Problem Solving & Quality · Root Causes

Ishikawa Diagram

Place the effect at the head of a fishbone and list possible causes along families — method, machine, material, manpower, measurement, environment.

Time1 h
FormatTeam
StageRoot Causes

Ishikawa Diagram: what it is and why it works

An Ishikawa diagram, or fishbone, organizes the possible causes of an effect into families. The deviation statement goes in the head box; the main bones carry cause categories, commonly Method, Machine, Material, Manpower, Measurement and Environment; and team members add possible causes on each bone, extending them into sub-branches by asking why. Each cause is then marked as verified, refuted or to be checked, using facts already collected, and the two or three most likely causes are selected for verification.

The diagram's main contribution is breadth. Teams under pressure tend to lock onto the first plausible cause, often the one closest to their own expertise; the fixed categories push them to consider materials, measurement and environmental factors they would otherwise skip. It is a structured brainstorming tool, not an analysis: it generates hypotheses but proves nothing. Its value depends on what comes before and after it, namely a precise deviation statement and a fact–opinion sort before, and 5 Whys and on/off cause verification after. For problems where several conditions must coincide, a fault tree shows the logic between causes better; the fishbone is stronger for open-ended quality problems with many possible contributors.

What you need

A verified deviation statement
Facts collected so far: data, observations, Is / Is-Not table
A cross-functional team that includes people who run the process
A large drawing surface or digital whiteboard

What you get

A fishbone diagram with causes grouped by family
A status for each cause: verified, refuted or to be checked
Two or three prioritized causes with verification plans
A record of ideas considered, useful if the first candidates are refuted

When to use it

When the team jumps to the first plausible cause and ignores the others.

How to do it, step by step

Write the deviation statement in the head box on the right.
Draw the main bones: Method, Machine, Material, Manpower, Measurement, Environment — adapt them if another set fits better.
Brainstorm possible causes on each bone, then ask “why?” to add sub-branches.
Mark each cause as verified, refuted or to be checked, using the facts gathered earlier.
Select the two or three most likely causes and plan how to verify each one.

Worked example: Short shots on an injection-molded connector housing

Illustrative scenario — figures are realistic but not from a real company.

An automotive electronics supplier molds nylon connector housings in an eight-cavity tool. Short shots (incompletely filled parts) run at 2.8% against a 0.5% target. Maintenance believes the barrel heater bands are failing; production blames the resin.

Head box: 'Connector housing short shots at cavities 7 and 8, 2.8% vs. 0.5% target, since week 14, per the vision-system reject log.'
The team kept the six standard bones. Examples: Machine (heater bands, check-ring wear, hot-runner zone control); Material (resin moisture, regrind ratio, lot viscosity); Method (dryer residence time, injection profile); Measurement (vision-system sensitivity); Environment (ambient humidity); Manpower (start-up practice across shifts).
Asking why on 'resin moisture' added sub-branches: dryer dew point, hopper residence time, and material left in an open box over the weekend.
Earlier facts refuted several branches: the heater bands tested within specification, and the same resin lot ran on another press without short shots. Hot-runner control for cavities 7 and 8 was marked 'to be checked' because the defect appeared only there.
Selected for verification: hot-runner zone control for cavities 7 and 8, and blocked vents at those cavities.

Result. A reference-thermocouple check showed the zone 7–8 thermocouple reading about 25 °F high, so the zone ran colder than displayed. Replacing it and cleaning the vents brought short shots down to 0.4%. The team kept the diagram on file, and its moisture branch later helped solve a brittleness complaint on another part.

Open it in the FieldKit Problem Solving app → Get the Problem Solving template pack — join the waitlist

Common pitfalls and how to avoid them

Treating the fishbone as the conclusion.Mark every cause as verified, refuted or to be checked, and verify the top candidates before acting.
Vague causes such as 'operator error' or 'bad material'.Write specific, testable causes, for example 'resin moisture above the drying specification' or 'mold not at temperature at start-up'.
Forcing causes onto all six bones.Adapt the categories to the process; for services or maintenance other families may fit better, and an empty bone is acceptable.
Brainstorming without facts in the room.Bring the deviation statement, the data and the Is / Is-Not table to the session so ideas can be filtered as they come up.

Frequently asked questions

What are the 6Ms of an Ishikawa diagram?

Method, Machine, Material, Manpower (often now called People), Measurement and Mother Nature, meaning the environment. They are common cause categories for manufacturing problems. Some teams add Management or Money, and service processes often use other families such as policies, procedures, people and place. The categories are prompts to widen the search, not a rule to fill in.

What is the difference between a fishbone diagram and 5 Whys?

A fishbone diagram goes wide: it lists many possible causes across several categories. 5 Whys goes deep: it follows one causal chain from a symptom to an actionable root cause. They are complementary. Use the fishbone to explore possibilities, then apply 5 Whys to the causes that survive verification to reach the level where a countermeasure can prevent recurrence.

How many causes should a fishbone diagram have?

There is no fixed number, and quantity is not the goal. What matters is that plausible causes from each relevant category are captured, that each one is specific enough to test, and that the list is then filtered with facts down to a few causes worth verifying. A diagram with a dozen precise, testable causes beats one covered in vague labels.

Origin

Cause-and-effect (fishbone) diagram — Kaoru Ishikawa, first used at Kawasaki Steel Works, 1943; the “6M” families are later common practice.

Used in these playbooks

Chronic scrap reduction month 1 month

One month against a loss everybody has learned to live with: count at the source, rank the losses, split the data, list the causes and prove the real one before spending money.

Related methods

More in “Root Causes”

List the possible causes, dig down the chains and prove the real one before acting.