A conveyor motor overheats and shuts down. Your team replaces the thermal overload relay, resets the breaker, and the line runs again. Three days later, it trips again. Same motor, same symptoms, same emergency repair.
You fixed the symptom — but the root cause is still there.
That's why Root Cause Analysis (RCA) exists. It's the discipline of asking "why" past the obvious answer until you find the real reason a failure happened — so you can fix it permanently instead of patching the same problem over and over.
What Is Root Cause Analysis?
Root Cause Analysis is a structured problem-solving method used after a failure or incident. Its goal is simple: identify the fundamental reason the failure occurred, then take action to prevent it from happening again.
RCA is not about blame. It's not about finding who made a mistake. It's about finding what in the system — the process, the procedure, the design, the environment — allowed the failure to happen.
In maintenance, RCA answers three questions:
- What happened?
- Why did it happen?
- What must change so it never happens again?
Why Fixing Symptoms Is a Trap
Most maintenance teams operate in a reactive cycle. A machine breaks. You fix the broken part. The machine runs. Repeat.
The problem is that the same failures keep coming back.
| Symptom-Fix Approach | RCA Approach |
|---|---|
| Replace the failed bearing | Find out why the bearing failed |
| Clear the jam | Find out why material keeps jamming |
| Rewind the motor | Find out why the motor burned out |
| Patch the leak | Find out why the pipe is corroding |
Fixing symptoms feels productive. You closed a work order. The line is running again. But you haven't addressed the underlying problem. Every repeat failure costs you: emergency labor, rush parts, lost production time, and frustrated technicians.
A single RCA can eliminate a recurring failure pattern that has been draining your resources for years.
Common RCA Methods
There's no single "right way" to do RCA. Different methods work for different situations. Here are the three most practical for maintenance teams.
5 Whys
The simplest RCA method. Start with the problem, then ask "why" five times — or as many times as it takes to reach a root cause.
Example — motor bearing failure:
- Why did the motor fail? → Bearing seized.
- Why did the bearing seize? → Lubrication starvation.
- Why did lubrication starvation occur? → Grease fitting was clogged.
- Why was the grease fitting clogged? → Old grease had hardened and blocked the passage.
- Why was old grease present? → The fitting had never been cleaned during regreasing — no procedure for it.
Root cause: No standard procedure for cleaning grease fittings before lubrication.
The 5 Whys is fast and requires no tools — just a whiteboard and a team that knows the equipment. But it works best for simple failures with a single root cause.
Fishbone (Ishikawa) Diagram
Also called a cause-and-effect diagram, this method is better for complex failures with multiple contributing factors.
You draw a horizontal arrow pointing to the problem (the "fish spine"), then add angled branches representing categories of possible causes. Common categories in maintenance:
- Man (people): Operator error, insufficient training, fatigue, communication gaps
- Machine (equipment): Design flaw, worn components, incorrect specs
- Method (procedure): Missing or unclear SOPs, shortcut culture, no verification step
- Material (parts/consumables): Wrong lubricant, substandard spare parts, contamination
- Measurement (inspection/data): Incorrect sensor reading, no trend monitoring, uncalibrated tool
- Environment (conditions): Heat, dust, humidity, vibration from nearby equipment
For each category, the team brainstorms possible causes until the diagram is full. Then you analyze which causes are most likely and drill into those.
The Fishbone diagram prevents tunnel vision. Instead of jumping to one explanation, you systematically consider all angles.
Fault Tree Analysis (FTA)
FTA is a top-down, deductive method that uses Boolean logic (AND/OR gates) to map how multiple failures combine to cause a top-level event.
Picture a tree: the failure is at the top. Below it, branches connect through "AND" gates (both conditions must be true) and "OR" gates (either condition triggers it). You work downward until you reach basic events — the root causes.
FTA is more rigorous than 5 Whys and more structured than Fishbone. It's useful for safety-critical failures, complex systems, and situations where you need to calculate the probability of failure paths.
Most factory maintenance teams won't use FTA every day. But for serious incidents — a major fire, a catastrophic equipment failure — it's the gold standard.
The RCA Process: Step by Step
Regardless of the method you choose, every effective RCA follows the same six steps.
Step 1: Define the Problem
Write a clear, specific problem statement. Avoid vague language.
- Bad: "The pump keeps breaking."
- Good: "Centrifugal pump P-102 tripped on thermal overload three times in the past two weeks, each time due to bearing seizure."
Include what, where, when, how often, and the impact (downtime hours, repair cost, production loss).
Step 2: Gather Data
Before you start analyzing causes, collect the facts. Don't rely on memory — review actual records.
- Work order history for the asset
- Operator logs and shift reports
- Condition monitoring data (vibration, temperature, pressure trends)
- Photos of the failed component
- Interviews with technicians and operators who were present
The data stage is where most RCAs fail. If you jump to conclusions without evidence, you'll end up with guesswork disguised as analysis.
Step 3: Identify Possible Causes
Using your chosen method (5 Whys, Fishbone, or FTA), generate a list of all potential causes. At this stage, don't filter. Brainstorm freely. A cause that seems unlikely might turn out to be the real culprit.
Organize causes into categories:
- Direct cause: The immediate physical failure (e.g., bearing seized)
- Contributing causes: Factors that made the direct cause more likely (e.g., grease contaminated with dust)
- Root cause: The fundamental deficiency that allowed contributing causes to exist (e.g., no sealed bearing housing specified for dusty environment)
Step 4: Find the Root Cause
Analyze the causes identified in Step 3 and determine which is the true root cause. Ask:
- If we fix this cause, will the failure stop recurring?
- Is this cause within our control to change?
