Developing an expert system of failure analysis

Yeong Ho Ho, Huei Sen Wang, Hei Chia Wang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

An Equipment Failure Analysis Expert System (EFAES) is to be developed to help the engineers diagnose the causes of the failure mechanism and provide a reliable remedy. This system is based on an innovative reasoning approach: integrating the rule-based reasoning (RBR) and the case-based reasoning (CBR) methods The architecture developed in the system consists of six major elements-"Factor and Attribute Editor", Knowledge Actuation Interface", "Knowledge Base", "User Interface", "Inference Engine" and "Explanation Facility". Here, the RBR system consists of 46 failure mechanisms and their rules. The CBR system consists of 586 failure cases which are coded and composed from 23 factors and their 265 attributes. Also, this system provides a variety of inference methods which allows retrieving the best answers to users. For the RBR system, performance is directly check the inferred order of the document ranking list. For the CBR system, the effectiveness of each inference method is evaluated by using "Recall", "Precision", and "F-Measure" approaches. From the test results, many recommendations are proposed.

Original languageEnglish
Title of host publicationInnovation for Applied Science and Technology
Pages2375-2379
Number of pages5
DOIs
Publication statusPublished - 2013 Feb 20
Event2nd International Conference on Engineering and Technology Innovation 2012, ICETI 2012 - Kaohsiung, Taiwan
Duration: 2012 Nov 22012 Nov 6

Publication series

NameApplied Mechanics and Materials
Volume284-287
ISSN (Print)1660-9336
ISSN (Electronic)1662-7482

Other

Other2nd International Conference on Engineering and Technology Innovation 2012, ICETI 2012
CountryTaiwan
CityKaohsiung
Period12-11-0212-11-06

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Fingerprint Dive into the research topics of 'Developing an expert system of failure analysis'. Together they form a unique fingerprint.

Cite this