3 Legged 5 Why Analysis
3 Legged 5 Why – Effective Root Cause
Analysis
“A Focused Approach to Solving Chronic
and/or Systemic Problems”
1
What is after Containment????
2
Agenda
When to Use 5 Why
3 Legged 5 Why Analysis
5 Why Examples
Resources and References
5 Why and Customer Problem Solving Formats
Where to Find the Blank Forms
3
When to Use 5 Why
Customer Issues
Required for all Covisint Problem Cases
May be requested for informal complaints
May be requested for warranty issues
Internal Issues (optional)
Quality System Audit Non-conformances
First Time Quality
Internal Quality Issue
4
When to Use 5 Why
5 Why Analysis can be used with various problem solving
formats
Internal Problem Solving
GM Drill Deep
Ford 8 D (Discipline)
Chrysler 8 Step
5 Why, when combined with other
problem solving methods, is a very
effective tool
5
3-Legged / 5-Why Form (Old Format)
Complaint Number: _______________
Issue Date: _____________
Define Problem
Use this path for the
specific nonconformance
ifi c
being investigated
Why?
ec Root Causes
Sp W
hy d
i dw
Why? e ha
ve t
Use this path to
Why? he p
n
investigate why the W robl
ct io
problem was not detected hy d
id tWhy? em?
t e he p A
De rob lem Why?
reac
Why?
h the c
Use this path usto
to investigate the Why?
W m er?
systemic root cause
hy d
mic id o Why?
ur sy
st e st e m
B
Sy
Why?
Why? a l lo w
it to
Why? occu
r ?
Why?
Why?
C
6
Problem Definition
New Format of 5 Why
if i c
pec
S
ion
c t
t e
De
mic
st e
Sy
5 Why Analysis
General Guidelines
A cross-functional team should be used to problem
solve
Don’t jump to conclusions or assume the answer is
obvious
Be absolutely objective
8
5 Why Analysis
General Guidelines
Ask “Why” until the root cause is uncovered
May be more than 5 Whys or less than 5 Whys
If you are using words like “because” or “due to” in any box, you will
likely need to move to the next Why box
Root cause can be turned “on” and “off”
Will addressing/correcting the “cause” prevent recurrence?
If not what is the next level of cause?
If you don’t ask enough “Whys”, you may end up with a “symptom” and
not “root cause”.
Corrective action for a symptom is not effective in eliminating the cause
Corrective action for a symptom is usually “detective”
Corrective action for a root cause can be “preventive”
Path should make sense when read in reverse using “therefore”
9
Problem Definition
New Format of 5 Why
if i c
pec
S
ion
c t
t e
De
mic
st e
Sy
10
Problem Definition
Define the problem
Problem statement clear and accurate
Problem defined as the customer sees it
Do not add “causes” into the problem statement
Examples:
GOOD: Customer received a part with a broken
mounting pad
NG: Customer received a part that was broken due to
improper machining
GOOD: Customer received a part that was leaking
NG: Customer received a part that was leaking due to
a missing seal
11
Problem Definition
New Format of 5 Why
if i c
pec
S
ion
c t
t e
De
mic
st e
Sy
12
Specific Problem
Specific Problem
Why did we have the specific non-conformance?
How was the non-conformance created?
Root cause is typically related to design, operations,
dimensional issues, etc.
Tooling wear /breaking
Set-up incorrect
Processing parameters incorrect
Part design issue
Typically traceable to /or controllable by the people doing
the work
13
Specific Problem
Specific Problem
Root Cause Examples
Parts damaged by shipping – dropped or stacked
incorrectly
Operator error – poorly trained or did not use proper tools
Changeover occurred – wrong parts used
Operator error – performed job in wrong sequence
Processing parameters changed
Excessive tool wear/breakage
Machine fault – machine stopped mid-cycle
14
Specific Problem
What if root cause is?
Operator did
not follow
instructions
Do we stop here?
15
Specific Problem
Operator did not
follow instructions
Do standard work Create a standard
instructions exist? instruction
Or do we attempt Is the operator
trained?
Train operator
to find the root
cause? Were work Create a system to
instructions assure conformity
correctly to instructions
followed?
Are work Modify instructions
instructions & check
effective? effectiveness
Do you have the right
person for this
job/task?
16
Specific Problem
Specific Problem
Column would not lock
in tilt position 2 and 4
Tilt shoe responsible for positions
2 and 4 would not engage pin
Shifter assembly screw lodged
below shoe preventing full
travel
THEREFORE
Screw fell off gun while
pallet was indexing
WHY??