- Is there evidence supporting this cause, or is it speculation?
Sometimes there are multiple root causes. That's fine — address all of them.
Step 5: Implement Solutions
For each root cause, develop an action plan. Don't just add another inspection. Ask what will permanently eliminate the cause.
| Root Cause | Weak Solution | Strong Solution |
|---|---|---|
| No grease fitting cleaning procedure | Remind technicians to clean fittings | Update PM procedure with a cleaning step |
| Bearing not sealed for dust | Replace bearing more often | Install sealed/shielded bearing housing |
| Operators start pump without checking suction valve | Add a sign | Install interlock that prevents start when valve is closed |
| Training gap on lubrication | Send one person to a course | Write a one-page lubrication standard, train all technicians, verify |
Assign owners and target dates for each action. Track them like any other work.
Step 6: Verify and Close
After the solutions are implemented, wait long enough to see if the failure recurs. Set a verification period — 30 days, 90 days, or one full maintenance cycle.
If the same failure happens again, your RCA wasn't complete. Go back to Step 3 and dig deeper.
If the failure doesn't recur, document the RCA, archive it against the asset, and close it out.
Real-World Example: The Conveyor That Kept Stopping
A food processing plant had a conveyor that stopped unexpectedly twice a week. Each time, maintenance found the drive motor tripped on overload. They reset the breaker, the conveyor ran again, and everyone assumed the motor was "sensitive."
After weeks of repeat failures, the team did an RCA.
Step 1: Define the Problem
"Incline conveyor C-102 trips on motor overload 2–3 times per week during the 2 PM shift. Each trip causes 20–30 minutes of downtime. No visible damage to the motor."
Step 2: Gather Data
- Work orders showed 17 overload trips in 8 weeks
- Vibration data was normal
- Motor temperature was within range at restart
- The 2 PM shift was the highest-throughput shift
- The conveyor had been operating for 6 years without issue
Step 3 & 4: Identify Causes and Find Root Cause
The team used a Fishbone diagram. Under "Method," they noted that the conveyor speed was increased 3 months ago to boost production. Under "Machine," they noted the gearbox oil had not been changed in 4 years.
Further investigation found: when the speed increased, the gearbox ran hotter. The degraded oil lost its viscosity, increasing friction. The higher friction required more torque from the motor, which eventually tripped the overload protection.
Root cause: Gearbox oil degradation caused by higher operating temperature after a speed increase. No one had considered that the oil change interval should be adjusted for the new speed.
Step 5: Implement Solutions
- Changed gearbox oil and restored the recommended change interval based on actual operating temperature
- Added gearbox oil temperature to the weekly condition monitoring checklist
- Created a standard operating procedure: any speed or load change must trigger a review of lubrication intervals
Step 6: Verify
Over the next 6 months, the conveyor had zero overload trips.
Total cost of the RCA: a 2-hour team meeting and an oil change. Total savings: 17 trips × 25 minutes × 26 weeks / 8 weeks = over 27 hours of downtime eliminated, plus the labor and frustration of emergency calls.
How RCA Feeds Into Your Maintenance Program
An RCA is not a one-off investigation. Its real value comes from using the findings to strengthen your entire maintenance system.
Preventive Maintenance Updates
Every root cause you discover should trigger a review of your PM program. If the RCA found that a bearing failed because the greasing interval was too long, update the PM schedule. If it found that a filter was never included in any PM, add it.
Your PM program should be a living document, constantly improved by RCA findings.
Technician Training
RCAs often reveal knowledge gaps. A technician greased the wrong fitting because the equipment drawing was unclear. An operator bypassed a safety interlock because nobody explained why it mattered.
Document these gaps and create targeted training. A 15-minute toolbox talk on a real RCA is more effective than a generic training module.
Spare Parts Strategy
If an RCA identifies a component that fails repeatedly, review your spare parts inventory. Do you stock that part? Is it a critical spare? Should you switch to a more durable version?
Procedure Improvements
Many root causes trace back to missing or unclear procedures. Use RCA findings to write better SOPs, checklists, and operating envelopes.
How a CMMS Tracks RCA Records
Without a system, RCA findings get lost. Someone writes a report, files it in a folder, and three years later nobody remembers the conveyor bearing issue.
A CMMS solves this by tying RCA records directly to assets.
Asset RCA History
In OpexMX, you can attach RCA reports to each asset. When a technician opens a work order for that asset in the future, they can see past RCAs — what was found, what was fixed, and whether the fix held.
Linked Work Orders
When an RCA identifies an action — "replace gearbox oil," "update PM frequency" — that action becomes a work order in the system. No loose ends.
Failure Code Integration
When closing a work order, technicians can tag the failure with a code that links back to the RCA. Over time, the system builds a failure history for each asset. If the same failure mode appears repeatedly, the system flags it — and you know another RCA is needed.
Trending and Reporting
With RCA data in the CMMS, you can answer questions like:
- Which assets have had the most RCAs?
- Which failure modes are recurring across the plant?
- Have our corrective actions actually reduced failure rates?
This turns RCA from a reactive investigation into a proactive improvement tool.
The Bottom Line
Root Cause Analysis is one of the highest-ROI activities a maintenance team can do. It doesn't require expensive software or consultants. It requires discipline: the discipline to stop patching and start asking "why."
Every time you fix the same failure twice, you've paid the cost of an RCA without doing it. An emergency work order costs 3–5 times more than planned maintenance. A single RCA that eliminates one recurring failure pattern pays for itself many times over.
Stop fixing symptoms. Find the root cause. Fix it permanently.
See how OpexMX helps you track RCAs, link them to assets, and turn findings into lasting improvements — built for maintenance teams on the factory floor.