Academic Commons

Reports

Are maintenance expert systems practical now?

Waldes, Peter; Lustgarten, Janet; Stolfo, Salvatore

This paper examines the question of whether Artificial Intelligence expert system technology can be effectively applied to real world maintenance problems. Expert maintenance systems Installed to date are reviewed and the ACE (Automated Cable Expertise) knowledge-based expert system presently marketed by AT&T is described in detail. ACE provides timely trouble-shooting reports and management analyses (or telephone cable maintenance). Many design decisions faced during the construction of ACE were guided by experience gained In the development of previous expert systems. The most significant departure from "standard" expert system architectures, however, is ACE's use of a conventional data base management system as its primary source of Information Its primary sources of knowledge are the expert maintenance personnel, who previously were users of the database system I and prImers on maintenance analysis strategy The coupling of "data base" and "knowledge-base" demonstrates In a forceful way the manner In which an expert system can significantly enhance the throughput and quality of automatic maintenance support systems. Based on the success of these maintenance expert systems, explosive growth of knowledge-based expert maintenance programs is predicted. Forthcoming maintenance expert systems will assist In the repair of high capital cost equipment, for which down time is expensive or unacceptable, or for which competent repair personnel are In short supply.

Subjects

Files

More Information

Publisher
Department of Computer Science, Columbia University
Publication Origin
New York
Series
Columbia University Computer Science Technical Reports, CUCS-166a-1985
Academic Units
Computer Science
Academic Commons provides global access to research and scholarship produced at Columbia University, Barnard College, Teachers College, Union Theological Seminary and Jewish Theological Seminary. Academic Commons is managed by the Columbia University Libraries.