You are on page 1of 18

purify

“Have You Found


The ‘Root Cause’ Yet?”
protect - A New Spin to Root Cause Analysis

Gary Jing, PhD, MBB


transport
Nov 14, 2006

The Tie Between the Speaker & the Topic

ƒ MS in Reliability & Applied Statistics

ƒ PhD in IE

ƒ Editorial Review Board of Six Sigma Forum Magazine

ƒ Founding MBB at Seagate TCO, the world’


world’s largest disc drive
design center (1998 - 2005)

ƒ 2 patents in disc drive modeling generated from his Sigma work

ƒ Sr. Mgr., Global Lean Sigma / MBB, Entegris (2005 - present)

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 2
Personal Philosophy – Striving for Balance

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 3

Topics to be Covered

ƒ The Tie Between the Speaker & the Topic

ƒ The Mindset of Infinite Chain of Causation (New Spin 1)

ƒ The Role of Root Cause Analysis (RCA) in Six Sigma


Methodology

ƒ Popular RCA Tools Used in Six Sigma

– Differentiating & Prioritizing Causes (New Spin 2)

ƒ Summary

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 4
Cause-Effect Relationship
A PROBLEM WHICH
HAS OCCURED

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

n
E

atio
CO nates

TI V

CT e
rd fa
E li

pro FFE s th

evi
RR

o
AP
mi USE em
of

t
EC
C

i
AD
Li m
ap
A bl

mo
T IV
ro

ble
E
the

ACTION
Dave Wessel, “An Ounce of Prevention”, Quality Progress, Dec, 1998
Root Cause Analysis - ASQ
Gary Jing 11/14/06 Slide 5

Cause-Effect Principle – Chain of Causation

ƒ The same thing is both a cause and an effect at the same time.
ƒ Causes and effects are part of an infinite continuum of causes.

EFFECTS CAUSES
Injury caused by Fall
Fall caused by Wet Surface
Wet Surface caused by Leaky Valve
Leaky Valve caused by Seal Failure
Seal failure caused by Poor Maintenance
...
5 Why’s Reference: Apollo Root Cause Analysis, Dean L. Gano

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 6
The Concept of Infinite Chain of Causation

ƒ Widely seen in philosophical level of discussion. e.g., Hegel’s philosophy.


ƒ “An important feature of causality is the continuity of the cause-effect
connection. The chain of causal connections has neither beginning nor end.
It is never broken... And no one can say where this chain began or where it
ends. It is as infinite as the universe itself. There can be neither any first (that
is to say, causeless) cause nor any final (i.e., inconsequential) effect. If we
were to admit the existence of a first cause we should break the law of the
conservation of matter and motion. And any attempt to find an ‘absolutely
first’ or ‘absolutely final’ cause is a futile occupation, which psychologically
assumes a belief in miracles.” - A. Spirkin, Dialectical Materialism
ƒ Applied by Dean L. Gano in Apollo Root Cause Analysis - A New Way Of
Thinking, He warned that “because this book challenges conventional
wisdom, it may not validate your existing belief system". -

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 7

Root Cause Analysis (RCA)

ƒ “Have you found ‘the root cause’?”


– The term “root cause” is somewhat misleading
– A so-called “root cause” is subjectively chosen to serve as
the “root cause”
ƒ RCA Theme
– Continuously migrate the focus to upstream causes
– Keep asking “why” like a 2-year old

ƒ Challenge!!!: With infinite layers of Cause and Effect, how do we


know where to stop and conclude we have reached “Root Cause”?
– Case study 1: Jefferson Memorial
Root Cause Analysis - ASQ
Gary Jing 11/14/06 Slide 8
So, Where to Stop?
- To conclude we have reached “Root Cause”?

ƒ Usually the lowest cause in a chain at which we


have some capability to cause the break is chosen
as the root cause.
ƒ The lowest level at which it is within our capability to
unilaterally control, or to influence, changes to the
cause

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 9

Three Important Principles . .

