What Is Root cause analysis in the 8D Process? #
The D4 Step – Root Cause Analysis is the heart of the 8D problem-solving process.
It helps to identify, verify, and document the correct root causes that lead to the problem. And it is for both, root cause for occurrence and root cause for non-detection (escape) of failure.
“You can’t fix what you don’t understand.”
Up to this stage, containment action (D3) has protected the customer, and now the focus shifts to why the defect happened in the first place.
Therefore, the Objective of D4 step is to:
- Identify all possible causes using brainstorming.
- Separate possible causes from verified root causes using evidence.
- Find out Occurrence Cause (why the defect happened) and Non-Detection Cause (why it wasn’t caught earlier).
- Use analytical tools such as 5 Whys and Fishbone (Cause & Effect) Diagram.
- Validated and document root causes in the 8D template for traceability.
How to update Root Cause Analysis Sections from 8D Analysis Template (Explained) #
In 8D analysis, the D4 step is structured with three sub-steps:
1. Identify Potential Causes using the Fishbone (Cause & Effect Diagram),
2. Verify those potential causes through a Cause Validation Table,
3. Perform Why–Why Analysis (5 Whys) only for the validated causes, to find the true root cause for both Occurrence and Non-Detection of failures.
Let’s go through each of these steps clearly using our problem:
“Weld spatter and weak joint on LH seat frame bracket during MIG welding.”
Step 1: Identify Potential Causes (Fishbone Diagram) #
To brainstorm the potential causes, we generally use the cause-and-effect diagram. This method use to categorized the potential causes under 6M categories:
- Man
- Machine
- Material
- Method
- Measurement
- Environment
Fishbone Analysis for Seat Frame Weld Defect
Effect (Problem):
Weak weld joint and spatter on LH seat frame bracket during MIG welding.
| Category | Possible Causes |
| Man | – Operator fatigue during night shift – Inconsistent torch angle – Lack of refresher training on parameter control |
| Machine | – Gas regulator malfunction – Clogged nozzle – Worn O-ring in gas line – Faulty welding timer |
| Material | – Welding wire moisture – Inconsistent wire feed speed – Contaminated base material (rust or oil) |
| Method | – No pre-cleaning step before welding – Wrong welding parameter setup – Missing PM step for gas regulator |
| Measurement | – No gas flow monitoring gauge – Lack of weld penetration check fixture |
| Environment | – High humidity – Poor lighting around welding area – Air draft affecting shielding gas flow |
Key Points in brainstorming:
- In this brainstorming stage, you don’t decide correct or incorrect cause, you collect ideas.
- Involve Quality, Production, Maintenance, and Welding Engineers together.
Step 2: Validate Potential Causes (Cause Validation Table) #
We validate the above causes to confirm which of potential causes are actually valid, to create the same observed defect.
Cause Validation Table
| Cause Type | Potential Cause | Specification / Standard | Observation / Evidence | Valid / Not Valid |
| Machine | Gas regulator malfunction | Gas flow 10–12 LPM | Observed fluctuation between 6–12 LPM during trial | ✅ Valid |
| Machine | Clogged nozzle | Nozzle clean every shift | Found clean during audit | ❌ Not Valid |
| Material | Wire moisture | Wire stored below 60% RH | Storage room humidity 50% | ❌ Not Valid |
| Method | Missing PM step for gas regulator | PM plan – weekly equipment check | Gas regulator not included in PM sheet | ✅ Valid |
| Environment | Poor lighting at inspection | Lux ≥ 800 | Actual 300 Lux measured | ✅ Valid |
| Man | Operator fatigue | Shift rotation standard | Operator worked 6 consecutive nights | ✅ Possible but not primary |
| Measurement | No gas flow gauge | Specified visual check only | No flowmeter installed | ✅ Valid |
Validated Causes:
From this analysis, the main valid causes are:
- Gas regulator malfunction / unstable gas flow
- Missing PM step for regulator (method-related)
- Poor lighting in inspection area
- No gas flow gauge installed (measurement-related)
These validated causes will now move to Why–Why Analysis to find root causes.
