A Logic Programming Approach to Requirements Modeling and Automated Test Generation

We present several techniques for construction of executable models of system requirements and automated test case generation using a Logic Programming (LP-based) approach to knowledge representation. We illustrate the LP-based approach with applications of the actual requirements modeling and test generation for several telecommunication services provided to customers from central office switching systems. This approach helped to prevent serious faults, which are typically introduced during the requirements phase and then remain undetected until a time when their correction is costly to the customers and suppliers. The automated generation of test case scenarios from machine executable requirements eliminated an otherwise tedious and error-prone effort, and created a better set of understandable and operational systen1 requirements. This approach can be used in any application domain and independently of any particular requirements analysis methodology.