Reliability-Centered Maintenance was developed for commercial aviation in the late 1960s, codified by United Airlines engineer F. Stanley Nowlan and Howard Heap in a landmark 1978 study. The finding that changed everything: most equipment failures are not prevented by time-based maintenance. They occur randomly regardless of age, and scheduled overhauls often introduce new failure modes rather than preventing existing ones.
RCM asks a different question than traditional maintenance. Not 'how often should we maintain this equipment?' but rather 'what are the functions of this equipment, what failures threaten those functions, and what is the most appropriate way to manage each failure mode?'
That question leads to a maintenance strategy that is sometimes more rigorous than your current PM program—and sometimes much less. Both outcomes are correct applications of RCM.
The Seven Questions RCM Asks About Every Asset
RCM analysis follows a disciplined seven-question framework applied systematically to each asset or functional group of assets.
Question 1: What are the functions of this asset in its current operating context? Not what it was designed to do—what it actually needs to do in your specific operation at your required performance standard. A pump that must deliver 300 gallons per minute to maintain process temperature has a specific function with a measurable failure threshold.
Question 2: In what ways can this asset fail to fulfill its functions? These are functional failures—the points at which the asset can no longer deliver its required function. For the pump, functional failure could be complete loss of flow or degraded flow below the required threshold.
Questions 3 and 4 identify failure modes (what causes each functional failure) and failure effects (what happens when each failure mode occurs). This is where the analysis becomes specific and valuable: different failure modes for the same asset may have very different effects and therefore warrant very different maintenance strategies.
Questions 5 through 7 address failure consequences, task selection, and defaults. Is this failure hidden or evident? Does it affect safety, environment, or operations? Is there a proactive maintenance task technically feasible and worth doing? If no proactive task is appropriate, what's the default action—redesign, run to failure, or change of operating procedures?
When RCM Is Worth the Investment—And When It Isn't
Full RCM analysis is rigorous and time-consuming. A complete analysis of one complex asset system—a compressor train or a heat exchanger network—can take 40-80 hours of skilled analyst time. Applied to an entire facility, RCM is a multi-year project.
RCM is worth this investment when: you're designing maintenance programs for new, complex, or safety-critical systems where no maintenance history exists; when your current maintenance program is demonstrably ineffective (high failure rate despite PM compliance); or when the cost of failure is catastrophic (aviation, nuclear, oil and gas).
For most industrial maintenance operations, Streamlined RCM or FMEA (Failure Mode and Effects Analysis) delivers 80% of the benefit at 20% of the effort. These approaches use the same seven questions but apply them at a higher level of abstraction and focus analytical effort on critical assets where the consequences of failure justify the investment.
How to Run an RCM Analysis in Practice
An RCM analysis requires a cross-functional team: a maintenance engineer who knows the equipment, an operator who uses it daily, a process engineer who understands the system context, and a facilitator who keeps the analysis structured.
Start by defining the system boundary. The compressor is not analyzed in isolation—it's analyzed as the compression system including its suction filters, discharge valves, cooling system, and control logic. Failure modes that seem unimportant for the compressor alone may be critical when their effect on the overall system is understood.
Document every function, functional failure, failure mode, and failure effect in a Failure Mode and Effects Analysis (FMEA) table. Software tools make this manageable—there are dedicated RCM software packages (available from specialized vendors) and some CMMS platforms with built-in FMEA modules. Spreadsheet-based FMEA is acceptable for initial analysis on smaller systems.
For each failure mode, select the maintenance strategy using the RCM decision logic: Is there a condition monitoring task that can detect the failure mode before function loss? If yes, use it. If not, is there a scheduled restoration or replacement task that reduces risk? If yes and cost-effective, use it. If no proactive task is appropriate, default to run-to-failure with appropriate consequence mitigation.
RCM in CMMS: Translating Analysis Into Action
An RCM analysis that stays in a document produces nothing. The value is in implementation: creating the specific maintenance tasks, intervals, and monitoring routines that the analysis defines, and tracking execution through your CMMS.
Each RCM-defined maintenance task becomes a PM template in your CMMS with the specific task description, interval, skill requirement, and parts list derived from the analysis. Condition monitoring tasks become recurring routes or sensor triggers. Run-to-failure decisions get documented as a deliberate strategy choice with appropriate spare parts stocked.
Review RCM decisions annually for critical assets. Failure mode patterns change as equipment ages, operating conditions shift, and new information accumulates. An RCM analysis that was correct when the asset was new may need revision after 5 years of actual failure history.
Streamlined RCM: Getting the Benefits Without the Full Commitment
For operations that can't commit to full RCM analysis, a pragmatic approach captures most of the value. Start with your top 10 critical assets and run a simplified FMEA covering the top 3-5 failure modes for each. This typically takes a half-day workshop per asset system with the right team.
For each identified failure mode, ask two questions: Is this failure mode detectable before function loss? What's the consequence if we don't detect it? The answers determine the maintenance strategy without the full seven-question RCM process.
Even this simplified analysis consistently produces valuable insights: maintenance tasks that can be eliminated (because they address failure modes that don't occur in your operating context), new tasks that should be added (because the analysis reveals failure modes not currently addressed), and condition monitoring opportunities that justify the sensor cost.