Where to stop?
- The trick is not to find the “true”
true” root cause per se.
- The trick is where to stop and treat what you have as
the root cause. That wisdom is what differentiates an
adult from a 2-
2-year old; an expert from a novice.
– The bottom-line is High ROI – low cost with high return.
- Three tips:
– Leverage Point Thinking
– Pareto Principle
– Span of Control / Sphere of Influence

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 10
Leverage Point Thinking & Pareto
How Do You Turn the Ship?
Knowing the leverage
points is critical for
determining priorities
and strategies for
improvement

Case study 2: The “Snow Map”

Vital Few Pareto Principle 80% of


the problems are caused
by only about 20% of the
Trivial Many contributing factors.

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 11

Span of Control / Sphere of Influence


Before we begin, we must establish the context in
which the Cause-Effect will be used.
Span of Control - areas where we
have a high degree of control over SPHERE OF INFLUENCE
parts or functions, virtually (Influence or persuasion only)
complete authority to change
anything
SPAN OF CONTROL
(Full authority)
Sphere of Influence - areas
where we can influence things to
varying degrees but don’t have
direct control.
OUTSIDE, UNCONTROLLED
Outside Environment - where we
ENVIRONMENT
have neither control nor influence

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 12
The Role of RCA in Six Sigma Methodology
- DMAIC Process (Roadmap)
Define Define the problem;
Establish project charter (goals, scope, team & timeline)
Measure Identify customer requirements / Critical To Quality (CTQ’s)
Translate into measurable Key Process Output Variables (KPOV’s)
Verify measurement system
Establish present capability
Identify any factors / Key Process Input Variables (KPIV’s) that may
potentially affect the outputs (KPOV’s) through process mapping
Establish high potential suspects (hypotheses) through cause-effect
analysis (subjective analysis using experience & expertise)
Analyze Validate the suspects / hypotheses using existing data through
statistical testing (objective analysis)
Reduce list of high potential KPIV’s to vital few
Improve Generate improvement ideas using soft tools (subjective)
Use DOE to generate new data when no existing data available and
to optimize & validate outputs (objective)
Control Put in place permanent controls
Root Cause Analysis - ASQ
Gary Jing 11/14/06 Slide 13

Five Types of Cause-Effect Analyses

1. Is / Is not Comparative Analysis


2. 5 - Whys
3. Fishbone Diagram (Cause-effect Diagram) - Traditional
approach to brainstorming and diagramming Cause-Effect
relationships. Good tool when there is one primary effect
being analyzed.
4. Cause & Effect Matrix - Used to relate and prioritize X’s to
customer CTQ’s through numerical ranking. Good tool
when there is more than one primary effect being
analyzed.
5. Root Cause Tree - A problem analysis diagram that begins
with undesirable effects and works back to root causes
and core problems.
Root Cause Analysis - ASQ
Gary Jing 11/14/06 Slide 14
Cause-Effect Analysis

I keep six honest serving men,

They taught me all I knew,

Their names are What and Why and How

and Where and When and Who


Rudyard Kipling

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 15

Is / Is Not Comparative Analysis


Problem Statement:
Is Is Not Differences
and Changes
What What is the specific What similar objects could
object has the defect? have the defect but do not?
What is the specific What other defects could be
defect? observed but are not?

Where Geographically?
Physically on the part?
When When was the defect
first observed? When Where, when and
since then? When in the what size could the
product life cycle?
defect have been
Size How many objects with
the defect? How many but it was not?
defects per object? Size
of defect? Trend?

