Due to the nature of module integration testing the hardware based method is very extensive in terms of equipment resources and time. Because of the sequential manner of module integration testing with an emulation system and test set-up time project schedules are often exceeded. This leads in many cases to significant financial impacts and image lost. This is a common business concern with the objective to reduce time and money within that part of the development cycle for safety critical software systems.
The baseline for achieving the objective described above is an existing Test Harness system which has only been used for single module testing on the host computer system. It has to be improved in a way to handle portions of modules with regard to their specific interface conditions. This feature has to be adapted to the three portions of the Test Harness (preparation, execution control, evaluation/protocol generation).
To find out whether the alternative method of host based module integration testing leads to applicable results, a set of representative modules (about 80) of the baseline application, will be taken for the experiment. The results will be analysed and then compared to the hardware based method. Common guidelines for future projects will be derived. People will be trained for the alternative test method and usage of the enhanced Test Harness system.
The development department consists of about 50 people, 10 people are working on software. Three of them are involved in the project team.
EXPECTED IMPACT AND EXPERIENCE
Nord-Micro expects to improve the module integration testing phase, in terms of time and efficient usage of development resources. Additionally maintenance activities will also be improved in terms of regression testing.