“Supply Chain Management Handbook”

Root Cause Analysis and Problem Solving

Updated September 2009

1
Objective

Root cause analysis and problem Solving:

Propose methodology to improve the way escapes
and problems are managed (including
communication between all actors - engineering,
MRB, supplier, customer, etc...) to reduce their
impacts, contain them as far upstream as possible
and prevent recurrence (ensure right measures
are taken at the right location and right time).

2
Rationale of the objective
• Objective of continuous improvement is to reduce number of issues
(undesirable conditions, defects and failures) but also to minimize their
impact on quality, delivery performance and costs.
• Several big issues originate with small problems that were discovered
but were never resolved due to incorrect analysis and/or ineffective
actions.
• Often, organisations and their suppliers do not deliver correct Root
Cause Analysis and Problem Solving results because:
– No clear criteria exist about what makes an acceptable corrective action plan
(people satisfied when they don’t receive any more defective parts and stop
here)
– People continue to accept bad answers (even if they say they will not, reality
of life makes them accepting )
– People (internal and external) have not the Root Cause Analysis culture,
don’t know any process or are not effectively trained
• Cultural change required to overcome the situation
3
Benefits of effective Corrective
Action process (Brainstorming)
• Sustainable performance (product conformity, on time delivery, etc…)
• Reduce cost of quality (reduces costs by eliminating wastes and unnecessary efforts)
• Protect internal and customer operations
• Solving and preventing problems on similar parts, processes, lines, etc
• Really understand the issue and identify corrective action at first time
• Ensures right measures are taken at the right location and right time
• Ensure stable processes
• Don’t repeat failures
• Customer satisfaction
• Supplier reputation
• Objective evidence for evaluation of corrective action effectiveness
• Compliance with international standards and aviation authorities requirements
• Improved communication between all stakeholders of the supply chain
• Motivate teams and recognise efforts
• Continual improvement
• Harmonise corrective action process between all stakeholders (suppliers, customer, etc)
• Allow use of unique IT system structure (same fields, definition, etc…) in the future
• Lessons learned captured and used
• Beneficial to the whole organisation (improve overall business results)
• reduce impact of problems to the minimum and contain them as upstream as possible
• Reduce time to get a fix (avoid diverting staff from key roles and objectives)
4
Risks and impacts of not having an effective
corrective action process (Brainstorming)
ƒ Believing you have fixed an issue but not having done it.
ƒ Stopping investigation at first identified cause
ƒ Believing one problem has only one cause, so one fix
ƒ All non-conformances cost money, which reduces investment, money available for pay rises, potential
to retain business
ƒ Fix the issue on one product or in one area and repeating similar mistake somewhere else later,
especially for new programs
ƒ Look for guilty, not for solutions
ƒ Used for blaming or transferring responsibility
ƒ Blame systematically on Quality, not looking for true responsibilities
ƒ Repeating mistakes
ƒ Permanently working in fire fighting mode
• Not enough time for analysis
• Dealing only with same big problems and never dealing with the other systemic ones (no time to fix
them)
• Going from one crisis to another
• Negative customer perception
• Loss of business
• Loss of international approvals
• Hidden costs
• Loosing opportunity for improvement
• Not attacking problems by adequate priority order (no risk based approach)
• Not attacking problems in a systematic way
• inability to protect your customer’s business
• Letting small problems develop until they become critical
• Spend valuable resources and substantial investment applying inappropriate methodology
• Waste more time to attempt to demonstrate there is no problem than to understand and fix it
• Loss of motivation due to increase of mistakes which leads to even more mistakes 5
• Not capitalising on experience (not updated instructions, paperwork, drawings, etc)
Key factors of success
(Brainstorming)
• Immediate containment “Stop the • Communicate (includes
bleeding” notification to customer)
• Build team and assign responsibilities • Identify solutions
• Identify common goals • Implement solutions
• Identify action owners and time scales • Verify effectiveness of the
• Identify external measures (actions) solutions
• Protect the operations (e.g. over • Manage change effects
production) • Measure new performance
• Contain the situation “Build the wall • Review performance
higher” (e.g. over inspection) • Assess success
• Map process • Document changes
• Identify internal measures (actions) • Capture learning (keep lessons
• Measure current performance learned register)
• Identify and prioritise key problems • Recognise team and celebrate
• Identify root causes success
Top Management must be committed to the
corrective action process to ensure effectiveness
6
3 Problem Solution Types

• Reactive mode
Solving the abnormality that has occurred, gathering
and analysing data aims to provide a customer
protection and countermeasure.

• Pro-active mode:
Analysing failures and looking for improvements.

• Preventive mode:
Putting in place solutions before undesirable
condition, defect or failure occurs.
7
Effective root cause analysis:
A cultural change
Reacting to an Event Systems Thinking
Traditional solutions Systemic solutions
• Fire Fighting • Many factors making up a complex
situation
• Quick Fix
• Fully understanding the problem and
• Not taking enough time for analysis then addressing the systemic root
• Going from one crisis to another cause(s)
• Look for the guilty party. ”Who did • Permanently fix and improve
that?” performance
• Seek total understanding of the
• Generate laundry list of solutions to process: How did that happen?”
firefight the symptoms.
• Take time to understand the big picture,
• Narrow focus results in sub- to dialogue, and to elicit diverse
optimization of system. perspectives, to apply the solution.
• Focus on performance metrics (e.g. •Optimize the whole enterprise.
sales and profits) and hope •Focus on improving processes that
processes improve. actually effect performance metrics.
8
Effective communication is mandatory

• Between Supplier and Customer to

immediately stop the problem getting worse,
ensure full understanding of the problem and
verify that implemented solutions are
satisfactory.
• Inside the organisation where the problem
originated, between all actors of the supply
chain to ensure effective root cause analysis
and definitive corrective action implementation
9
ISO 9000:2000 Terminology

• Nonconformity: Non-fulfillment of a requirement.

