EXPERT AND KNOWLEDGE-BASED SYSTEMS IN FAILURE ANALYSIS

All areas of industry are affected by premature failure of components which must be investigated so as to avoid future repetition. However, the complexity of the interrelations between design stress and environment often makes this a daunting task, except for the engineer with considerable knowledge or experience in failure analysis. Modern developments in information technology should allow this task to be made easier by storing and subsequently accruing this ''experience'' by using expert and knowledge-based systems. The application of these systems to the task of failure analysis, and design against failure, is reviewed, particular emphasis being placed on the reasoning methodologies used and the knowledge domains for which these methodologies are appropriate. Case-based reasoning techniques are considered to be the most suitable for generic failure analysis due to the complexity of knowledge required. It is concluded that future trends in diagnostic expert systems will be based on the holistic hybrid rule-case-based reasoning approach combined with a number of stand-alone engineering failure analysis calculating software tools, and a multimedia-type knowledge-based system for different failure modes.