The process of developing software systems involves a series of activities in which the chances of human error appears are enormous. Errors can be started from the beginning of the process, in which targets may be specified incorrectly or imperfectly, as well as in subsequent steps. Due to the impossibility of human work and communicate seamlessly, software development must be accompanied by an activity that ensures quality, software testing.

Software testing is a critical element for ensuring the quality of software. The importance of the costs associated with errors, promotes the definition and implementation of a testing process increasingly detailed and well planned.

The commitment of our group is that the test planning as early as possible and that is what we call Early Testing. Thus the objectives of our group with reference to this area are mainly focused on trying to answer the following questions:

How early can be the planning of software testing?
To what level of detail can be reached with these schedules?
Is it possible from the requirements to generate efficient test plans?
What strategies are appropriate in the context of early testing?
Is it possible to automate 100% of software testing?
How to make an early testing of Web applications in context MDWE?

Currently, the group have failed to demonstrate that there are ways to automate the generation of test cases from requirements specifications. More accurately, has achieved a semi-automatic way to obtain test cases from use cases. In addition we are also investigating early tests in the field of Web application development. More specifically, we focus on an important feature of these applications, navigation. Therefore, at present we are focusing on automated test generation from navigational models.