Note: It may be a non conforming product but also a late delivery, an
incorrect paperwork, incorrect process (production or QMS related),
etc…

• Correction: the action taken to fix the nonconformity

• Corrective Action: the action taken to fix the cause of the
nonconformity and to prevent recurrence
• Preventive Action: the action taken to prevent
nonconformities or problems from occurring

10
Terminology: New definitions
(IAQG Standard 9136 - draft May 09)

• Apparent Cause: The event or action that immediately

results in or precedes the nonconformity. May also be called
Obvious Cause, Direct Cause, Immediate Cause
• Root Cause(s): The original event(s), action(s), and/or
condition(s) generating (directly or in cascade) an actual or
potential undesirable condition, situation, nonconformity or
failure.
– Note: There are often several root causes for one problem

• Contributing Causes: Contributing causes are causes that

taken alone would not cause the problem but can increase
the risk of the issue to happen. Analysis for these causes
generally require taking small steps (or a finer look) to be
identified and fixed. 11
Terminology: New definitions
(IAQG Standard 9136 - draft May 09)

• Root Cause Analysis (RCA): The process of identifying all

the causes (root causes and contributing causes) that have
or may have generated an undesirable condition, situation,
nonconformity or failure.
• Containment - the action to mitigate impact of the problem
and protect the operations/customers (stop the problem
getting worse). Includes correction, immediate corrective
action, immediate communication and verification that
problem does not degrade

12
Terminology: New definitions
(IAQG Standard 9136 - draft May 09)
• Immediate Correction: Action(s) taken to immediately fix the
nonconformity.
– Note: For a product non conformity, correction might be understood as
reworking the part, accepting the non-conformance through concession
process, or ultimately scrapping it. For a system issue, it may include
correcting the paper work or issuing a new purchase order. For a
delivery issue, it may include air transportation instead of by truck or by
ship, increasing production rate, etc
• Immediate Corrective Action: Action(s) taken to eliminate,
prevent, or reduce the probability of any additional non-
conformances related to the apparent cause from happening
again in the short term.
– Note: These actions may be temporary and should remain in place until
root cause(s) is(are) identified and permanent root cause corrective
action(s) is(are) implemented and verified to be effective.
13
Terminology: New definitions
(IAQG Standard 9136 - draft May 09)

• Root Cause Corrective Action: The corrective action(s)

implemented to address the root cause(s) of the
nonconformity that will prevent recurrence.
• Root cause corrective action verification: Actions taken
to verify that the planned actions were taken as scheduled.
Note: This includes specific actions, milestones, completion
dates, and responsibilities.
• Root cause corrective action effectiveness: Actions
taken to verify that the planned actions have permanently
prevented recurrence of the identified root cause(s).
– Note: This may include auditing, monitoring of specific metrics, or
any other reporting methodologies.

14
Relationship between some existing
Root Cause Analysis methodologies
7-Step 8D MTU & Airbus - 8D Boeing model Rolls-Royce 7 Step Safran Impact 8D
1. Identify team 0. Immediate 0. Define the problem and make sure
1.Select
1.Select 1.Define Problem
project/theme containment that the methodology is necessary
2. Define problem
1. Identify team Correct, 2. Contain 1. Build the team
Contain, Corrective problem
2. Define problem Communicate Action 2. Describe problem
3. Contain 3. Find root cause
2. Grasp present symptom 3. Contain of escape 3. Define and launch temporary
status symptom actions to contain the risks
4. Prevent further
4. Identify root escapes 4. Find the root
4. Identify root Identify Root Cause
causes causes
causes
3. Analysis 5. Find root cause
5. Choose corrective 5. Choose corrective of problems 5. Find and grade
action action corrective actions
Root Cause
4. Implement Corrective Action 6. Implement
corrective action corrective action
6. Implement 6. Implement 6. Do corrective actions and
corrective action corrective action make sure they are efficient

5. Check the Root Cause Corrective

effects Action Verification

6. Standardise and
control 7. Verify fix

7. Make change 7. Make change 7. Capitalize, perpetuate,

Root Cause Corrective
permanent permanent generalize to avoid problem’s
Action Follow Up
7. Conclusions recurrence
and future plans

8. Recognise the 8. Recognise the 8. Conclude the process

team team and congratulate the team
15
Process Steps assembling based on
these existing methodologies
Immediate correction and Start Immediate containment
containment actions S0 actions

Build the team

S1 Build the team
Define problem

Find why problem S2 Define problem

was not detected

Contain Complete and optimise

symptom S3 containment actions

Identify Root Cause(s)

S4 Identify Root Cause(s)

Define and prioritise

corrective actions Define and select root
S5 cause corrective
Implement actions
corrective actions
no
Implement root Is the root cause
Verify Corrective cause corrective
Actions are effective S6 actions and check corrective
effectiveness action effective ?
Make change
permanent and
yes
Standardise and transfer the
transfer the S7 knowledge across business
knowledge

Recognise the team Recognise and

S8 close the team 16
Process Steps and Results
Start Immediate containment
S0 actions Immediate protection

S1 Build the team

S2 Define problem

Complete and optimise

S3 containment actions Temporary fix

S4 Identify Root Cause(s)

Define and select root

S5 cause corrective
actions

no
Implement root Is the root cause
cause corrective
S6 actions and check corrective Durable solution
effectiveness action effective ?

yes
Standardise and transfer the
S7 knowledge across business Systemic Improvement

