Fishbone Diagram
Fishbone Diagram
2. State the need(s) clearly and concisely. Make sure that all group members agree
with the needs as stated. For example, the national administration has managed to
present 25% (on average) of requested reports on time in the last 10 years. Now,
the objective of your program/project is that 100% of the reports requested each
year are transmitted on time and that their content is informative and relevant.
3. On a long sheet of paper, draw a horizontal line along its entire length. This line will
represent the "backbone" of the fish. Write the need along this spine on the left
side.
5. With the help of brainstorming, the group must identify the factors that may affect
the cause and/or need. For each category of causes, ask the group "Why does this
happen?" Add each "reason" to the diagram, indicating it around the category of
the main cause it affects.
6. Repeat the procedure asking the question "Why does this happen?" for each effect,
until there are no more answers to the question (see image 2).
7. Once the group has established that the diagram contains enough information,
proceed to analyze the diagram. Look in particular for causes that appear in more
than one section of the diagram.
8. Draw a circle around all that appear to be root causes at the origin of the need.
Classify the causes in order of priority and define the action to be taken. Such action
may be a further study of the root causes.
Using a fishbone diagram, the group's attention can be drawn to the "whole situation"
from the point of view of the causes or factors that may have an effect on a
problem/need.
Even after the need is addressed, the fishbone diagram indicates weaknesses that can
be rectified – once presented – before they cause further difficulties.
Disadvantages
The simplicity of a fishbone diagram can represent both a strength and a weakness. A
weakness, because the simplicity of this type of diagram can make it difficult to
represent the very interdependent nature of problems and causes in very complex
situations.
Unless there is a large enough space to draw and develop the diagram, it may happen
that the necessary conditions are not available to delve into the cause-effect
relationships as would be desirable.
The goal of this technique is to help us discover vital information in a systematic way,
analyze hidden causes and develop solutions to the questions raised. This analysis can be
applied to resolve a conflict, diagnose a problem or make decisions.
For more information, read the article Root Cause Analysis , which explains how to use
this methodology and others similar.
This technique was first used at Toyota during the evolution of its manufacturing
methodologies, which would later culminate in the Toyota Production System (TPS). This
technique is currently used in many areas, and is also used within Six Sigma.
Example
The following simple example shows us the use of this method. We start from a postulate:
My car won't start. (the problem)
Why doesn't it start? Because the battery is dead.
Why is the battery dead? Because the alternator doesn't work.
Why is the alternator not working? Because the tape broke.
Why did the tape break? Because the alternator is beyond its useful life and was not
replaced.
Why wasn't it replaced? Because I am not maintaining my car according to the
manufacturer's recommendations.
Obviously, this example could be followed further, with more questions. This would be
correct, since the "five" in the "Five Whys" technique is not fixed, but rather an
encouragement to do several iterations to find the root cause.
Root cause analysis
A root cause is the initial cause of a chain of causes that lead to an effect of interest.
Generally, the root cause is used to describe the place in the chain of causes where an
intervention could be implemented to prevent undesirable outcomes.
It is important to know when to stop the analysis. In the previous example, one could
continue asking why the car did not have maintenance, and then why the vehicle had a
design that needed this type of maintenance.
In general, it is the analyst's own framework that determines when the analysis should
stop. For example, if viewed from the point of view of the car owner, then the analysis
might stop at the fifth why. However, if the frame of reference is the car manufacturer,
who is responding to thousands of complaints about this problem, the stopping point of
the analysis would have to reach the scope of design.