Continuous Improvement Toolkit

Fishbone Diagram

Dr. Shaymaa Elsawaby

 Fishbone Diagram
❑ Identifies and organizes the potential causes of a business
problem.
❑ Uses an easy and understandable format.
❑ Used to identify the sources of process variation.
❑ It is also called Ishikawa Diagram and Cause and
Effect Diagram.
 What do you mean by “ Cause & Effect”?

A PROBLEM WHICH A POTENTIAL

HAS OCCURED FUTURE PROBLEM
(FMEA)

CAUSE EFFECT CAUSE EFFECT

Events/conditions Symptoms that Events/conditions Symptoms that
that led to provide evidence that would lead to would result
the problem of the problem the problem from the problem
Cause - Effect Relationship
A PROBLEM WHICH
HAS OCCURED

CAUSE EFFECT
Events/conditions Symptoms that
that led to provide evidence
the problem of the problem

ACTION
What is a Fishbone Diagram?
A visual tool used to identify, explore and
graphically display all the possible causes related
to a problem to discover root causes.

A Fishbone diagram is also known as a Cause and

Effect Diagram or Ishikawa Diagram.
C/N/X
Materials Methods

C C

N N
Problem/
Desired
Improvement
N N
N
C
C
C

Machinery Manpower
Dr Kaoru Ishikawa
Quality control statistician
Professor in University of Tokyo
One of the pioneers of Japan’s quality
revolution in the 1940s
Played major role in growth of QC circles
Best known for formalizing use of Cause-
and-Effect Diagram
Won the Deming Prize and Shewhart
Medal
ASQ established the Ishikawa Medal to
recognize the human side of quality
 Why Use Fishbone Diagrams?
To discover the most probable causes to a problem
(or effect)
◦ Sometimes, the effect can be a desirable effect.
◦ When something desirable has happened, it is useful to
find out what caused it so that you can make it happen
again

To visual possible relationships between causes for a

given problem under investigation
 Fishbone Diagram

❑ Identifies the causes of an undesirable effect of a problem.

❑ It is also possible to identify the hierarchy of causes
including the possible root causes.
❑ Brings attention to the primary factors affecting the quality
of a product or service.
❑ These factors need to be optimized in
order to reduce the amount of process
variation.
❑ The outcome can provide initial
information to later problem solving tools.
 Fishbone Diagram
❑ Often used during brainstorming sessions.
❑ Everyone gains insight into the cause and effect
❑ This makes the solution easier to find later on.
 Fishbone Diagram
❑ Helps to create ideas about the possible causes of a
problem before it happens.
❑ An effective cause prevention tool.
❑ Helps to identify the potential
factors causing an effect to prevent
future problems.
❑ Can also be used in product design and
to plan new processes.
 Fishbone Diagram
❑ Potential causes are sometimes called the
process input variables.
❑ They are normally grouped into categories for
easier sharing and reference.
❑ These are often called the 6 Ms.
 Fishbone Diagram
❑ Man – anyone involved with the process and
contributes to the effect.

❑ Methods – how the process is performed and the

specific requirements for doing it, such as:
• Policies.
• Procedures.
• Rules.
• Common practices.
 Fishbone Diagram
❑ Machines – including the equipment and tools
required to do the process.

❑ Materials – The material needed to produce

the product or provide the service, including:
• Raw materials.
• Parts.
• Papers.
• Packing.
• Consumables.
 Fishbone Diagram
❑ Environment (or Mother Nature) – the conditions in
which the process operates such as:
• Location.
• Time.
• Temperature.
• Culture.

❑ Measurements – including the data that are used to

evaluate the performance of the process.
 Fishbone Diagram
❑ Typical Fishbone Diagrams for Manufacturing and Non-
manufacturing Processes:
Manpower Environment Machines

Manufacturing
Materials Measurement Methods

People Environment Equipment

Administration
and Services
Policies and Procedures
How to Construct a Fishbone Diagram??

❑ Clearly define the effect of the problem.

❑ Write it down at the center right of a large piece of
paper.
❑ Determine the major categories of causes.
❑ Write them on the branches of the diagram.
❑ Use brainstorming to log all possible causes under
the appropriate branches.
❑ Use 5 Whys to search for root causes
then add them to the diagram.
How to Construct a Fishbone Diagram??

❑ Take time to ensure the appropriateness of the recorded

information.
❑ Update the chart as new causes become apparent.
❑ Mark or assign numbers near to the key causes to show their
relative priority.
❑ Collect data and investigate to verify the key causes are actual.
❑ Plan and implement actions to address the key causes.
 Ways of Finding More Causes

❑ Keep asking questions to encourage everyone to

participate.
❑ Involve other people, especially those who have
experience in the process.
❑ Leave the chart on the wall for few
days and encourage passers-by to
contribute.
❑ Take a break or do something to take
the team’s mind off the current
thoughts.
Constructing a Fishbone Diagram
Fishbone Diagram for Surface Flaws

Measurements Materials Man

List specific
causes in each
category

Surface Flaws

Problem (effect)
Environment Methods Machines
at the “head of
the fish”

Major categories of causes

(or sometimes call major bones)
Constructing a Fishbone Diagram
Identify possible causes through Brainstorming
• Identify specific causes within each major category that
may be affecting the problem.

