PM, PdM, Proactive Maintenance

Here are my definitions of these terms:

Preventive Maintenance: A general category covering maintenance tasks (called policies) that seek to anticipate and deal with failures that could likely occur in the future. Preventive Maintenance is a super-set of strategies that include Time Based Maintenance (TBM) and Predictive Maintenance (PdM).

Time (Age) Based Maintenance (PM): A maintenance policy  intended to reduce the number of failures that are expected to occur. The maintainer performs a scheduled repair or replacement of components at fixed age or calendar intervals. The term “Preventive Maintenance” is often used to mean Time Based Maintenance (TBM) as opposed to Condition Based Maintenance. The challenge in TBM is to choose task intervals (in appropriate units e.g. hours of operation, widgets produced, energy consumed, etc) that are neither too short not too long. If the intervals are too short they would result in unnecessary interventions thereby reducing the asset’s availability and increasing overall maintenance cost. Intervals that are too long would result in too many failures occurring before the moment of TBM. The course “Achieving Reliability from Data” presents the methodology for deciding upon the appropriate (called “optimal”) PM interval.

Predictive Maintenance (PdM): Also called condition based maintenance (CBM), is designed to preempt functional failure by monitoring condition indicators and intervening at the propitious moment,  during the degradation process before the full ramifications of failure would be incurred by the organization.  TBM accepts that a certain number of failures will occur. CBM is more conservative because it is less tolerant of a functional failure, one having severe consequences. CBM aims to detect failures while they are still in their formative or “potential failure” stages. The challenge in PdM/CBM is to select a “condition indicator” that accurately tracks the failure degradation process. The course “Achieving Reliability from Data” presents a three part methodology:

1. Acquiring the relevant data needed for building the CBM decision rule.
2. Selecting the influential CBM condition indicator(s).
3. Monitoring and continuously improving CBM performance.

RCM: A process to determine which maintenance policies, TBM or PdM, will mitigate satisfactorily the consequences of a given failure mode. If neither of these policies meet the organization’s requirements, RCM allows a policy of no scheduled maintenance  as long as the consequences of failure do not involve health, safety, or environmental risk. Otherwise a one-time redesign action is mandatory.

Here’s a short video that helps to clarify these different maintenance concepts. The Basic Problem In Physical Asset Management