Magnet on the screw bit
was weak
Exceeded the bits workable life
17
Specific Problem
Specific Problem
Loss of torque at rack inner tie
rod joint
Undersized chamfer (thread
length on rack)
Part shifted axially during
drill sequence THEREFORE
Insufficient radial clamping
load. Machining forces
WHY?? overcame clamp force
Air supply not maintained
Various leaks, high demand at full
plant capacity, bleeder hole plugs
caused pressure drop
18
Problem Definition
New Format of 5 Why
if i c
pec
S
ion
c t
t e
De
mic
st e
Sy
19
Detection
Detection:
Why did the problem reach the customer?
Why did we not detect the problem?
How did the controls fail?
Root Cause typically related to the inspection system
Error-proofing not effective
No inspection/quality gate
Measurement system issues
Typically traceable to /or controllable by the people
doing the work
20
Detection
Detection
Example Root Causes
No detection process in place – cannot be detected in our plant
Defect occurs during shipping
Detection method failed – sample size and frequency inadequate
Error proofing not working or bypassed
Gage not calibrated
21
Detection
Detection
Column would not lock
in tilt position 2 and 4
On-line test for tilt function is not
designed to catch this type of defect
THEREFORE
Test for tilt function is applied
before shifter assembly
Process flow designed in
this manner – would not
detect shifter assy screw
WHY?? lodged below tilt shoe
22
Detection
Detection
Loss of torque at rack
inner tie rod joint
Undersized chamfer/thread length
undetected
THEREFORE
Inspection frequency is
inadequate. Chamfer gage
is not robust
WHY??
Process CPK results did not reflect
special causes of variation affecting
chamfer.
23
Problem Definition
New Format of 5 Why
if i c
pec
S
ion
c t
t e
De
mic
st e
Sy
24
Systemic
Systemic
Why did our system allow it to occur?
What was the breakdown or weakness?
Why did the possibility exist for this to occur?
Root Cause typically related to management system issues or
quality system failures
Rework/repair not considered in process design
Lack of effective Preventive Maintenance system
Ineffective Advanced Product Quality Planning (FMEA, Control Plan)
Typically traceable to /controllable by Support People
Management
Purchasing
Engineering
Policies/Procedures
25
Systemic
Systemic
Helpful hint: The root cause of the specific problem leg
is typically a good place to start the systemic leg.
Root Cause Examples
Failure mode not on PFMEA – believed failure mode had zero
potential for occurrence
New process not properly evaluated
Process changed creating a new failure cause
PFMEA generic- not specific to the process
Severity of defect not understood by team
Occurrence ranking based on external failures only, not actual
defects
26
Systemic
Systemic
Column would not lock
in tilt position 2 and 4 Root Cause
Detection for tilt function done
prior to installation of shifter
assembly
THEREFORE
PFMEA did not identify a
dropped part interfering with
tilt function
WHY?? First time occurrence for this
failure mode
27
Systemic
Loss of torque at rack inner tie
Systemic Root Cause
rod joint
Ineffective control plan related to
process parameter control (chamfer)
THEREFORE
Low severity for chamfer control
Dimension was not
WHY?? considered an important
characteristic – additional
controls not required
Insufficient evaluation of
machining process and
related severity levels during
APQP process
28
Corrective Actions
Corrective Actions
Corrective action for each root cause
Corrective actions must be feasible
If Customer approval required for corrective action, this
must be addressed in the 5 why timing
Corrective actions address processes the “supplier”
owns
Corrective actions include documentation updates and
training as appropriate
29
Specific Problem
•Corrective Action:
•Reset alarm limits to sound if <90 PSI.
•Shah 10/12/10
Loss of torque at rack inner tie
•Disable machine if <90 PSI.
rod joint
•Jamil 9/28/10
•Dropped feed on drill cycle to .0058
Undersized chamfer (thread from .008.
length on rack) •Durai 10/10/10
•Clean collets on Kennefec @ PM
Part shifted axially during
frequency
drill sequence •Samy 10/12/10
•Added dedicated accumulator (air) for
system or compressor for each
Insufficient radial clamping Kennefec
load. Machining forces •Shah 10/12/10
WHY?? overcame clamp force •Verify system pressure at machines at
beginning , middle, and end of shift
Air supply not maintained •Shah 10/12/10
Various leaks, high demand at full
plant capacity, bleeder hole plugs
caused pressure drop
30
Detection
Corrective Action:
•Implement 100% sort for chamfer length and
Loss of torque at rack thread depth.
inner tie rod joint •Shah 9/26/10
•Create & maintain inspection sheet log to validate
Undersized chamfer/thread length •Durai 8/22/10
undetected
•Redesign chamfer gage to make more effective
•Jamil 11/30/10
Inspection frequency is
•Increase inspection frequency at machine from 2X
inadequate. Chamfer gage
is not robust per shift to 2X per hour
•Johari 10/14/10
•Review audit sheets to record data from both
Process CPK results did not ends on an hourly basis
reflect special causes of
•Durai 10/4/10
variation affecting chamfer.
