I have been giving information a lot of thoughts, what I see around me is companies building data warehouses to gain insight to processes and risk through data.

Traditional Data warehousing is a limited pattern

An organization registers data in different systems, which support the various business processes. In order to create an overall picture of business operations, customers and suppliers in one single version of the truth, the data must come together in one place and made compatible. The structure of a data warehouse is specifically designed to quickly analyze such large amounts of data.

The structure of both data warehouses and data marts is aimed for flexible and quick reporting. This allows for interactive analysis on the basis of various predefined dimensions. End users can quickly gain knowledge about business operations and performance indicators.

Acting on and understanding data is left to the business users; and it may quickly lead to an information overload that drowns the business; and results in an inability to optimize the processes based on the very same information.

Information Sophistication reaches a level where information becomes a necrology over what happened, and rather than a tool steering through icebergs as they arises on the path.

Maturing from data necrology to prescriptive acting

Increasing Information Sophistication and with the aspiration of avoiding to drown in the amount of data, we take three new areas of analysis in use.

Descriptive analysis or statistics does exactly what the name implies they “Describe”. By using data aggregation and data mining techniques the aspiration is to provide insight into what has happened. Descriptive analytics are useful because they allow us to learn from past behaviors, and understand how they might influence future outcomes.

Predictive analytics seeks ability to “Predict” what might happen. Where descriptive was aimed against the path is the aspiration Predictive analytics the ability to understand “What could happen”. Predictive analytics provide estimates about the likelihood of a future outcome.

Prescriptive Analytics, in a nut-shell, these analytics are all about providing advice. Through simulation algorithms we try we seek possible outcomes to advice on what action should be taken. The goal of Prescriptive analytics is to quantify the effect of different decisions, allowing the decision maker to base his decisions on simulated scenarios.

These three kinds of analysis create the fundament of Optimization. By feeding back actions based on an optimal Prescriptive scenario we can fine tune operational application and processes. Optimization these analytic techniques allows for quick adapting to events. These advanced analytical patters are much needed where the amount of data is vast and complex, and timely reaction is key to operate the market.

Examples:

• Supply chain management profitability maximization
• Solar and wind farm energy optimization
• Stock trading
• Optimal bank loan credit scoring