Title :
From fault trees to fuzzy relations in managing heuristics for technical diagnosis
Author :
Ulieru, Mithafla
Author_Institution :
Inst. of Autom. Control, Tech. Univ. Darmstadt, Germany
Abstract :
A generic possibilistic framework for embracing through fuzzy relations the causal-abductive duality co-existent within the diagnostic information is proposed. As a challenge of the classical fault tree analysis, the proposed approach overpasses the main limitations of the restrictive framework of reliability theory by processing linguistic expert knowledge through the deterministic trees. On a very simple example it is illustrated on how fuzzy relations represent an excellent tool for unified integration of heuristic diagnostic information, as a basis for managing complex diagnostic situations via approximate reasoning
Keywords :
fault diagnosis; fault trees; fuzzy logic; inference mechanisms; reliability; reliability theory; uncertainty handling; approximate reasoning; causal-abductive duality; deterministic trees; fault diagnosis; fault trees; fuzzy relations; heuristic diagnostic information; linguistic expert knowledge; reliability theory; technical diagnosis; Artificial intelligence; Cause effect analysis; Connectors; Equations; Fault diagnosis; Fault trees; Reliability theory; Tree data structures; Uncertainty;
Conference_Titel :
Systems, Man and Cybernetics, 1993. 'Systems Engineering in the Service of Humans', Conference Proceedings., International Conference on
Conference_Location :
Le Touquet
Print_ISBN :
0-7803-0911-1
DOI :
10.1109/ICSMC.1993.384903