Recognise and
S8 close the team 17
Sequencing Consideration
This diagram provides the typical sequence in
Start Immediate containment
applying the 9 steps. Time relationship
S0 actions between steps shall be defined according to
the type and criticality of the problem and/or in
S1 Build the team line with contractual requirements. In some
cases, one step may start before the previous
S2 Define problem has ended or different steps require to be
conducted in an iterative mode
Complete and optimise
S3 containment actions

S4 Identify Root Cause(s)

Define and select root

S5 cause corrective
actions

Implement root
cause corrective
S6 actions and check
effectiveness

Standardise and transfer the

S7 knowledge across business

Recognise and
S8 close the team
18
Time
When to launch a structure root cause
analysis and problem solving process?
• Launching a formal Root Cause Analysis and problem
solving process shall always be considered when an issue
(undesirable conditions, defects and failures) is detected.
• Decision not to apply the process shall be made based on
objective evidence of absence of risks.
• In any case it shall be applied if one or more of these
conditions exist
– Safety impact
– Product strength, performance, reliability issue
– High impact on operations
• Stop the line, prevent next operations to occur satisfactorily, etc
• Regulatory authorities and/or customer dissatisfaction
• Costs issue (generated to your Customer or to your organisation)
– Repetitive problems (on one part, similar activity or similar process)
– Difficulty to detect
– Customer request
– Significant QMS issue 19
When to launch a structure root cause
analysis and problem solving process?

Customer request shall always

Impact take precedence

M
an
da
Re to
co ry
m
m
en
d ed
Op
t io
na
No l
tr
eq
ui
re
d

20
Frequency
Step 0: Start Immediate containment
actions “Stop the Bleeding”
• Objective: To mitigate the impact of the problem, protect the
customer operations and the organisation (stop the problem
getting worse) and verify that problem does not degrade until
the root causes are known.
• Output: Immediate containment actions implemented,
Customer protected
• WHAT: First thought “ What do we have to do to protect the Customer and the
organisation (eliminating the impact of the effect) ”?
• WHY: The problem that has been identified is having an impact now on the
customer or on the organisation. If we don’t do anything, the problem will
degrade: This justifies a containment plan.
• WHEN: Action generally required within 24 hours (or less in critical cases).
• WHO: Must be assigned to an individual for implementation.
21
Step 0: Start Immediate containment
actions “Stop the Bleeding”
• HOW:
– Identify, isolate, perform immediate correction of all defective parts or data
– Identify apparent cause and perform immediate corrective action to eliminate,
prevent, or reduce the probability of any additional non-conformances from
happening again in the short term.
– Typical immediate containment actions may include
• immediate stop of the working process
• Stop deliveries
• Recall product (still within the organisation or already delivered)
• Over inspection - “Build the wall higher”
• Inventory checks and segregation of defective parts
– Identify immediate potential risk on same parts if not detected => Determine
apparent criticality: “Am I able to assess criticality”? If not, who should I
inform (customer, design office, type certificate holder, airworthiness office
?…) to assist me in evaluating the criticality.
Note: Immediate containment actions terminate when corrective actions are in
place or when study has found a more effective containment action (Step 3 or
Step 6): Immediate containment actions must have an agreed effective life span
22
Step 0: Start Immediate containment
actions “Stop the Bleeding”
• COMMUNICATION: You are likely to require communicating and implementing
action(s) across various entities and organisations.
– Identify who is affected by the issue
– Has the issue an impact now (internally or to your customer)?
– Inform all impacted parties (next cell, sub-tiers, customer, etc)
• Product or data concerned
• Nature of the issue as known at this time
• How and when it was detected
• Apparent consequences
• Containment actions taken and recommended actions at customer level when
relevant
• What is the next step (e.g. if Root Cause Analysis process is being considered -
see specific page later)
• If customer help is required (e.g. to determine criticality)
• When next communication will occur
• Who is the focal point at this point in time
– Immediate information to the customer is mandatory if product has been
delivered which is known or suspected to be affected by the issue, impacting
safety and more generally having a significant impact on customer’s
operations. 23
Step 0: Start Immediate containment Menu

actions “Stop the Bleeding”

• TO BE CONSIDERED:
– Who is suffering or may suffer from the issue and where?
– How and where is it necessary to contain problem?
– What are the investigations to be undertaken to know more about the issue?
– Are there other products or production lines or data that could be affected by a
similar issue and that require similar containment actions to be launched at other
sites, at customer, …?
– What is being done to contain this issue & how is the customer informed and
being protected?
– How are finished products in stores, WIP (Work In Process), at suppliers and
customers being managed?
– What checks are being performed or should be launched, by who and where?
– What is the planned date for permanent resolution (allows to evaluate how long
the containment actions are likely to remain in place)?
– Which data do we need to collect to support deeper analysis ?

24
Step 1: Build the team
• Objective: To ensure that all different actors (organisation,
suppliers, customers) and functions that may have an
influence on the corrective action process, including
identification of the root cause(s), are in the team
• Outputs: Cross functional team of experts in place
• WHAT: Gather a team representing different functions that may have an
influence on the problem and that are prepared to assist in its resolution.
• WHY: The corrective action process, including root cause analysis is always
more successfully conducted by a team knowing the process and owning the
data, than by individuals.
• WHO:
• Top management must support the team approach to the corrective action
process
• Note: For big issues, it is recommended to set up a steering committee composed
of members at adequate level of management (not participating to the working
Root Cause team) able to validate decisions within the entire organisation and
ensure right level of empowerment. 25
Step 1: Build the team

• WHO (continued):
• Team leader (facilitator) shall be nominated
• Should preferably be someone without any hierarchical role (not directing the
analysis but ensuring targets and timing are met)”
• Must be empowered by the appropriate management level commensurate to the
issue
• Must be nominated based on experience of root cause analysis techniques or
must have access to the appropriate help by specialists, but must not be
necessarily an expert in the processes/problems being analysed.

