Knowledge Management; A System for Controlling the AI Development Environment for an Expert System Which Performs Satellite Anomaly Resolution

This paper introduces the concept of knowledge management as a comprehensive system for managing and controlling the development and validation of a knowledge base which is a part of an expert system application. In most applications the complexity of the system and the representation used tend to obscure the structure and data which describe the knowledge domain which effectively inhibits Knowledge Verification and Validation and places severe constraints on the ability to extend and maintain a knowledge base. Ford Aerospace has pioneered a unique Knowledge Management system through the evolution of knowledge representation systems and knowledge acquisition tools which overcomes these traditional limitations. It can be anticipated that as expert systems experience wide spread application, Knowledge Management will become an important component of the standard development methodology employed by most firms working in the field. In the near term, researchers need to continue to refine and extend the concept so its application achieves its potential for improving the quality and scope of AI projects.