLAP, online analytical processing and multidimensional analysis refers to software for manipulating multi-dimensional data. Even though we can have multi-dimensional data in a data warehouse, the OLAP software can create various views and more dimensional representations of the data. According to Nigel Pendse (see http://www.olapreport.com/FASMI.HTM), OLAP software provides fast, consistent, interactive access to shared, multi-dimensional information. Pendse calls this the FASMI test, fast analysis of shared, multi-dimensional information test. What does it mean?

Figure 7.1 An example of a multi-dimensional data cube.

FAST means that the system delivers most responses to users within about five seconds. ANALYSIS means that the system can cope with any business logic and statistical analysis that is relevant for the application and the user. SHARED means that the software has security capabilities needed for sharing data among users. MULTI-DIMENSIONAL is an essential requirement. An OLAP system must provide a multidimensional conceptual view of the data. The term INFORMATION means the software can support all of the data and derived information that managers need.

OLAP software often accesses Multi-dimensional databases. A multi-dimensional database captures and presents data as arrays. Variables hold data in a multi-dimensional database. The multi-dimensional database management system creates arrays of values, usually numeric, that are "dimensioned" by relevant attributes. For example, YEAR, PRODUCT MAKE, and COLOR may dimension a UNITS SOLD variable. This three-dimensional array is often visualized as a cube of data (see Figure 7.1).

Multi-dimensional databases can have multiple variables with a common or a unique set of dimensions. This multi-dimensional view of data is especially powerful for OLAP applications. We can sum units in dimensions. A relational database software package can also be used to structure data to support rapid, multi-dimensional queries. A Star schema is a typical structure implemented for multi-dimensional data using a relational data based management system (see Raden, N. Star Schema 101, http://www.netmar.com/~nraden/str101.html). A Star schema has a central fact table and dimension tables linked by keys (see Figure 7.2). The star is a picture of the way the data is being stored. The basic factual information is in the middle of the star. This type of application where multi-dimensional data is stored in a Relational Data Base Management System has been called ROLAP for relational OLAP.

Figure 7.2 Star Schema Diagram

OLAP usually provides drill down and drill up capabilities. This is an analytical technique that lets a DSS user navigate among levels of data ranging from the most summarized data (up) to the most detailed data (down).

Software reviewer Jay Tyo has divided OLAP tools into five broad types. First are stand-alone desktop OLAP tools, including products like Cognos’ PowerPlay (http://www.cognos.com) and Andyne Computing’s Pablo. Second are integrated desktop tools, such as Business Objects (http://www.businessobjects.com) and BrioQuery Enterprise (http://www.brioquery.com). Third are relational OLAP tools, including IQ Software’s IQ/Vision. Fourth are personal multi-dimensional databases, such as Pilot Software’s Pilot Desktop (http://www.pilot.com) and Oracle’s Personal Express (http://www.oracle.com). And finally, he identified some OLAP tools, which he found harder to classify, including the SAS Institute’s SAS System (http://www.sas.com). Categorizing Data-Driven DSS products is complex and difficult. New products are continually being introduced. So, managers and DSS analysts are confronted with a wide array of different types of OLAP products. The OLAP terminology is also somewhat confusing, so for help with OLAP terms, check the Guide to OLAP Terminology created by the OLAP Council at http://dssresources.com/glossary/olaptrms.html and check the DSS Glossary at DSSResources.COM (http://dssresources.com/glossary).

Business Intelligence (BI) is sometimes used interchangeably with OLAP. BI is however a popularized umbrella term introduced by Howard Dresner of the Gartner Group in 1989 to describe a set of concepts and methods to improve business decision making by using fact-based support systems. The term is used interchangeably with OLAP, briefing books and Executive Information Systems. A Business Intelligence System is a marketing-oriented term for a Data-Driven DSS.

[an error occurred while processing this directive]