A fuzzy-statistical reasoning model for bearings fault diagnosis

When searching for faults threatening a system, the human expert sometimes performs an amazingly accurate analysis of available information, eventually by only using some elementary statistics. Automating the reasoning mechanisms founding such an analysis is, in general, a difficult attempt, but also a possible one, in some cases. The paper aims to introduce a nonconventional method of bearings faults diagnosis, based upon some statistical and fuzzy concepts.