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 16
Fish Bone Diagram

- Focusing on single effect with simple


cause-effect relationships

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 17

“Fishbone” Cause-Effect Diagram


Advantages Drawbacks
ƒ Helps organize and relate factors ƒ Might become very complex
ƒ Provides a structure for brainstorming ƒ Requires patience
ƒ Does not rank the causes in an if- then
ƒ Involves everyone manner

Materials Methods C/N/X C = Control Factor


N = Noise Factor
C C X = Factor for DOE (chosen later)

N N
Problem/
Desired
N N Improvement
N
C
C
C New spin 2: Categorizing / labeling Causes:
Machinery Manpower
Opportunity: Controllability:
H: High C: Within control
M: Medium I: Have influence
L: Low N: No influence

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 18
Product: The 5M’s

Man

Machine Methods
Five Key
Sources of + Environment
Variation

Materials Measurement

Use
UseCause
CauseandandEffect
EffectDiagram
Diagramto toSingle
SingleOut
Out
Variation
VariationSources
Sourceswithin
withinthe
the“5M’s
“5M’s++E”E”

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 19

Transactional or Process: 4P’s + M&E

People

Policies Procedures
Five Key
Sources of + Environment
Variation

Place Measurement

You
Youcan
canmerge
mergethe
thecause
causeand
andeffect
effectwith
withthe
theaffinity
affinity
diagram
diagrambybyplacing
placingyour
yourPost-Its
Post-Itsdirectly
directlyonto
ontothe
thefish.
fish.
This
Thishelps
helpsreduce
reducethe
thenumber
numberofofheadings
headingsgenerated.
generated.

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 20
Fishbone Example #2
AGB project: Inv e ntor y Adjus tme nts New spin 2:
Retu r n s P r o c ed u res C y c le C o u n ts
Differentiate
causes.

G o o ds issue no t done B a d T y ping


w o rk orde r
v a ria nce
o v e r sta te d inv e nto ry N o D a ta a v a ila ble

R
eq
u
is
i ti
e xce ss

o
ns
inv e nto ry Line shut do w n w rong co unts
H ig h
N u mb er o f
in v en to r y
ad ju stmen ts

o
O

rd
w
p

e
e

rk

rs
A ccura cy o f co unts

n
em
N o I nv e nto ry by inco rre ct

pl
lo ca tio n qua ntity

o
T

ye
e
m

es
La ck o f tra ining

p
La ck o f S pa ce w ro ng pa rts
re ce iv e d N o t be ing do ne

E n v ir o n men t Rec eip ts D ata E n tr y

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 21

Weaknesses of the Fishbone

Although the “fishbone” cause-effect diagram has been very


popular over the years, it has some inherent weaknesses:

ƒ _______________________________________

ƒ _______________________________________

ƒ _______________________________________

ƒ _______________________________________

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 22
Cause and Effect Matrix

- To evaluate the impact of a group of


causes on a group of effects

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 23

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
12
C&E Matrix.xls 0
0
13 0
14 0
15 0
16 0
17 0
18 0
19 0
20 0
0
0

Total
Lower Spec
Target
Upper Spec

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 24
C-E Matrix: Example
Customer
CustomerECO
ECOThroughput
ThroughputReduction
Reduction

Rating of Importance
10 6 9 8
to Customer

1 2 3 4 5 6

Implementation
TOTAL %
me it’s

Commonality
Someti

(accuracy)
PROCESS INPUTS

Efficiency
rank
used to puts.

Change
ne in

Yield
Fishbo
1 Customer Input 8 6 8 8 252 11.36%
2 Equipment Specs 8 5 10 8 264 11.90%
3 Bill of Materials 7 5 10 5 230 10.37%
4 # of Revisions 8 6 10 8 270 12.17%
Label
5 Documentation 8 2 8 9 236 10.64%
6 CC Drawings 5 3 2 2 102 4.60%
7 Pre-CCP Metting 5 5 2 2 114 5.14%
8 Ownership 8 10 8 8 276 12.44%
9 Approval Cycles 8 8 5 8 237 10.69%
10 AMK Delays 8 8 5 8 237 10.69%
2218

