Using an expert system to test a logistics information system

The reliability and maintainability information system (REMIS) is in the software development phase of its first subsystem. For REMIS to be accepted by the US Air Force Logistics Command (AFLC) all program lines of code and all program branches must be executed in a controlled test. Since the development of test data and test cases to perform such tests is typically labor intensive; the authors propose using knowledge-based artificial intelligence (AI) programs to generate test data and test cases. It is suggested that AI languages are well suited to the development of such test support tools due to their special capabilities to categorize information and to search/track through data. The test data and test case creation procedures will use the REMIS software application programs as the source of data for the knowledge-base. >