TABLE OF CONTENTS Traditional RCM analysis focuses on equipment reliability from the system down. New system analysis ignores pre-existing work (including current PM program tasks) and their basis. Rigorous, traditional RCM presupposes that existing work biases analysis. Streamlined RCM processes do not accept this assumption. Foundation-up analysis development constrains analysis performance speed ( see Fig. 7-1).
Failure data used in analysis may be inferred without fieldwork. Organizations tend to justify ("grandfather") their current scheduled maintenance program, although controls can counter this temptation. Managing bias takes planning. Legacy tasks should not be blindly accepted; otherwise, the there is no inherent reason not to consider existing tasks in new RCM program development.
Practically, analysts can conduct their analysis so that hiding legacy tasks origin is virtually impossible. PM tasks must meet and pass applicability and effectiveness (A/E) tests. There can be no PM program task grandfathering in an RCM-based program. Every task must stand on its own failure-prevention merit.
Dominant failure modes (DFM) emerge in plant operations as the failures that matter. They drive reliability and cost. Finding dominant failure modes is an essential part of RCM analysis. Only actual, substantiated failures and their modes should be considered as DFM candidates. Statistics should be used as one tool to identify DFM, but interviews and benchmark comparisons are also required. Engineering design, experience, and an open mind help define DFM for safety, production, and operating costs.
Successful DFM identification takes failure-engineering experience,...
