Preventive maintenance (PM) is usually self imposed downtime (although it may be possible to perform some PM while the product is operating). PM consists of actions intended to prolong the operational life of the equipment and keep the product safe to operate. Ideally, a product will require no servicing or other preventive maintenance and either the probability of failure is remote or redundancy makes failure acceptable (however, one often-required PM task is to verify the operational status of redundant components prior to a mission.). For such an ideal product, only corrective maintenance, if any, would be required. Most often, however, failure is not a remote possibility. Moreover, most products of any complexity require some servicing, even if that only consists of cleaning. Sometimes failures can actually be prevented by preventive maintenance. The goal, then, is to identify only that preventive maintenance that is absolutely necessary and cost-effective. Figure 5 illustrates the two major categories of maintenance, PM and CM, and the tasks associated with each.
Figure 5 -
Major Categories of Maintenance
Reliability Centered Maintenance (RCM) is an analytical method used to identify essential and cost effective preventive maintenance. Preventive maintenance for a product may be scheduled based on condition, a number of events, rounds fired, cycles of operation, seasonal time period, operational profile changes, and sometimes as the result of failures of other equipment. For example, automobile manufacturers provide buyers with a range of recommended mileage-based and time-based preventive maintenance. The initial overall maintenance program should reflect the RCM-based schedule for preventive maintenance. Figure 6 summarizes the steps in an RCM approach to identifying preventive maintenance.
Figure 6 -
The Steps in an RCM Approach to Identifying Preventive
Reference has been made to "on-condition" and "based on condition" in determining the frequency or need for PM. On-condition monitoring, also called predictive maintenance or performance monitoring, is a process whereby one or more parameters are unobtrusively monitored and trended over time. These parameters must have a direct relationship with the "health" of the equipment being monitored. On the basis of some threshold value (determined through analysis or experience) of a parameter or combination of parameters, the equipment will be repaired or replaced prior to any actual failure. This approach to "scheduling" PM can significantly reduce costs, prevent failures, and increase safety. The approach has long been used to "detect" an impending problem in rotating machinery by monitoring vibration.
Improvements in sensors and recording devices and an
understanding of which parameters truly indicate health have increased the
number of applications of condition monitoring. One device developed under Air
Force sponsorship, the Time Stress Measurement Device (TSMD), is one example
of a technological improvement related to on-condition monitoring. The TSMD
collects, records, and stores stress data for subsequent analysis. Although
initially studied as a method of dealing with false alarms (see 184.108.40.206 ), TSMDs can be used with appropriate sensors to record health-related parameters. The recorded levels of the parameters can then be compared with pre-established limits. When the limits have been exceeded or the data indicates that the limits will be exceeded in the near future, an inspection would normally be performed to confirm that a failure is imminent. If the inspection confirms the analysis, then appropriate maintenance can be performed before an actual failure occurs. If the inspection shows that no problem exists, it may be necessary to revise the pre-established limits.
Corrective maintenance includes those activities required to return failed equipment to acceptable operational status. Corrective maintenance consists of recognizing that a failure has occurred (fault detection - FD), determining what has failed (fault isolation - FI), accessing and replacing or repairing the failed component, and confirming that the failure has been corrected. It is the maintainability engineer's responsibility to strike the best balance between preventive and corrective maintenance, keeping in mind such factors as safety, availability requirements, and the customer's operating and support concepts.