Fishbone Diagram for Surface Flaws

Measurements Materials Personnel

3. Continue Micrometers
Calibration Method
2. Repeat this
Alloys Shifts
asking: ‘Why is Calibration Interval procedure with each
Precision
this happening?’ Accuracy Lubricants Supervisors specific cause to
until you no produce sub-causes.
Microscopes Suppliers Training
longer get useful
information. Inspectors Operators

Surface Flaws
Machine feedrate
Speed
Machine rpm

Brake Lathes
Brand of bit
Condensation Engager Bits
Size of bit
1. The team should ask :
Moisture% Angle Sockets
‘What are the machine
Environment Methods Machines issues affecting/causing
the problem?’

When do we know we have reached the root cause ?

 Example –
The Increased Invoice Errors for a Company

Environment Equipment Method

Poor ergonomics Cumbersome process

Inadequate printers
Noisy office
System incompatible Manual data entry
System not integrated
Invoice
Faults
Poor audit system Low morale
Paper quality
Inadequate Increased workload
validation process
Inadequate paper supply
Poor technical support

Material Measure Man

 Example –
The Increased Coolant Consumption in a Factory

Environment Machine Method

Adjusting
Machine leakage Machine breakdown Concentration settings
Evaporation
in hot seasons Adding WBLFto
Trench gets the tank directly
other materials Blocked hoses
High Coolant
Consumption
Rounding Lack of training
Biocide level WBLF not measured measurement
per line separately Poor communication
Antifoam WBLF not between shifts
level measure on time

Material Measure Man

 Further Information
❑ Different names may be used for the same category.
• Procedures could be used instead of Methods.
• Equipment could be used instead of Machines.
❑ You may even create your own branches that suit your needs.
❑ An affinity diagram could be used to create the branches of
the diagram from the titles of the affinity sets.
❑ Don’t overload categories.
• Establish another category if needed.
 Further Information
❑ You could also create sub categories under the main
categories.
❑ For example, under Man there might be:
• Poor Communication
• Absenteeism
• Lack of Training
• …
❑ You should, however, avoid writing real names of persons.
❑ Beware of adding causes which are actually solutions.
❑ A Cause and Effect Matrix could be used to prioritize the causes
of the problem.
❑ Prioritizing and selecting the key causes will minimizes the need
for more statistical evaluation of inputs that are unlikely to have
an impact on the output.
Y1 Y2 Y3

X1
X2
X3
X4
Cause & Effect
Matrix

26
 Cause and Effect Matrix
Cause and Effect
Matrix
Rating of
Importance to
Customer
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement

Requirement
Total

Process Step Process Input

1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
10 0
11 0
12 0
13 0
14 0
15 0
16 0
17 0
18 0
19 0
20 0
0
0

0
Total
Lower Spec
Target
Upper Spec
Description: Cause and Effects Matrix
Simple QFD (Quality Function Deployment) matrix.
Used to relate and prioritize X’s to customer Y’s through numerical
ranking using the process map as the primary source.
Y’s are scored as importance to the customer
X’s are scored as to relationship to outputs

Results
Pareto of Key Inputs to evaluate in the FMEA and Control Plans
Input into the Capability Study
Input into the initial evaluation of the Process Control Plan
This is the team’s first stab at determining Y = f(X)
Constructing a Cause & Effect Matrix
1. List key outputs (Y’s)

Cause and Effect

Matrix

Rating of
Importance
to Customer
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Homogeneity

Temperature
Consistency
Cleanliness
Viscosity
Gel Time

Solids
Color

Process Inputs Total

1 0
2 0
3 0
4 0
5 0
6 0
7 0
Constructing a Cause & Effect Matrix
2. Rank Y’s with respect to customer importance

Cause and Effect

Matrix

Rating of
Importance to 9 9 7 10 10 9 3 2 6
Customer
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Homogeneity

Temperature
Consistency

Digets Time
Cleanliness
Gel Time

Viscosity

Solids
Color

Process Inputs Total

1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
Constructing a Cause & Effect Matrix
3. List key inputs (X’s)

Rating of
Importance to 9 9 7 10 10 9 3 2 6
Customer
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Input

Homogeneity

Temperature
Consistency

Digets Time
Cleanliness
Gel Time

Viscosity
Variables

Solids
Color
Process Inputs Total

Scales
1 0
Accuracy
Preheating
2 0
DICY TK
DMF Load
3 0
Accuracy
DMF
4 0
Cleanliness
DMF Raw
5 0
Materials
DICY Load
6 0
Accuracy
DICY Envir.
7 0
Factors
DICY Raw
8 0
Materials
DICY Mixer
9 0
Speecd
Constructing a Cause & Effect Matrix
You are ready to correlate customer requirements to
the process input variables

Avoid confusion and inconsistency by establishing

scoring criteria:
0 = no correlation
1 = the process effect only remotely affects the
customer requirement
4 = The input variable has a moderate effect on the
customer requirement
9 = The input variable has a direct and strong effect on
the customer requirements

Note: Not recommended to use more than 5 different criteria.