Decision Support Systems

A data warehouse is a specialised database having techniques for the extraction of data from many databases. This data is used to construct a potentially vast database that has a historical perspective. Managers and decision makers within organisations can look for trends and patterns in order to assist in decision making. Decision support systems have the following characteristics:

• Aimed at less-structured, loosely specified problems.
• Combine use of models/analytic techniques with data retrieval.
• Able to be used interactively by non-computer people.
• Flexible to be able to change as environment and decision approach alters.

A DSS usually has to answer a factual question. A major constraint is having the right data available at the right time. This data may be spread around many, possibly hundreds of diverse application and hardware systems. A DSS has to be able to communicate across these diverse ranges.

Strategic or Operational

Operational matters relate to the day to day running of a business. They will answer questions of the 'right now' variety. How many widgets are on shelf number 4? As such they provide a 'Snapshot' of the data at that time. At any time earlier or later the answer would probaly be different. This is a 'Twinkling' database. Constantly changing.

Strategic matters are concerned with policy and planning. They require data that is not subject to such volatile change. The characteristics of a data warehouse that could provide the right kind of strategic data are:

• Subject oriented - subjects such as sales.
• Non-volatile - data placed in the warehouse is not subject to change. The data warehouse will grow, but at a slow rate.
• Integrated - data is consistent. Data that is comparable must be in the same format. Different applications may hold different data in different formats. For this data to be comparable it will to undergo an integration process.
• Time-variant - historical data is recorded.

Comments, suggestions, ideas to
Stuart Banner