A common problem in the training simulator community is the effective sharing and reuse of data. Each new simulator effort tends to set up its own infrastructure to acquire the data necessary to run the simulator. This was one of the specific problems addressed by the Universal Threat System for Simulators (UTSS) program. The goal was to define common data requirements for a family of training devices and to map these requirements from source databases to the individual data elements required by the training simulators.

The Integration Definition for Information Modeling (IDEF1X) data modeling language was employed with a reverse engineering methodology to accomplish this task. Using this methodology, the UTSS team constructed IDEF1X data models to document the data required for each of the simulators to be served by UTSS. The format of the data varied widely from simulator to simulator, with some simulators employing relational database management systems while others used formatted ASCII files. The resulting data models were then integrated into a common data model which described the data requirements for the family of simulators. Source databases were also modeled using the reverse engineering methodology and the resulting data elements were mapped to the common data requirements model.

In addition to identifying the common data requirements and the sources of data necessary to meet these requirements, several side benefits were realized. Many of the simulator administrators gained additional insight into the rules governing their data. This discovery and documentation of the rules governing the simulator data provide the opportunity to improve the quality of data, thereby reducing down-time from data errors encountered at run-time. Another major benefit of this effort is the identification of a common set of data requirements, data sources, and a data architecture that can be used for the development of new simulators.