Parallelization of Integration Tests (PINT)

Important results from the project

PINT addresses the lack of variability and interaction between different functionalities in integration testing of automotive systems. In particular, we have proposed parallelization of integration tests using Guarded Assertions as a strategy for increasing real-world representativeness, reducing testing time, and increasing defect detection. Experiments show that our proposed method provides improvement with respect to the two former objectives, but so far not with respect to defect detection.

Expected long term effects

The main result of PINT is a method for expressing; executing and analyzing parallelized automotive system test cases in a laboratory environment. The PINT results furthers the knowledge on how to effectively and efficiently establish an adequate quality assessment, and is thus highly relevant for the overall FFI objectives of shortened time-to-market and road and vehicle safety.

Approach and implementation

The project has been run as an iterative process, with recurring experimentation of preliminary results and prototypes included in the loop. During the project, we have held bi-weekly half-day working meetings with all operational project members. In addition, we have held quarterly strategic steering group meetings with all project members, including the members of the steering group of the project.