Verification and Validation

Abstract

Traditional hardware and software development efforts typically incorporate well-defined approaches to testing and verifying the functionality of a system according to a set of requirements and standards. Today, with increased complexity in system design, diagnostic tools used to maintain and support these systems are having to rely on approaches from artificial intelligence to meet their requirements. Unfortunately, the discipline of testing and verifying systems incorporating AI techniques is less well understood. AI-based diagnostics has been demonstrated through several applications to offer tremendous potential to maintaining complex systems, but concern exists as to how to determine the reliability of using AI to drive complex diagnostics.