• WHEN: As soon as problem is considered as justifying a root cause analysis.

Note: Step 1 (Build the team) and Step 2 (Define problem) may be conducted
concurrently in an iterative process

26
Step 1: Build the team
• HOW:
• Identify representatives from functions that may contribute to the corrective
action process, including identification of the root causes (see table)
• Assign responsibilities and objectives
Notes:
• Remember, those performing the job (operators, inspectors, assemblers, etc) are the
best to identify the real causes: Don’t leave them out of the team !
• Size and composition of the team depend on the complexity and on the impact of the
problem
• The composition of the team is not fixed forever and may evolve depending on the
analysis results and needed actions: New actors may join the team if analysis shows
they are identified as being in the scope, some others will leave if their area is
definitely identified as out of the scope.
• However, consideration should be made that expending the size of the core team
over 6 to 8 members generally results in less efficiency. When more members or
special skills are required, sub teams should be considered.
• Don’t forget a root cause analysis shall NOT be used for blaming or transferring
responsibility

27
Step 1: Build the team
• COMMUNICATION:
• Team leader needs to know that he has been nominated, why he has been
assigned this role and the team objectives and constraints
• Each team member needs to understand his role and objectives
• The management of each team members needs to know level of
involvement of his staff (time, duration, role)
• All stake holders must be informed of the team composition and objectives

28
Step 1: Build the team
People joining the team must be selected based on how they are impacted
by the problem and how they can help to find an effective corrective action.

Could contribute to

or disturbed by the
Could help solving

Will help selecting

Sees the problem

May be impacted
Suffers from the

Is in charge of

the problem

the problem

fixing of the
solving the
problem

problem

problem
solution

Others
Name 1

Name 2

Name 3

Name 4

Name 5

Name 6

Above selection criteria for example only, others may be found 29

Step 1: Build the team
• To be considered: Team dynamic can have positive or negative results: It is
therefore important to ensure that the team leader is selected on ability to effectively
manage team dynamics:
Positive Negative
Outcomes Outcomes
Leads to new ideas Diverts energy from work

Stimulates creativity Threatens psychological well-being

Motivates change Wastes resources

Promotes organizational vitality Creates a negative climate

Helps individuals & groups establish

Breaks down group cohesion
identities
Serves as a safety valve to indicate Can increase hostility & aggressive
problems behaviors 30
Step 2: Define problem
Go - Look - See
• Objective: To understand the significance, impact and size
of the problem and ensure the situation is fully described
and understood
• Output: Problem defined and agreed (which product, which
process, which defect or deviation)
• WHAT: This step helps to fully describe a situation, precisely analyse all its
elements and gain a common understanding of them, allowing defining an
action plan
• WHY: Often, the first time a problem is put into words, it is vague, subjective
or even abstract. Without a proper problem definition, it is likely that root cause
will not be identified and wrong or insufficient actions will be put in place
• WHEN: As soon as the team is formed
Note: Step 1 (Build the team) and Step 2 (Define problem) may be conducted
concurrently in an iterative process
• WHO: All the team

Well begun is half done…. 31

Step 2: Define problem
Go - Look - See
• HOW:
– A good problem statement is supported by objective evidence, based on
facts and figures, not on perception
– Must understand the scope and extent of the problem and the symptoms
which are being experienced by the customer (either internal or external)
• Effect - Something resulting from a cause .
• Cause - Something producing an effect .
Notes
– This may be an iterative process and could take some time.
– The problem statement should be updated as more information is gathered
• COMMUNICATION:
– Continual communication between all team members is mandatory, for
instance through regular reviews until problem is clearly identified, defined
and agreed by all.
– Problem definition and extent must be communicated to all stake holders
and agreed, especially by the customer when he is impacted.
32
Step 2: Define problem
Go - Look - See
• TO BE CONSIDERED:
– Which are the elements (operations, products, materials, defects,
malfunctions, …) that may characterise the situation ? What is it about
(process step, …) ?
– Who is concerned with the problem? Who is reporting the problem? Who is
rectifying the problem? Who is the problem affecting?
– Where are all the places where the event takes place (shop floor, services,
machine, process step, …)?
– Where is it seen ? Where does it originate ?
– When does the event appear (time, date, when does it start, how long does it
last, how often, …)
– When is the problem reported?
– Has it occurred before? If yes, what is the history?
– How do we know there’s a problem (how is it detected)?
– How does the event appear, how does it stop ?
– How frequently is the problem experienced? 33
Step 2: Define problem
Go - Look - See
• TO BE CONSIDERED (continued):
– How is the effect of the problem being measured (costs, delays, scrap rate,
customer complaints, return rate, concessions, reliability rate, etc)?
– How is the problem currently addressed ? How is it corrected?
– Why is there an action to take place? (customer impact, cost impact, safety
impact, …)
– Natural tendency is to try finding the root causes at this stage, whereas you
are not sure yet of what the real problem is
– Check every answer above by asking: ”Do we know this for sure, or are we
just assuming that this is the way things are?”

Step 2 should always be concluded by confirming

that all team members agree about the definition of
the issue and resulting impact
34
Step 2: Define problem Menu

Go - Look - See
• Problem recognition
- Effect - Something resulting from a cause .
- Cause - Something producing an effect . ?
? ?

Compare a problem to an Iceberg !

