Integrating Graphical and Natural Language Specifications to Support Analysis and Testing

The ongoing trend towards distributed development activities causes a growing need for specification activities and techniques. Each component leads to a large number of specification documents being exchanged, change managed and committed. The quality of the specifications influences the timing, costs and success of the development task. However, the quality of such specifications is often far from optimal, exhibiting gaps, inconsistencies, redundancies, and unbalanced structures. At every release or delivery milestone, acceptance and integration testing take place. Therefore, test-cases have to be created from the requirements exchanged. This paper presents a model-based approach for improving the quality of comprehensive requirements sets. The presented solution is based on a combination of a graphical notation and natural language and can be used to drive model-based testing. The approach has been implemented using state-of-the-art tools. We present experience from field application in the automotive industry.