A Diagnostic Expert System Which Generates Explanations

  • Soheir Ghallab
  • Janet Efstathiou


Fault diagnosis is one of the largest application domains in expert systems. But most of diagnostic systems concentrated on constructing systems for simulation and diagnosis, and neglected the user’s (operator) need for explanations for the reasoning. The objective of this work is to build an expert system for fault diagnosis of a process control system, which is able to cope with both single and multiple faults, and generates explanations describing the reasoning underlying a particular diagnosis. The explanation does not merely consist of just a trace of rules which have been activated. Instead the explainer responds to the user’s interrogation of the diagnosis system’s decisions for choosing a particular fault. The user can also propose his own hypotheses which are then evaluated by the system, then the explainer will generate explanations for the possibility of these alternative solution(s).


Expert System Fault Diagnosis Diagnosis System Multiple Fault Process Control System 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Davis R. [1984]:“Diagnostic Reasoning Based on Structure and Behaviour”, Artificial Intelligence, vol 24.Google Scholar
  2. 2.
    Davis R., Shorb H., Hamscher W. [1982]:“Diagnosis Based on Description of Structure and Function”, In proceedings of AAAI-82, August 16–20 1982.Google Scholar
  3. 3.
    Koukoulis C.G. [1986]: “The Application of Knowledge Based Techniques to Industrial Maintenance Problems”, PhD Thesis, Department of Electrical and Electronics Engineering, Queen Mary College, University of London.Google Scholar

Copyright information

© Kogan Page Ltd. 1989

Authors and Affiliations

  • Soheir Ghallab
    • 1
  • Janet Efstathiou
    • 1
    • 2
  1. 1.Department of Electrical Engineering, Queen Mary CollegeLondon UniversityUK
  2. 2.Step CentreOxford TrustOnsey MeadUK

Personalised recommendations