26 October 2009. Researchers at the Universidad Politécnica de Madrid’s School of Computing have developed a time management model applied to data warehouse systems to optimize company business intelligence.
The model is part of the response to the need to improve information processing within companies, which has led to what is known as business intelligence. The goal of business intelligence is to convert simple business transactions data into information and knowledge to support and enrich business decision making.
To set up an infrastructure that supports business intelligence it is imperative and vital to have the technology to manage, debug and integrate data from more than one source and store all these data at a single destination, repository or database for later analysis and exploration.
This technology is what is known as a data warehouse. A data warehouse is basically responsible for consolidating, integrating and centralizing the data that a company generates in all fields of business activity through a high-performance structure enabling access and exploration of the required information, then outputting a wide range of analyses for strategic and tactical decision making.
Time management model
The research developed at the UPM’s School of Computing provides a time management model that can be developed using relational database management systems (DBMS) capabilities and the latest database standard query language (SQL). This model accounts for both data structure issues and data query and modification operations, as well as selected constraints.
This time management model has adapted data structures to be able to store time-variable data. It also accounts for the extension of data modification and query operations to include time semantics. Additionally, the time management model implements the equivalent temporal version of any constraint expressable in the non-temporal model.
As a result, the model gives an overview of all aspects that need to be taken into account to build time management to a data warehouse system. Additionally, the model provides a work guide that sets out all issues in checklist format, showing all the aspects to be taken into account when undertaking a project requiring time management.
This research is a response to the complexity of time management in data warehouse systems and the need to systematically undertake system design, as the success of a data warehouse project largely depends on how correct the time management of the information it stores is. The provision of a good work guide and time management model eases the process for building the data warehouse.
Almost twenty years after the first data warehouse systems were released, we are still researching and learning about time management. And even though what are known as temporal databases, a field that has been under research since the 1970s, are still not commercially viable products, all the theory and concepts developed in this field are now being exploited to build temporal functionality into commercial relational DBMS.
But until there are temporal DBMS for the business world and temporal functionality is completely consolidated in relational DBMS, DIY is the only alternative for addressing time management. In this respect, this time management model offers a practical and systematic approach for undertaking projects requiring time management.
The model was presented as a final-year project by Laura González Macho and supervised by professor Juan Pedro Caraça-Valente.