KNOWLEDGE-BASED TEST PLANNING - FRAMEWORK FOR A KNOWLEDGE-BASED SYSTEM TO PREPARE A SYSTEM TEST PLAN FROM SYSTEM REQUIREMENTS

Early planning for system acceptance testing will result in better prepared and executed tests of a system being delivered. Since system acceptance testing is tightly coupled to system requirements, a test plan can be derived directly from requirements statements. For complex systems with hundreds of requirements, knowledge-based support has the potential to improve the quality and timeliness of a test plan document. This article discusses a framework for such knowledge-based support. First, it describes a taxonomy for classifying requirements an two exercises given to testers to validate the taxonomy and the classification concept. Then, the test-types taxonomy, a rudimentary knowledge base, and matching algorithms are presented. The article gives a complete example of matching requirements statement to an appropriate test type for demonstrating system compliance. An overview is presented of ongoing efforts to construct a robust, knowledge-based product that will implement this framework to automatically prepare a system test plan from requirements.