Abstract
A methodology is proposed for inferencing and decision-making under uncertainty in production-rule analysis systems where the knowledge about a given system is both probabilistic and possibilistic. In this approach, uncertainty is considered as consisting of two parts: Randomness, meaning the uncertainty of occurrence of an object, and Fuzziness, expressing the imprecision of the meaning of the object. The analysis uses the concepts of the probability of a fuzzy event and of information granularity. Application to the monitoring of engineering devices is demonstrated, where the propagation of the coupled probabilistic and possibilistic uncertainty is carried out over a model-based system using the Knowledge Engineering Rule-Based paradigm. The approach provides a quantification for combining the concepts of 'Reliability' from probability theory, and of 'Performance Level' from the theory of fuzzy subsets.
| Original language | English (US) |
|---|---|
| Pages | 97-105 |
| Number of pages | 9 |
| State | Published - 1986 |
ASJC Scopus subject areas
- Industrial and Manufacturing Engineering
- Mechanical Engineering