Implementation of a fault tree method

Satu Kärki, Juhani Hyvärinen

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review


Information concerning a fault in a process can be broken down to components using a tree diagram where a so-called major event forms the root of the diagram and the reasons, contributing to the event taking place, form the branches of the tree. The tree diagram and sets of symptoms can be combined relatively easily in order to make it possible to handle both top-down and bottom-up approaches in the reasoning process. A fault tree consists of a hierarchical structure in which one fault can explain one or several faults on a lower level. A set of symptoms can be used for indicating that a specific fault has occured. In this paper, the implementation of a fault tree as a basis of a fault diagnosis system is discussed. General requirements for the implementation are identified and the structure of a fault diagnosis system is defined. Examples of user interfaces of fault tree implementations are shown. In addition, solutions of different user interface possibilities are described.
Original languageEnglish
Title of host publicationApplications in Artificial Intelligence in Engineering XI, AIENG 96
EditorsR.A. Adey, G. Rzevski, A.K. Sunol
Place of PublicationSouthampton
PublisherComputational Mechanics Publications
ISBN (Print)1-85312-410-9
Publication statusPublished - 1996
MoE publication typeA4 Article in a conference publication

Publication series

SeriesWIT Transactions on Information and Communication Technologies


Dive into the research topics of 'Implementation of a fault tree method'. Together they form a unique fingerprint.

Cite this