Symptoms
EFFECT

Problems CAUSES

Problem statement is a factual description of the PROBLEM EFFECT

35
Step 2: Define problem Menu

Go - Look - See

• TOOLS :
– Some tools may be used to gather and analyse data for problem definition
– Brainstorming
– Is/is not
– Comparison sheet
– Check Sheets and Tally charts
– Histograms
– Scatter Diagrams
– Control Charts
– Pareto Analysis
– Etc

36
Step 2: Define problem
Go - Look - See

The
IS / IS NOT
analysis

(Helps to identify
exactly what the
problem effect
IS…
But also, what it IS
NOT).

37
Step 2: Define problem
Go - Look - See
Comparative analysis: allows identifying at which date
differences appeared, or when changes - process , staff, design,
etc - were introduced, possibly to have generated the issue.

Description and comparaison chart

« IS » « IS NOT « Particularity Changes Dates
object
WHAT ?
anomaly

location
WHERE ?
object

calendar
WHEN ?
cycle

object
HOW MANY ?
anomaly

38
Step 2: Define problem
Go - Look - See
The check sheet and the Tally chart
7.09 7.10 7.09 7.08 7.07 7.08 7.09 7.06 7.07 7.08 7.10

7.13 7.12 7.11 7.10 7.09 7.10 7.08 7.05 7.11 7.06 7.08

7.07 7.06 7.05 7.04 7.03 7.06 7.09 7.07 7.08 7.07 7.09

7.08 7.10 7.07 7.11 7.08 7.07 7.10 7.09 7.06 7.10 7.06

7.09 7.08 7.09 7.06 7.09 7.08 7.06 7.07 7.08 7.06 7.08

7.08 7.05 7.06 7.08 7.07 7.10 CLASS7.11 7.09

7.08 TALLY
7.07 7.09 TOTAL
7.13 I 1
7.12 7.07 7.10 7.07 7.06 7.11 7.07 7.12 7.05 7.08 7.10
7.12 III 3
7.08 7.08 7.11 7.08 7.05 7.08 7.04 7.09 7.07 7.05 7.08
7.11 IIII III 8
7.09 7.07 7.06 7.07 7.10 7.07 7.09
7.10 7.08IIII 7.05
IIII III 7.09 7.04 13
7.10 7.08 7.09 7.11 7.09 7.08 7.09 7.07IIII 7.10
7.06 IIII IIII 7.08
III 7.11 18
7.08 IIII IIII IIII IIII IIII 25
7.05
7.07 IIII IIII IIII III 18
7.06 IIII IIII III 13
7.05 IIII III 8
7.04 III 3
7.03 I 1
39
 •Total

•0
•10
•20
•30
•40
•50
•60
•70

•Missing Labels

•Cap Missing
The Pareto chart
Go - Look - See

•Foreign Bodies

•Torn Labels

•Loose Labels
Step 2: Define problem

•Glass Chipped

•Incorrect Fill Height

•Glass Broken

•Damaged Closure

•Damaged Bottle
•0
•20
•40
•60
•80
•100

•Cumulative Sum (%)

40
Step 2: Define problem
Go - Look - See
The Run chart

Drill bit broke Drill bit broke

Jig recalibrated

Target value 41
Step 2: Define problem
Go - Look - See
The control chart
1. Distance between Widgets - X Bar
1. 50

1. 40

1. 30

1. 20

1. 10

1. 0
1. 41. 51. 61. 71. 81. 1.
9 10
1. 11
1. 12
1. 13
1. 14
1. 15
1. 16
1. 17

1. Actual 1. UCL 1. LCL 1. xbar

1.Distance Between Widgets - Moving Range

1.20

1.15

1.10

1.5

1.0
1.4 1.5 1.6 1.7 1.8 1.9 1.10 1.11 1.12 1.13 1.14 1.15 1.16 1.17

1.Moving Range 1.UCLr 1.R bar

42
Step 2: Define problem
Go - Look - See
The histogram
Number of Concessions / product between wk 01/02 & wk 36/02
14

12

10
Frequency

6-9 9 - 12 12 - 15 15 - 18 18 - 21 21 - 24
Concessions/product
43
Step 2: Define problem
Go - Look - See
The scatter diagram (shows relationships and how
variables correlate)
Strong, positiv e relationship Strong, negative relationship
Y

Y
Weak, positive relationship
X
X
Y

W eak, negative relationship No relationship

X
Y

X X
44
Step 3: Complete and optimise
containment actions
• Objective: To ensure containment actions suitably address
the problem definition and to verify that immediate
corrective actions are commensurate with the problem,
implemented and effective
• Output: Completed and optimised containment actions
implemented (symptoms contained)
• WHAT: To check that all non-conforming product or data has been isolated
and corrected to prevent their escape, and optimise immediate corrective
actions to minimise impact on the customer operation and the organisation until
the root cause of the problem is understood and permanent effective corrective
actions are taken
• WHY: When problem is well defined, it is very likely that immediate
containment actions need to be further developed, added, or optimised and
some may be removed.

45
Step 3: Complete and optimise
containment actions
• WHEN: As soon as all team members agree about the definition of the issue
and resulting impact
• WHO: The owner of each action and all team members to verify effectiveness
of actions taken to date
• HOW:
– Identify action holders and time scales
– Identify potential risk on same and similar parts or data if not detected
– Confirm criticality with all team including suppliers and customer if required.
– Generally, addition containment actions may include:
• Temporary increase of production
• Over inspection upstream in the process
• Stock segregation in sub-tiers
• Similar product recall (still within the organisation or already delivered)
Note: Customer might be internal or external customer

46
Step 3: Complete and optimise
containment actions
• COMMUNICATION:
– Continual updating of communication between all team members is
mandatory, for instance through regular reviews until containment actions
are clearly identified, agreed by all and implemented.
– Nature of containment actions must be communicated to all stake holders
and agreed, especially by the customer if some product has been delivered
and/or if he might soon be impacted (e.g. deliveries will stop)
• TO BE CONSIDERED:
– Find why problem was not detected and act accordingly
– Is initial (immediate) containment still required?
– Containment plan may include other equipment, areas, data, etc… that
could potentially be affected.
– The containment plan should prevent the same issue arising at other sites,
products or production lines of the suppliers
– What checks need to be performed, by who and where ?

