A fishbone (Ishikawa) diagram lays out every plausible cause of one problem under category headings — usually people, machine, method, material, measurement and environment — so a team can see the whole field before picking causes to test. Use this template to run the session and to record the evidence that confirms or rules out each cause.
Problem statement
Problem / effect, Part or process, Where found, Quantity affected, Requirement not met, Date, Team members, Linked CAPA / 8D number
Causes by category
Category, Possible cause, Sub-cause (why?), Evidence / how checked, Verified? (Y/N/Partly)
Verified causes and next steps
Verified cause, Next step, Owner, Due date
Example Example Manufacturing Co., a fictional manufacturer, finds burrs on drilled holes in a steel bracket at final inspection. All figures are illustrative.
| What | Burr at exit side of Ø6 mm hole, bracket BRK-110 |
|---|---|
| Where found | Final inspection, visual check |
| How many | 12 of 200 brackets in lot 0457 |
| Requirement | Drawing note: all edges free of burrs |
| Category | Possible cause | Evidence / how checked | Verified? |
|---|---|---|---|
| Machine | Drill worn past tool-life limit | Tool-change log vs part sequence | Yes |
| Machine | Spindle runout | Indicator check on spindle | No — within limit |
| Method | Deburr step skipped on second shift | Observed both shifts; step done on both | No |
| Material | Harder steel than specified | Mill cert and hardness check of lot | No — matches spec |
| Measurement | Burr criterion unclear to inspectors | Asked 3 inspectors to judge sample parts | Partly — added a limit sample |
| People | New operator on drill press | Training record and observation | No |
| Environment | Coolant concentration low | Refractometer reading | No — within range |
| Field | What to enter |
|---|---|
| Problem statement | What, where, when found, how many, and the requirement not met. One specific effect per diagram; “quality issues” is too broad. |
| Team | Names and functions of people in the session. Include someone who does the work every day. |
| Categories | The 6M headings, or headings that fit your process. Service and office processes often use policies, procedures, people, place. |
| Possible cause | A specific condition that could produce the effect. Write causes, not solutions ("drill worn", not "replace drill"). |
| Sub-cause | A branch that explains why the cause could exist. Ask "why could this happen?" once or twice per bone. |
| Evidence / how checked | The data, observation or test used to confirm or rule out the cause. This column turns a brainstorm into an investigation. |
| Verified? | Yes, No or Partly. Only verified causes move on to corrective action. |
| Next steps | Deeper analysis (5 Why), owner and date. Link the CAPA or 8D number so the record is traceable. |
Man (people), machine, method, material, measurement and mother nature (environment). They are a prompt, not a rule — adapt them to your process.
It is a structured way to list possible causes. It becomes a root cause analysis when each likely cause is tested with data and the verified cause is explained, often with a 5 Why.
Use both. The fishbone widens the search across categories; the 5 Why drills down on a verified cause to find why it was able to happen.
It is named after Kaoru Ishikawa, the Japanese quality engineer who popularized it, which is why it is also called an Ishikawa or cause-and-effect diagram.
No standard named here mandates a specific tool. They require that causes be determined; the fishbone is a common, accepted way to show how you did it.
COPLAIN AI PLATFORM
When the verified cause is an unclear procedure, Coplain’s Procedure Simplifier rewrites it into short, shop-floor-ready steps for you to review.
Generate yours automatically with Coplain →Related Tools & Templates