Title :
An adaptive-learning expert system for maintenance diagnostics
Author :
Tran, Luc P. ; Hancock, John P.
Author_Institution :
McDonnell Douglas Corp., St. Louis, MO, USA
Abstract :
The Artificial Intelligence Applications to O-level Maintenance (AIATOM) expert system is the product of research into how artificial intelligence experience-based learning can improve the accuracy and cost effectiveness of fault diagnosis in a military maintenance environment. An adaptive diagnostic maintenance advisor, developed using LISP on a VAX system, learns new symptoms and forms new associations between sets of known symptoms and maintenance actions. However, the system does not learn new maintenance actions. The current system has avionic and nonavionic knowledge bases for the nose wheel steering system and the stores management system, respectively, of the US Navy F/A 18 Hornet. The general capabilities of the user interface in AIATOM and the design of an adaptive-learning maintenance-assistant system are described
Keywords :
adaptive systems; expert systems; learning systems; maintenance engineering; military computing; Artificial Intelligence Applications to O-level Maintenance; adaptive-learning expert system; design; diagnostic maintenance advisor; experience-based learning; fault diagnosis; knowledge bases; maintenance actions; military maintenance; symptoms; user interface; Aerospace electronics; Artificial intelligence; Costs; Diagnostic expert systems; Fault diagnosis; Knowledge management; Learning; Nose; Steering systems; Wheels;
Conference_Titel :
Aerospace and Electronics Conference, 1989. NAECON 1989., Proceedings of the IEEE 1989 National
Conference_Location :
Dayton, OH
DOI :
10.1109/NAECON.1989.40337