Step 3: Why–Why Analysis (5 Whys) #
We do why-why analysis to dig deep into each validated cause until you reach the real, controllable, root cause.
Because, we always take a corrective action on root cause of problems.
Now, you need to perform two separate 5-Why analyses:
- One for Occurrence Cause (why the defect happened)
- One for Non-Detection Cause (why it was not caught earlier)
A. Occurrence Cause Analysis #
| Why Level | Question | Answer |
| Why-1 | Why did the weld become weak and spattered? | Because the gas flow during welding was unstable. |
| Why-2 | Why was gas flow unstable? | Because the regulator connection was loose and O-ring damaged. |
| Why-3 | Why was the O-ring damaged? | Because it was worn out and not replaced for a long time. |
| Why-4 | Why wasn’t the O-ring replaced in time? | Because the Preventive Maintenance checklist didn’t include regulator inspection. |
| Why-5 | Why was the PM checklist incomplete? | Because PM schedule was prepared only for welding machine, not its accessories. |
Root Cause of Occurrence:
“Preventive Maintenance schedule did not include gas regulator inspection; worn O-ring caused inconsistent gas flow leading to weak weld and spatter.”
B. Non-Detection Cause Analysis #
| Why Level | Question | Answer |
| Why-1 | Why was the weak weld not detected in visual inspection? | Because the lighting in the inspection area was poor. |
| Why-2 | Why was the lighting poor? | One inspection lamp was damaged and not replaced. |
| Why-3 | Why wasn’t the lamp replaced? | Maintenance request was pending for more than a week. |
| Why-4 | Why was the maintenance request delayed? | There was no escalation process for critical quality equipment. |
| Why-5 | Why no escalation process existed? | Because lighting was not classified as a quality-critical control in the maintenance system. |
Root Cause of Non-Detection:
“Inspection lighting not defined as quality-critical equipment; no escalation for maintenance backlog, leading to poor visibility and missed defect detection.”
Key Concepts: Occurrence vs. Non-Detection Causes #
| Type | Definition | Example |
| Occurrence Cause | The direct reason why the defect was produced. | Worn gas regulator O-ring. |
| Non-Detection Cause | The reason why the defect was not detected earlier. | Poor lighting in inspection area. |
Both must be identified and corrected; fixing only the occurrence cause does not ensure prevention.
Tools Used in root cause analysis #
| Tool | Purpose |
| 5 Whys | To drill down to the real cause. |
| Fishbone (Ishikawa) Diagram | To categorize all potential causes (Man, Machine, Material, Method, Measurement, Environment). |
| Cause Validation Table | To prove cause-effect relationship. |
| Process Flow Review | To identify failure points in production sequence. |
| FMEA Cross-Check | To verify if the cause was already identified as a potential failure mode. |
How D4 Links to next steps
Once you find out the root cause, the next steps are:
- D5 – Corrective action = Root causes identified now guide selection of corrective actions.
- D4 to FMEA & Control Plan = Causes found here must be reflected as lessons learned in D7.
Common Mistakes in Root cause Analysis #
- Jumping to conclusions: verifying the first cause that looks logical.
- No evidence: failing to validate cause with data.
- Confusing symptom with cause: “low gas flow” is a symptom, not a root cause.
- Skipping non-detection cause: only fixing occurrence leads to repeat failures.
- Not updating FMEA: missing linkage weaks your prevention.
Summary of root cause analysis #
- Start wide (Fishbone Diagram), then narrow down (Cause Validation), then deep dive (5 Whys).
- Always separate Occurrence and Non-Detection causes.
- Support every conclusion with data or physical evidence.
- Don’t stop at surface-level reasons, ask why until you find a system-level cause.
- Verified root causes are the foundation for next step D5 (Corrective Actions).
“Guessing is not problem-solving. Evidence is.”
