A Guide to Reliability-Centered Maintenance (RCM)

Walk most plant floors, and you'll find maintenance running on a calendar: Every asset is serviced on a fixed schedule, whether it needs it or not. The trouble is the schedule rarely matches how or when the equipment actually fails. That gap is expensive: Unplanned downtime costs manufacturers between $36,000 and $1.4 million an hour, reports Siemens. 

Reliability-centered maintenance (RCM) starts with the failure instead of the calendar. Before scheduling anything, RCM asks how a given asset breaks, what that failure costs, and whether there's a task worth doing to get ahead of it. Sometimes the answer is a monthly check on a sensor; sometimes it's to run the asset until it fails. 

What Is Reliability-Centered Maintenance?

RCM is a structured process for determining the most effective maintenance strategy for each asset based on its functions, the ways it can fail, and the consequences of those failures. The goal is to keep equipment reliable enough to do its job at the lowest sensible cost, rather than to prevent every failure at any price.

The core insight is that not all assets deserve the same treatment. A backup light that fails once a year with no real consequence doesn't warrant a maintenance schedule; you replace it when it dies. A pump whose failure stops the line and risks a safety incident warrants close monitoring. RCM gives you a repeatable way to distinguish and match each case to the right level of effort.

That's why RCM isn't a single tactic. For each way an asset can fail, it points you toward one of a few outcomes — from doing nothing until it breaks, to predictive monitoring, to redesigning the problem to fix it permanently. The maintenance strategy follows the failure instead of the calendar.

{{callout1}}

A Brief History of RCM

RCM was born in commercial aviation. In 1978, United Airlines engineers F. Stanley Nowlan and Howard F. Heap published a landmark report commissioned by the US Department of Defense that overturned a long-held assumption about how equipment fails.

The prevailing belief was that most components wear out on a predictable schedule, so overhauling them at fixed intervals would prevent failures. Nowlan and Heap's data said otherwise: Only about 11% of failures were tied to age, while the other 89% struck randomly.  Worse: The intrusive overhauls meant to prevent failure often introduced it.

If most failures aren't age-related, scheduled teardowns are frequently wasted effort — and sometimes counterproductive. The methodology Nowlan and Heap built around that evidence became RCM, later codified in the SAE JA1011 standard that still defines an RCM process.

RCM vs. Traditional Maintenance

Most plants run on some mix of two older approaches. Reactive maintenance fixes things after they break. It’s cheap until the breakdown is expensive. Preventive maintenance (PM) services assets on a fixed time or usage schedule. It's safer, but it spends the same effort on a critical pump and a conveyor that rarely fails.

RCM doesn't replace those approaches or newer ones like total productive maintenance. Instead, it decides which approach is best for each asset or failure. Traditional maintenance starts with a breakdown or prevention schedule: How often should we service this? RCM starts with the failure: How does this asset fail, does that failure even matter, and is there a task that prevents it?

The practical effect is less wasted work. Instead of servicing everything on the same cycle, you stop over-maintaining low-consequence assets and redirect resources toward the equipment that can actually hurt you when it fails.

The RCM Process

RCM is a decision process you run on one asset — or failure mode — at a time. The SAE JA1011 standard frames it as a sequence of seven questions, but in practice, it comes down to five working steps:

1. Define the asset's functions and standards: Spell out what the equipment is supposed to do and the performance it should hit — throughput, output quality, equipment availability, safety, and compliance. You can't judge a failure until you've defined "working."
2. Identify how it can fail: List the functional failures (the ways the asset stops meeting its standards) and the failure modes behind each one. A failure mode and effects analysis (FMEA) is the usual tool for this step.
3. Trace the effects and consequences: For each failure mode, document what happens and, more importantly, whether it matters. Sort consequences into safety and environmental, operational, and cost. Use a criticality assessment to separate the failures worth preventing from the ones you can live with.
4. Assign a maintenance strategy: Based on the consequence and whether the failure can be predicted, choose the right response: run-to-failure, preventive maintenance, condition-based/predictive maintenance (PdM), or redesign. 
5. Turn it into a plan and review it: Convert each decision into a work plan with steps, parts, and intervals, then track the results and adjust. RCM assumes the first version is a starting point, not the final answer.

Applied across your critical assets, those steps produce something a generic PM schedule can't: a maintenance program where every task exists for a documented reason.

{{callout2}}

Pros and Cons of RCM

RCM earns its reputation, but it isn't free, and it isn't right for every asset. Knowing the trade-offs keeps expectations honest.

Pros

• Less wasted maintenance: You stop servicing assets that don't need it, freeing labor and parts for the work that matters.
• Fewer surprise failures: Condition-based and predictive tasks catch warnings before they become downtime.
• Effort aimed at consequences: Criticality analysis concentrates your attention on the assets that threaten safety, compliance, or production.
• Documented decisions: Every task traces back to a specific failure mode, which makes the program auditable and easy to defend.
• Longer asset life: Dropping intrusive overhauls that introduce defects keeps equipment healthier for longer.

Cons

• It's analysis-heavy: A full RCM study takes time, cross-functional input, and reliable failure data that smaller teams may not have on hand.
• It exposes problems; it doesn't fix them: RCM names the right task, but you still need trained people, parts, and discipline to do the work.
• It can stall on paper: Without a system to turn analysis into scheduled, tracked work, the findings gather dust, and nothing changes.
• It can become outdated: Failure patterns shift as equipment and processes change, so an RCM program that isn't reviewed regularly drifts out of date.

RCM is one of the strongest ways to spend a maintenance budget well, but it rewards only the teams that act on the analysis.

How To Run an RCM Program

You shouldn't start an RCM program on a whole plant at once. Pick a handful of assets, prove the approach works, and grow from there. A workable sequence:

1. Start where failure hurts most: Rank your equipment by what a breakdown costs(e.g., lost production, a safety or compliance exposure, a long wait on a replacement part) and begin at the top of that list. Teams that try to tackle every asset at once tend to burn out before they finish the first one.
2. Get the right people in the room: An RCM analysis is only as good as the people doing it. Pull in operators, maintenance techs, and an engineer. The ones who run and repair the equipment know failure modes that never make it into a manual.
3. Run the five steps on each asset: Define what each asset does, map how it fails, weigh the consequences, and assign it a strategy.
4. Write the plan down: Each strategy becomes a concrete task with a schedule, parts, tools, and a clear standard for "done."
5. Put it in a system: Load the job plans into your computerized maintenance management system (CMMS) or shop floor platform. This tracks that work gets scheduled, signed off, and recorded, and creates a failure history that feeds the next review.
6. Revisit it on a schedule: Quarterly review is a reasonable start. Adjust intervals and tasks based on what happened (run a quick root-cause analysis on anything that fails twice). Add new equipment to the plan as it arrives.

Avoid treating RCM as a project with an end date. The first analysis earns its keep, but the real payback comes from working that review loop, year after year.

The Bottom Line

Reliability-centered maintenance replaces one-size-fits-all upkeep with a simple discipline: Understand how each asset fails, weigh the cost of that failure, and allocate maintenance efforts accordingly. Done right, it saves you from servicing machines that are fine and frees that time and money for serious failure risks. But it only pays off if the analysis makes it to the floor and gets used.

See how Redzone's connected workforce platform gives frontline teams real-time visibility to turn reliability strategy into daily action.