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 25

Cause-Effect Matrix Analysis Example


Output Variables that Reflect the Impact of Adjustment AGB project: Inventory Accuracy
Weight 0.1 0.5 1 3 5 1 -2 1 -2 Rank
# of Substrac Rank
adjustm Substra Added ted Added Substra without
ent Added cted Substra quantity/ quantity/ $/ cted $ / Consum Rank by
entry / quantity quantity Added $ cted $ / consum consum consum consum ption Substra Overall
Material Total / total / total / Total Total Overall ption ption ption ption Info cted $ Rank
830R25 0.001 0.000 0.001 0.000 0.027 0.138 -2.840 -2.840 11 4 1
225 0.001 0.000 0.036 0.000 0.013 0.102 0.028 -2.807 0.028 -2.807 18 17 2
7500-0585-03 0.007 0.020 0.037 0.003 0.004 0.076 0.358 -0.317 0.357 -0.316 28 51 3
01-001638 0.004 0.000 0.003 0.042 0.184 1.050 0.011 -0.036 0.011 -0.036 1 1 4
01-006154 0.005 0.004 0.010 0.008 0.014 0.108 0.183 -0.244 0.157 -0.209 15 15 5
4002-5973-02 0.002 0.002 0.006 0.004 0.012 0.081 0.102 -0.202 0.102 -0.202 24 20 6
Input Variables - Materials

01-004488 0.003 0.001 0.005 0.009 0.061 0.337 0.027 -0.134 0.027 -0.134 2 2 7
8100-0100-01 0.012 0.009 0.000 0.067 0.000 0.206 0.564 0.564 7 403 8
200 0.003 0.009 0.034 0.005 0.013 0.117 0.071 -0.135 0.071 -0.135 13 18 9
7500-0585-07 0.003 0.024 0.017 0.008 0.004 0.071 0.258 -0.092 0.258 -0.092 30 52 10
1032-047 0.018 0.052 0.000 0.014 0.000 0.069 0.570 0.569 32 403 11
PKG-782 0.017 0.131 0.000 0.009 0.000 0.096 0.530 0.530 20 403 12
01-009871 0.002 0.002 0.005 0.018 0.030 0.212 0.072 -0.083 0.071 -0.084 6 3 13
01-009644-17C02 0.001 0.000 0.001 0.006 0.017 0.103 0.047 -0.107 0.046 -0.103 17 11 14
9700-7036-01 0.003 0.004 0.002 0.049 0.018 0.241 0.141 -0.037 0.141 -0.037 4 8 15
01-001651 0.003 0.003 0.005 0.015 0.020 0.152 0.073 -0.070 0.073 -0.070 9 7 16
160 0.001 0.000 0.001 0.000 0.016 0.079 -0.095 -0.094 25 13 17
01-018514 0.016 0.034 0.000 0.073 0.000 0.238 0.203 0.203 5 403 18
01-021524 0.001 0.001 0.003 0.008 0.016 0.109 0.062 -0.069 0.047 -0.071 14 12 19
01-009897 0.003 0.002 0.008 0.003 0.011 0.076 0.029 -0.076 0.029 -0.077 27 21 20
01-021554 0.001 0.000 0.002 0.001 0.012 0.069 0.013 -0.073 0.014 -0.087 31 19 21

Sometime the ranking can be objective, data driven.


Root Cause Analysis - ASQ
Gary Jing 11/14/06 Slide 26
Root Cause Tree (RCT)

- For more complicated cause - effect relationships


(causes may be dependent on each other)

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 27

Example of RCT
AGB project: Potting Leaks

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 28
Distinguishing Undesirable Effects From Neutrals

Two “Rules of Thumb”


1. What bearing does the effect have on the system goal?
2. Does the effect pass the “So What” test?

ƒ UDE’s are negative on their own merit


ƒ No further explanation why is needed
ƒ Effects are negative at face value
“We lost money last quarter”
*Undesirable Effects