47
Step 3: Complete containment
actions
• TO BE CONSIDERED (continued):
– How products in stores, WIP (Work In Process), at suppliers and
customers are being managed?
– Do you have an estimation (order of magnitude) of how long it will take to
implement actions that will permanently solve the problem (the answer to
this question will give you an estimation of how long these containment
actions are likely to remain in place).
– How do you manage communication and action implementation across
various entities and organisations ?

The risk and natural tendency are that these

containment actions, which should be temporary in
nature, become permanent and people are satisfied
with, so will never find the root causes.
48
Step 4: Identify root cause(s)
• Objective: To identify, through structured root cause
analysis, all causes that have or may have generated or
contributed to the undesirable condition, situation,
nonconformity or failure, then select the most critical ones
that require to be addressed.
• Output: All Root Causes identified
• WHAT: To follow a structured process to identify all causes that when
addressed by appropriate corrective actions, will durably prevent the
undesirable condition, situation, nonconformity or failure from recurring.
• WHY: No problem can be effectively and durably solved unless all its origins,
apparent or hidden, direct or indirect are identified and understood.
• WHO: All team members
• WHEN: Starts as soon as all team members agree about the definition of the
issue and resulting impact and effective containment actions are in place
49
Step 4: Identify root cause(s)
• HOW:
• Understand and map the Process
• Define what actions need to be taken in order to find the root cause(s)
• Measure current performance
• Identification of root causes by Root Cause Analysis must be supported by
objective evidence (facts and figures, not by perception) based on proven
tool(s) which shall include verification of the relation between the cause(s) and
the issue
• Prioritise causes by performing a risk analysis based on:
• Impact on the product, operations, customer, etc
• Probability to occur
• Detectability
• Select those root causes that you want to address and those that do not
require a corrective and preventive actions due to their low criticality
Note: Team leader must be nominated based on experience of root cause
analysis techniques or must get the appropriate help by specialists
50
Step 4: Identify root cause(s)
• COMMUNICATION:
• Continual communication between all team members is mandatory, for
instance through regular reviews until root cause are clearly identified and
agreed by all
• Root causes must be communicated to all stake holders and agreed,
especially by the customer when he is impacted.
• Communication internally and between various tier levels
• TO BE CONSIDERED:
• Take care not to disperse in too many directions (better performing few
effective investigations than many ineffective checks) but do not “jump” only
on what appears as evident.
• Interaction and combination of various causes
• What failed in the quality system to allow the undesirable condition,
situation, nonconformity or failure not to be detected?
• Check if problem could have been detected earlier or somewhere else

51
Step 4: Identify root cause(s)
• TOOLS:
• Some tools must be used which enable problems to be defined, data to be
collected and analysed
• Several tools may be used in step 4 (non exhaustive list, see also step 2
“Define Problem “as some are common to both steps):
Checksheets Data Collection
Histograms
Scatter Diagrams
Run and Control Charts
Data Collection & Analysis
Process mapping (Flow Charts)
Design of Experiment
Pareto Analysis
Fishbone
5 Why
Cause and effect
FMEA (Failure Mode and Effect Analysis Techniques
Analysis)
FTA (Fault Tree Analysis)
Root Cause Chain 52
Step 4: Identify root cause(s)
The Process Mapping

Start
Skin Installation Process
No R eturn for R ew ork

Load Skins Inspected: Y es Install in Jig Apply Spray

in R acks C orrect ? and D rill D ot

Trim , R em o ve
ove No
D rill M anual R ivet, C hecked
Lug, D eburr,
Identification Identify and Prior to Trim :
C lean and Apply
H oles M ark Y es C orrect?
Sealing

Fastener
Autom atic Apply Liquid Final D rill,
B racket
R iveting Shim C ountersink
Installation
R eturn for R ew ork No

Ship to Y es Inspected: C lean, Identify

EN
ENDD
N ext Step C orrect ? Scratch R epair

53
Step 4: Identify root cause(s)

The Fishbone Analysis (Ishikawa)

Environment Method Machinery

Method Sheet
Wrong Specification After sales
service Poor Lighting
Method sheets do
not contain No plant
In process measuring
inspection maintenance Spindle Speed
node
instructions

Gauges not calibrated M/c capability

No visual standards

Out of
specification
Lack of Training measuring equip
ineffective

Shift Changeover
Overloaded Out of date clamping devices

Careless checking
New Operator Foreign bodies found on Raw M/c

Measurement Man Materials

54
Step 4: Identify root cause(s)
FMEA (Failure Mode and Effect Analysis)

55
Step 4: Identify root cause(s)
The 5 why’s

Q : WHY has machine stopped ? Q : WHY overload trip ? Q : WHY Insufficient oil ?
A : Overload tripped out ! A : Insufficient oil on shaft! A : Oil pump is inefficient!

Is “Oil filter blocked with

swarf” the real root cause?
You should finish asking
“Why?” where you can
launch actions to fix the
problem. Q : WHY is pump not efficient ?
A : Pump drive shaft worn !
Q : WHY is this shaft worn ?
56
A : Oil filter blocked with swarf !
Step 4: Identify root cause(s)
The Failure Tree Analysis is a failure analysis in which
an undesired state of a system is analyzed to combine a
series of lower-level events.