•Conduct machine capability studies on thread
depth
•John 9/22/10
•Perform capability studies on chamfer diameters
•10/14/10
WHY?? •Repair/replace auto thread checking unit to
include thread length.
•10/18/10
31
Systemic
Loss of torque at rack
inner tie rod joint Corrective Action:
•Designrecord, FMEA, and Control Plan to be
reviewed/upgraded by Quality, Manufacturing Engineering
Ineffective control plan
related to process parameter •Update control plan to reflect 100% inspection of feature
control (chamfer)
•PM machine controls all utility/power/pressure
•Implement layered audit schedule by Management for
Low severity for chamfer robustness/compliance to standardized work
control
Lessons Learned:
Dimension was not •PFMEA severity should focus on affect to subsequent internal
considered an important process (immediate customer) as well as final customer
characteristic – additional •Measurement system and gage design standard should be
controls not required robust and supported by R & R studies
•Evaluatethe affect of utility interruptions to all machine
Insufficient evaluation of
machining process and processed (air/electric/gas)
related severity levels during
APQP process
WHY??
32
5-Why Critique Sheet
General Guidelines:
Don’t jump to conclusions..don’t assume the answer is
obvious
Be absolutely objective
A cross-functional team should complete the analysis
Step 1: Problem Statement
State the problem as the Customer sees it…do not
add “cause” to the problem statement
33
5-Why Critique Sheet
Step 2: Three Paths (Specific, Detection, Systemic)
There should be no leaps in logic
Ask Why as many times as needed. This may be fewer than 5 or
more than 5 Whys
There should be a cause and effect path from beginning to end of
each path. There should be data/evidence to prove the cause and
effect relationship
The path should make sense when read in reverse from cause to
cause – this is the “therefore” test (e.g. – did this, therefore this
happened)
The specific problem path should tie back to issues such as design,
operations, supplier issues, etc.
The detection path should tieback to issues such as control plans,
error-proofing, etc.
The systemic path should tie back to management systems/issues
such as change management, preventive maintenance, etc
34
5-Why Critique Sheet
Step 3: Corrective Actions
There should be a separate corrective action for each root
cause. If not, does it make sense that the corrective action
applies to more than one root cause?
The corrective action must be feasible
If corrective actions require Customer approval, does
timing include this?
Step 4: Lessons Learned
Document what should be communicated as Lessons
Learned
Within the plant
Across plants
At the supplier
At the Customer
Document completion of in-plant Look Across
(communication of Lessons Learned) and global Look
Across
35
5 Why Analysis Examples
Group Exercise
Review a 5 Why using the Critique Sheet and what you
have learned
Note: These are actual responses as sent to our Customers!
Has probable root cause been determined for:
Non-conformance leg
Detection leg
Systemic leg
Do corrective actions address root cause?
Have Lessons Learned/Look Across been noted?
If any above answers are “no”, what recommendations would you
make to the team working on the 3 Leg 5 Why?
36
Is this a good or bad “Specific” leg?
Missing o-ring
on part number
K10001J
WHY?
Parts missed the
o-ring installation
process
WHY?
Parts had to be Why did they
reworked have to rework?
WHY?
Operator did not return
parts to the proper process
step after rework
WHY?
No standard
rework
procedures exist
This is still a systemic failure
& needs to be addressed,
but it’s not the root cause.
37
Is this a good or bad “Detection” leg?
Missing threads
on fastener part
number LB123
WHY?
Did not detect What caused
threads were the sensor to
missing get damaged?
WHY?
Sensor to detect
thread presence
was not working
WHY?
Sensor was
damaged
WHY?
No system to
assure sensors
are
This is still a systemic failure working properly
& needs to be addressed,
but it’s not the root cause
of the lack of detection.
38
Summary of Key Points
When do you use it?
Use a cross-functional team
Never jump to conclusions
Ask “WHY” until you can turn it off
Use the “therefore” test for reverse path
Strong problem definition as the customer sees it
Specific Leg – Typically applies to people doing work
Detection Leg – Typically applies to people doing work
Systemic Leg - Typically applies to support people
Start with root cause of specific leg
Corrective
actions with date and owner
Document lessons learned and look across
39
40
