Accession Number:

ADP003930

Title:

Artificial Intelligence Approaches to Monitoring System Integrity,

Descriptive Note:

Corporate Author:

COLORADO UNIV AT BOULDER INST OF COGNITIVE SCIENCE

Personal Author(s):

Report Date:

1984-06-01

Pagination or Media Count:

17.0

Abstract:

This paper introduces the concept of artificial intelligence-based expert systems for enhancing system availability through system integrity monitoring and system diagnosis. Such expert systems use knowledge and inference mechanisms to solve problems which would ordinarily require the expertise of the best human practitioners in the field. The paper demonstrates the seriousness of failure and explains why several traditional methods of ensuring system availability, reliability, and maintainability fail in certain important cases, e.g., intermittent failures. An outline of the human diagnostic problem-solving process is presented with a computational analog. A symptom-based expert diagnostic system is introduced with its early and successful results in fault prediction. Author

Subject Categories:

Distribution Statement:

APPROVED FOR PUBLIC RELEASE