A scheme to prioritize classes at the early stage for improving observable reliability
Mitrabinda Ray, Durga Prasad Mohapatra · 2010
While testing a program using a test suite of finite size, testing the frequently used classes more thoroughly within the testing budget would lead to higher observable reliability compared to testing all classes uniformly. To achieve this, we first propose a technique to prioritize classes at the design stage according to their impact on the overall reliability of the system. The priority value for a class is calculated based on UML diagrams (Use Case and Sequence diagrams) and operational profile of the system. Then we propose a genetic algorithm-based technique to select test cases for a test suite out of a large pool of test cases such that the intensity with which each class is tested is proportionate to its priority value and the test suite is optimal under other constraints.