57
Step 4: Identify root cause(s)
The Root Cause Chain

58
Step 4: Identify root cause(s)
ƒ Poor maintenance of
machines
ƒ Common Causes (Environmental) ƒ Normal wear and tear
ƒ 85% of Variation (Many Small
ƒ Insufficient training
Problems)
ƒ Predictable
85% ƒ Not one way of working
ƒ Poor working conditions
ƒ Difficult to Eliminate
ƒ Measurement error
ƒ Ambient temperature /
humidity

ƒ Poor Batch of Material

ƒ Special Causes (Assignable) ƒ Inexperienced Operator
ƒ 15% of Variation (Few Large ƒ Out of Date Drawings
Problems)
ƒ Unpredictable
15% ƒ Tool Damage
ƒ Maintenance Check overlooked
ƒ Easily Detected & Corrected ƒ Misread Drawing / Planning
Instruction
ƒ Machine Breakdown
59
Step 4: Identify root cause(s)

ƒ Action Plan
ƒ Eliminate Special Cause Variation
ƒ Identify when it happens
ƒ Identify root causes Stabilise Process
ƒ Eliminate root causes
ƒ Reduce Common Cause Variation
ƒ Identify amount of variation
ƒ Establish if it is excessive Control variations
ƒ Identify root causes

60
Step 5: Define and select root
cause corrective actions
• Objective: To define, prioritise and select corrective actions
that must be implemented to address the root causes and
prevent the undesirable condition, situation, nonconformity
or failure from recurring.
• Output: Root cause corrective actions defined
• WHAT: To ensure corrective actions addressing the most likely or critical root
causes are taken, considering operational and business constraints (costs, lead
time, difficulty of implementation, resources)
• WHY: Often the problem is not durably solved because wrong actions are taken
based on insufficient relation between the causes, the effects and the corrective
actions.
• WHO: All the team members.
WHEN: When all root and contributing causes have been identified and their
effect understood
61
Step 5: Define and select root
cause corrective actions
HOW:
• Based on results of risk analysis (Step 4 outcomes), identify solutions for
each selected root cause
• Determine for each solution :
• The probability of correcting the cause
• The risk of creating a new or worse problem
• The difficulty of implementation (practicality, time scale, costs, return
on investment, etc)
• The stability over time
• How to verify implementation and effectiveness
• Clear ownership (in particular if they have to be implemented internally
or externally)
• Select solutions that optimise value and effectiveness for all stake holders

62
Step 5: Define and select root
cause corrective actions
• COMMUNICATION:
– Continual communication between all team members is mandatory, for
instance through regular reviews until root cause corrective actions are
clearly identified and agreed by all
– Root causes corrective actions must be communicated to all stake holders
and agreed, especially by the customer when he is impacted
– Communication internally and between various tier levels

• TO BE CONSIDERED:
• When contributing causes can be easily minimised or removed,
documented action should also be taken to reduce their impact

63
Step 6: Implement root cause corrective
actions and check effectiveness
• Objective: To ensure all selected actions are implemented as
defined and to assess their effectiveness in permanently
preventing the undesirable condition from recurring.
• Output: Root cause corrective actions implemented and
efficiency demonstrated (problem durably solved)
• WHAT: Implement the solutions that have been selected, verify that all actions
have been completed to schedule and that they have prevented and will continue to
prevent the undesirable condition, situation, nonconformity or failure from recurring
• WHY: Too often, at this step of the corrective action process, focus is only on the
implementation of the easiest and quickest solutions and there is no plan to verify
timely implementation of the actions and their overall effectiveness, resulting in
corrective action plan to be only partially implemented.
• WHEN: As soon as actions have been prioritised and selected (i.e. when decision
has been made to launch the plan) and until effectiveness has been demonstrated
and agreed by all stake holders including the customer
64
Step 6: Implement root cause corrective
actions and check effectiveness
• WHO: The owner of each action to implement it as defined and with other team
members including when relevant the customer, to verify effectiveness
Note: At this stage, key role of team leader is to ensure that the entire corrective
action plan (solutions) has been implemented in due time and is effective
• HOW:
• Identify action owners (individuals, not functions) and due dates
• Get commitment from all action owners and ensure they agree with the plan
• Plan detailed actions (e.g. issue detailed working instructions, order parts or
tools, etc…)
• Establish a review process to ensure actions are completed to the plan and
will continue to be effective over time (e.g. by performing a product or
process audit) )
• Process confirmation (confirming you have done what you have planned)
• Definition of type and number or frequency of additional checks and audits

65
Step 6: Implement root cause corrective
actions and check effectiveness
• HOW (continued)
• Identify measures required to verify effectiveness of actions (who, what,
where, frequency, conditions,…)
• Measure and analyse new performance as planned and compare results
with performance measured at steps 2 and 4
• Verify effectiveness of the solutions
• If the root cause corrective action is not effective, return to step 4 (identify root
causes) and revisit the analysis process to check if the failure was the root
causes identification and/or the development of the solutions
• If they are effective, evaluate which containment actions may be eliminated (e.g.
stop over inspection and over production, return to normal transportation means,
etc…) without adversely affecting the product and process output.
• Record evidence of actions completed and associated results (what works
and what does not)