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 29

Rules, Guidelines

Purpose: To help us construct our own logical


relationships and to evaluate the logic of others

Clarity: Someone stating a cause-effect relationship


must take care to ensure that others will:
– Understand the individual words
– Comprehend the meaning of the idea
– See a clear connection between cause and effect

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 30
Rules, Guidelines
Causality Existence
– Tests the connections
– One basic criterion: Does the proposed cause really produce the
observed effect?
– The Cause-Effect relationship must be direct and unavoidable
– If not clear - keep looking, you missed something

I can’t buy I can’t buy


a new car a new car

I have less
? disposable income

Taxes are Taxes are


raised raised

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 31

Rules, Guidelines
Cause Insufficiency
– Be sure you have identified and included all of the major contributing
causes
– To be contributing, causes must depend on one another
– Two questions:
• Is the cause alone enough to deliver the effect?
• If I take away one of the causes, will I still have the effect?

My book gets wet My book gets wet

“AND”

I leave my I leave my It rains


book outdoors book outdoors today

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 32
Rules, Guidelines

Additional Cause
– Consider all possible independent causes
– A cause may consist of several independent causes, any of
which would have produced the effect
– Each independent cause would increase the magnitude of
the observed effect
Accounts receivable
are too high

Payments Bills are Bills are


are late ignored partially paid

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 33

Rules, Guidelines
Circular reasoning
The effect is the rationale for the existence of the cause
– More common in conversation than on graphs
– Example:
• The fisherman brought home several large fish
• WHY? - The fishing was good
• HOW DO YOU KNOW? - He brought home several large fish, didn’t he?

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 34
Example RCT
(SRA – Sulfur related anion) Customer suffered a corrosion Opportunity: Controllability: Color code:
problem, meantime they tested a H: High C: Within control • Being
surge of sulfate with our product. They M: Medium I: Have influence studied
suspect it’s the primary contributor and L: Low N: No influence • Should
demand for the root cause and control. investigate

A. Understand the B. Internal C. SRA D. SRA introduced E. SRA introduced


meaning and “reliability” measurement of originated from during manufacturing after manufacturing
of customer data Sulfur related anion original material process

What’s the The readings B1. Where do C1. Lot SRA transform Cross E1.
repeatability are the we measure to lot during contamination Product
and calculated SRA, R&R. H,C variation. manufacturing. when lines absorb
reproducibility Sulfate H,N M,I shift products. SRA
of customer concentration L,I overtime.
data. L,N of product / H,I
resin? B7.
(surface B2. Resin
Sample Bag
only?) or the B3. Test test. M,I During
What’s the size. M,I contamination.
Correlation sulfate prime molding. L,C
between concentration product? L,I SRA
our data of test H,C transform
During Contamination
and their coupon / Cleaning. overtime.
B6. Air Exposing through leaking M,N
data. L,N water? L,N L,C to air. M,N
B4. Test quality bag. L,I
cleaned . M,C
product?
Correlation doesn’t mean H,C Resin absorb
Outside
causation. Is Sulfate really the cleanroom SRA from Product
primary contributor? air. M,C Cleaning release
(Otherwise it may not help the B5. Test / removes / SRA
problem.) – Need to work with monitor cleaning Inside Parts absorb SRA from absorb overtime.
customer and their data. M,I solution? H,C cleanroom air inside cleanroom. M,C SRA. M,C L,I

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 35

Summary

ƒ New Spin 1: The Mindset of Infinite Chain of Causation

ƒ The Role of Root Cause Analysis (RCA) in Six Sigma


Methodology

ƒ Popular RCA Tools Used in Six Sigma


– Is / Is not analysis
– Fishbone Diagrams (Cause-effect Diagram)
– C&E Matrix
– Root Cause Tree

ƒ New Spin 2: Differentiating & prioritizing causes

Root Cause Analysis - ASQ


Gary Jing 11/14/06 Slide 36

You might also like