Artificial Intelligence
KARDIO: a study in deep and qualitative knowledge for expert systems
KARDIO: a study in deep and qualitative knowledge for expert systems
Hierarchical model-based diagnosis based on structural abstraction
Artificial Intelligence
A layered approach to automated electrical safety analysis in automotive environments
Computers in Industry
Simulating electrical devices with complex behaviour
AI Communications
A layered approach to automated electrical safety analysis in automotive environments
Computers in Industry
Automated FMEA based diagnostic symptom generation
Advanced Engineering Informatics
Qualitative order of magnitude energy-flow-based failure modes and effects analysis
Journal of Artificial Intelligence Research
Hi-index | 0.00 |
The Failure Mode and Effects Analysis (FMEA) design discipline involves the examination at design time of the consequences of potential component failures on the functionality of a system. It is clear that this type of information could also prove useful for diagnostic purposes. Unfortunately, this information cannot be fully utilised for diagnosis when FMEA has been performed by human engineers, because of inconsistencies in effect descriptions. The FMEA process is also very. time consuming, with the consequence that the engineer can only deal with single point failures. Automation of the electrical FMEA process facilitates information reuse for diagnosis by providing consistent descriptions of failure effects, and by speeding up the FMEA process to such an extent that it becomes feasible to examine multiple failures. This paper introduces the advantages that automated FMEA provides for diagnosis, and describes its use for generating fault trees from the FMEA report. The paper examines the current limitations of FMEA use for diagnosis, and reports on how these limitations may be overcome.