66
Step 6: Implement root cause corrective
actions and check effectiveness
• COMMUNICATION:
• Ensure feedback of all information between each action owner, the team
leader and the customer as required
• Content and frequency of reviews and status reports shall be defined between
the team leader and the stakeholders
• Escalation to management and the customer in case of implementation
difficulties or failures
• TO BE CONSIDERED:
• At this step, the team leader has to check that the composition of the team is
still appropriate. It is likely that new members will be invited to join the team
(new action owners, people in charge of implementing and verifying the
effectiveness of the corrective actions, etc…) and some may leave.
• When actions have to be implemented/verified in other business areas,
production lines, factories or suppliers, members of this organisation shall
become part of the team
67
Step 6: Implement root cause corrective
actions and check effectiveness
• TO BE CONSIDERED (continued):
• Consideration should be made, when validating the improved process, to the
way possible mistakes will or will not be detected.
• Source detection: The process detects the error making it impossible to do wrong.
• Self detection: The process points out the error during the operation.
• Successive detection: A later process detects the error after the operation
• Detection not possible further in the process (e.g. Special Process)

68
Step 7: Standardise and transfer
the knowledge across business
• Objective: To document analysis, results and changes, to
capture and share learning with all the stakeholders to
prevent similar undesirable condition, situation,
nonconformity or failure occurring on other products,
production lines, factories or suppliers
• Output: Occurrence of similar problems prevented (system
improved)
• WHAT: Formalise and standardise decisions made and actions completed
throughout the whole process and develop effective knowledge management
to transfer ideas, lessons learned, best practices, etc to all other stake holders
and other similar production lines, factories or suppliers that may require
similar actions to be implemented
• WHY: As product lines, people, processes and systems change, the same or
similar issue will recur if introduced changes are not captured and frozen.
Moreover, if an undesirable condition, situation, nonconformity or failure
happened somewhere, it is very likely that it will also happen in a similar
environment or on a similar or new product of the same nature. 69
Step 7: Standardise and transfer
the knowledge across business
• WHO: The team leader and all team members with involvement and support
of all levels of management
• WHEN: When actions have been successfully implemented and their
effectiveness is demonstrated
• HOW:
• Compile all documentation, instructions, analyses, flow charts, etc…
• Ensure all working documentation (working and inspection instructions,
purchase orders, etc…) is updated and available at every point of use
• Update skills matrices, training packages and deploy training accordingly
• Update IT systems and tools to support all design and process changes
• Capture and share knowledge
• Identify all other data, products, production lines, factories or suppliers that may
potentially be affected by same type of undesirable condition, situation,
nonconformity or failure (similar design, process, material source or supplier,
location, function or use, environment, training, machines and tools)

70
Step 7: Standardise and transfer
the knowledge across business
• HOW (continued):
• Identify all what can be shared from your experience and can be transferred
across these identified business units, production lines, factories or suppliers
• Get agreement from appropriate levels of management and other process
owners and functions (internally and externally) to launch actions and verify
there are implemented and effective
• Register lessons learned: Summary of content and results of analyses, flow
charts, data bases, performance data, main actions and decisions, location
where detailed data can be retrieved, difficulties encountered when
managing the issue, etc…
• COMMUNICATION:
• Inform all main process owners (internally and externally) of experience
gained (what worked, what failed and the associated reasons)
• Ask or encourage all staff and all functions to cascade information,
implement similar changes but checking possible adverse effects and
develop training in their area of activity
71
Step 7: Standardise and transfer
the knowledge across business
• TO BE CONSIDERED:
• When the decision is made to implement actions in other business areas,
production lines, factories or suppliers, which are not under direct control of
the team, implementation and the verification of effectiveness is not
necessarily the responsibility of the team: Escalation to top management or
transfer to another function (procurement, engineering, etc..) may be
required to ensure proper leverage and action follow-up

72
Step 8: Recognise and close the
team Champagne !
• Objective: To ensure all team members and stakeholders
are aware of the successful implementation of all actions,
to confirm that the activity is closed, and to recognise and
reward their work and accomplishment.
• Output: Synthesis of actions distributed and recorded,
team acknowledged and Root Cause Analysis process
closed
• WHAT: Confirm that all actions have been successfully implemented, record
synthesis of causes, actions and methodology, inform all those having been
affected by the undesirable condition, situation, nonconformity or failure that
the activity is complete, recognize those who have been involved in the
corrective action process and disband the team.
• WHY: Too often, action items are left open, diverting people from their main
roles, closed loop corrective action process is not achieved because there is
no feedback of actions and results to stake holders, and team efforts are not
recognised which negatively affects the dynamics of the root cause corrective
action culture. 73
Step 8: Recognise and close the
team Champagne !
• WHO: All the team and stake holders, especially the customer if he has been
involved
• WHEN: When all corrective actions have been implemented and their
effectiveness has been demonstrated internally and externally, and if relevant,
when knowledge has been transferred to other business units, production lines,
factories or suppliers.
• HOW:
• Check that all steps and all actions have been completed, in particular that::
• The undesirable condition, situation, nonconformity or failure has not recurred and
there is a high level of confidence it will never do
• All actions and decisions have been adequately documented and filed
• Lessons learned have been reviewed, in particular the way the problem solving and
root cause analysis processes were managed (successful actions and behaviour,
possible mistakes to be avoided in the future, etc)
• Ensure the customer has been informed and is satisfied by the final result
• Recognise team, inform all members that the team is now disbanded, and
celebrate success 74
Step 8: Recognise and close the
team Champagne !
• COMMUNICATION:
• Last information must be ensured that may include:
• Close out meeting
• Sending complete dossier to stake holders and customer
• Visual management board update
• TO BE CONSIDERED
• When the decision is made to implement actions in other business areas,
production lines, factories or suppliers, which are not under direct control of
the team, the team activity can be closed and the team can be disbanded
when:
• Proper information has been escalated to top management
• Decision to transfer implementation and the verification of effectiveness of
remaining actions to another function (procurement, engineering, etc..) has
been agreed by all
• The appropriate management has acknowledged that the remaining activities
are no longer the responsibility of the team
75
Contact

• For any question or comment, please

contact:

[email protected]

76