← Back to Blog
GUIDE7 min read · March 19, 2026

Fishbone Diagram: How to Run an Ishikawa Analysis That Actually Finds the Root Cause

Most fishbone diagrams produce a wall of brainstorm output and no actionable finding. The problem isn't the tool — it's how teams use it. The fishbone is a structured hypothesis framework, not a whiteboard free-for-all.

What the fishbone diagram actually is

The Ishikawa diagram — named after Kaoru Ishikawa, who developed it at Kawasaki in the 1960s — is a cause-and-effect analysis tool. It structures potential causes of a problem into categories, with the goal of generating a complete picture before narrowing to the most likely root cause.

The key word is "before." The fishbone is not meant to identify the root cause on its own. It is meant to surface every plausible category of cause so that the team does not miss something obvious by focusing too early. The 5 Why analysis then drills into the most credible branch.

The 6M framework for manufacturing

For manufacturing environments, the standard framework is 6M. Each category prompts a different line of investigation:

Man (People)
Training, certification, experience, fatigue, adherence to procedure. The most common category and the most misused — "operator error" is a symptom, not a root cause.
Machine (Equipment)
Wear, calibration, setup, maintenance intervals, age. Always check whether the problem correlates with a specific machine or shifts across all machines.
Method (Process)
Standard work gaps, SOP accuracy, sequence variability, measurement methods. If the standard allows variation, the variation isn't the cause — the standard is.
Material
Supplier variation, incoming inspection, storage conditions, batch traceability. Problems that start intermittently and improve or worsen after a delivery are often material-driven.
Measurement
Gauge R&R, calibration status, measurement system consistency. A significant fraction of quality escapes turn out to be measurement problems rather than process problems.
Mother Nature (Environment)
Temperature, humidity, contamination, shift timing, seasonal variation. Often overlooked, sometimes decisive — particularly in electronics, food, and pharmaceutical environments.

How to run the session correctly

Start with a precisely defined problem statement. "High defect rate" is not a problem statement. "Weld joint rejection rate at Station 4 is 3.2% against a target of 0.5%, occurring on the day shift between 10:00 and 14:00 since March 11" is a problem statement. The more specific the effect, the more specific the causes will be.

Fill every category before evaluating any. The discipline of the tool is in completing the full picture before narrowing. Teams that short-circuit to their favourite explanation skip the category that contains the actual cause. Go around the full 6M before anyone argues for a specific branch.

Use evidence to score each branch. After brainstorming, ask for each potential cause: do we have data that supports or contradicts this? A cause with supporting evidence gets prioritised. A cause that is plausible but unverified gets flagged for investigation. A cause contradicted by existing data gets removed.

Transition to 5 Why on the highest-priority branch. The fishbone finds the most credible direction. The 5 Why goes to the bottom of it. They are sequential tools — the fishbone is not complete until it feeds a deeper analysis.

A real manufacturing example

Problem: Dimension out of tolerance on machined bore — 4.2% rejection rate, target 0.5%.

Man
Operator B is new and hasn't been certified on the post-shift warm-up procedure
Shift handover notes not consistently completed
Machine
Spindle bearing showing early wear — vibration measurement elevated
Tool holder showing 0.003mm runout at inspection
Method
Warm-up procedure specifies 15 min but SOP says 5 min — discrepancy not resolved
First-piece inspection sometimes skipped when line is behind
Material
Incoming billet hardness varies 12% across suppliers
Last batch from Supplier B showed higher hardness
Measurement
Gauge last calibrated 14 months ago — 12-month interval
Two gauges in use — no study on inter-gauge correlation
Environment
Machine located near dock door — temperature swing of 8°C over shift
Coolant concentration not checked since maintenance window

After evidence review: the gauge calibration lapse and the warm-up procedure discrepancy both had supporting data. The team ran a 5 Why on each. The warm-up discrepancy traced to a standard work update in January that was not propagated to the SOP — a change management failure, not an operator failure. That's the root cause.

Fishbone in VeSiMy

VeSiMy's Fishbone tool supports 6M Manufacturing, 8P Service, 4S, and Custom frameworks. Causes are added by category, and the AI can generate initial cause suggestions based on the problem statement and your process context. When the analysis is complete, it feeds directly into the 5 Why tool on the same step — the problem statement carries over and the team doesn't re-enter context. The combined analysis exports as an ISO 9001:2015 §10.2.1 compliant root cause report.

Run a Fishbone analysis free

Connected to 5 Why and your VSM. AI-assisted cause generation. ISO export when you need it.

Start free →