Overview
Root Cause Analysis (RCA) helps you systematically investigate failures to identify underlying causes, not just symptoms. By understanding why failures occur, you can implement effective corrective actions and prevent recurrence.MicroCBM’s RCA module uses browser localStorage for data persistence. Records are stored locally and not synced to the backend API.
RCA Methodology Options
Choose the investigation method that best fits your failure type:5 Whys
Linear cause chainStart with a problem and ask “Why?” repeatedly to drill down to root causes. Best for simple failures with clear cause-effect relationships.
Logic Tree
Hierarchical hypothesis testingMap potential causes in a tree structure. Test each hypothesis with evidence (Confirmed/Rejected/Pending). Ideal for complex failures with multiple contributing factors.
Fishbone (Ishikawa)
6M categorical analysisOrganize causes by category: Man, Machine, Method, Material, Measurement, Environment. Useful for process-oriented investigations.
Creating an RCA
Enter RCA Header
Fill in the investigation details:Required Fields:
- Title: Brief summary of the incident (e.g., “Gearbox failure - Unit 3 compressor”)
- Investigation Method: Select 5 Whys, Logic Tree, or Fishbone
- Asset: The failed equipment
- Department: Responsible department
- Event Date: When the failure occurred
- RCA Leader: Person leading the investigation
- Severity Level: Low, Medium, High, or Critical
- Production Impact (hours): Downtime caused by failure
- Estimated Cost Impact: Financial loss or repair cost
- Map Location: Physical location or address
- Groups: Category (Operations, Safety, Quality, Other)
- Notes: Additional context
- Types: Incident, Near Miss, Audit, or Other
- Tags: Keywords for searching
An RCA ID is automatically assigned in the format
RCA-YYYY-XXXX (e.g., RCA-2026-0001).RCA Workflow Tabs
Once created, each RCA uses a tabbed interface to guide you through the investigation:1. Evidence
Gather all relevant information before analyzing:-
Evidence Items: Document facts, observations, and data
- Photos of failed components
- Oil sample results showing abnormal trends
- Maintenance logs and operating conditions
- Witness statements from operators
- Sensor data or alarm logs
- Attachments: Upload supporting files
2. Problem Statement
Define the problem clearly and completely:- Focal Point: What failed? (e.g., “Gearbox bearing seizure”)
- Timing: When did it happen? Start/end dates and times
- Location: Where did it occur? Map location, business unit
- Impact: What were the consequences?
- Actual Impact: Production loss, safety incident, environmental release
- Potential Impact: What could have happened
- Costs: Investigation, repair, lost production
- Frequency: How often does this problem occur?
A well-defined problem statement focuses the investigation and prevents scope creep.
3. Analysis
This is where you build your cause-and-effect investigation using your chosen methodology:5 Whys Analysis
Continue to Root Cause
Keep asking why until you reach a cause that, if corrected, prevents recurrence.Typically 3-7 levels deep. Stop when:
- You’ve identified a controllable root cause
- Further “whys” lead to factors outside your control
- Multiple whys from different branches converge
Logic Tree Analysis
Branch Hypotheses
Add child nodes for possible causes:
- “Contamination”
- “Lubrication failure”
- “Overload”
- “Manufacturing defect”
Test Each Hypothesis
For each hypothesis:
- Add evidence (oil sample data, visual inspection, load records)
- Mark status: Confirmed, Rejected, or Pending
- Add supporting evidence notes
Drill Down Confirmed Causes
For confirmed causes, add sub-hypotheses and continue testing until root cause is identified.
Fishbone Analysis
Organize causes by the 6M categories:| Category | Examples |
|---|---|
| Man | Operator error, training gaps, fatigue, communication failure |
| Machine | Wear, design flaw, inadequate maintenance, capacity exceedance |
| Method | Incorrect procedure, missing work instruction, poor process design |
| Material | Wrong lubricant, contaminated parts, substandard materials |
| Measurement | Faulty sensors, calibration error, inadequate monitoring |
| Environment | Temperature extremes, humidity, dust, vibration |
Identify Causes per Category
For each category, add specific causes that contributed to the problem.
4. Corrective Actions
Document actions to prevent recurrence:- Action Type: Corrective, Preventive, or Systemic
- Description: What will be done
- Owner: Person responsible for implementation
- Department: Responsible team
- Due Date: Target completion
- Priority: Low, Medium, High, or Critical
- Status: Open, In Progress, Completed, Verified, or Overdue
- Verification Required: Whether follow-up validation is needed
- Effectiveness Review Date: When to check if action worked
5. Final Report
Summarize the investigation:- Executive Summary: High-level overview for management
- What happened
- What was the impact
- What were the root causes
- What actions are being taken
- Cause and Effect Summary: Detailed technical explanation
- Investigation timeline
- Evidence findings
- Root cause analysis results
- Corrective actions with justification
The final report is often shared with stakeholders who weren’t involved in the investigation. Write for clarity and completeness.
RCA Status Workflow
RCAs progress through defined statuses:- Draft: Initial creation, evidence gathering
- Investigation: Active analysis and hypothesis testing
- Review: RCA complete, awaiting approval
- Actions Open: Corrective actions in progress
- Verification: Actions completed, checking effectiveness
- Closed: Investigation and actions complete
Viewing and Managing RCAs
The RCA list shows:- RCA ID: Auto-generated identifier
- Title: Investigation summary
- Status: Current workflow stage
- Updated: Last modification date
- Actions: Open (view/edit) or Delete
Deleting RCAs
To delete:- Click Delete next to the RCA in the list
- Confirm deletion in the browser dialog
Best Practices
Investigation Principles
- Assemble a Team: Include operators, maintenance, engineering, and management
- Define Scope Early: Clear problem statement prevents mission creep
- Gather Evidence First: Don’t jump to conclusions before collecting facts
- Test Hypotheses: Don’t assume causes — prove them with evidence
- Look for Systemic Issues: Individual failures often reveal process weaknesses
- Focus on Prevention: Root cause fixes prevent recurrence, not just repair
Common Pitfalls to Avoid
- Blame Culture: Focus on system failures, not individual fault
- Confirmation Bias: Test hypotheses objectively, don’t cherry-pick evidence
- Fixing Symptoms: Replacing a failed component doesn’t prevent future failures
- Incomplete Actions: Corrective actions must be specific, measurable, and assigned
When to Use Each Method
| Method | Best For | Example |
|---|---|---|
| 5 Whys | Simple, linear failures | Single-component failure with clear cause chain |
| Logic Tree | Complex failures, multiple causes | Intermittent failures, process upsets |
| Fishbone | Process problems, systemic issues | Quality defects, recurring incidents |
Example: Complete 5 Whys RCA
Problem: Hydraulic pump bearing failed on Pump 023 5 Whys Chain:- Why did bearing fail? → Oil contamination exceeded limits
- Why was oil contaminated? → Filter was saturated with particles
- Why was filter saturated? → Filter change interval too long
- Why was interval too long? → Scheduled based on calendar, not condition
- Why not condition-based? → No monitoring system for filter differential pressure
- Install differential pressure gauge on filter (Owner: Maintenance Lead, Due: 2 weeks)
- Establish condition-based filter replacement criteria (Owner: Reliability Engineer, Due: 1 week)
- Train technicians on new procedure (Owner: Training Coordinator, Due: 3 weeks)
Related Features
- Recommendations: Link RCA corrective actions to tracked recommendations
- Alarm Monitoring: Use alarm trends as RCA evidence
- Oil Sample Analysis: Sample history provides critical evidence
- Asset Management: Track RCA history by asset