A Fuzzy Model for Object Oriented Testability & Its Performance

For large software systems, the testing phase tends to have comparatively much higher cost than all the previous life-cycle phases taken together, obviously resulting in much more efforts. A good measure of software testability can help better manage the testing phase effort. This paper proposes a fuzzy model for measurement of software testability based on five-object oriented metrics. Empirical data of testing time of five software projects has been collected and used to evaluate the performance of the model.