Professional Documents
Culture Documents
Problem Solving
1
Trainer: Carol Kurtz
2
Course Objectives
3
Corrective Action
4
Problem Solving
5
Corrective Action Origins
The origins of the 8-D system
actually goes back many
years.
The US Government first
‘standardized’ the system in
Mil-Std-1520 “Corrective
Action and Disposition
System for Nonconforming
Material”
Mil-Std-1520 - First released:
1974
Last Revision was C of 1986
Cancelled in 1995
6
What is 8D?
7
Why 8D?
8D is a structured approach to solving
problems
Fact Based
Data Collection & Analysis
Tests progress and results
Verify & Validate
Documented
History – An information database
Anticipate future problems
Prevent recurrence
8
8D Corrective Action
D4:
D5: Choose &
D0: Recognize the Identify Potential
Verify Corrective
Problem Causes
Actions
Yes
9
Six Sigma DMAIC
D1: Team Approach
D0: Recognize Problem
Define D2: Describe Problem
D3: Containment
Measure
10
Plan Do Check Act - PDCA
D1: Team Approach
Plan: D0: Recognize Problem
Identify the Problem D2: Describe Problem
Analyze The Problem D3: Containment
Do:
Develop Solutions
Implement Solutions
D4: Define & Verify Root Causes
D5: Select & Validate Corrective Actions
Check: D6: Implement Corrective Actions
Evaluate Results
Achieve Desired Results?
Act:
Standardize Solution D7: Prevent Recurrence
11
Some Vocabulary
Problem Effect
Symptom Cause
Concern Special Cause
Root Problem Common Cause
Failure Mode Root Cause
12
Verification & Validation
Verification and Validation are often not
well understood. Verification and
Validation work together as a sort of
‘before’ (Verification) and ‘after’
(Validation) proof.
Verification provides ‘insurance’ at a point in
time that the action will do what it is intended
to do without causing another problem.
Predictive.
Validation provides measurable ‘evidence’
over time that the action worked properly.
13
Verification & Validation
14
Structure of a Problem
15
Structure of a Problem
16
Structure of a Problem (continued)
Actual
Performance
17
Structure of a Problem (continued)
Positive change:
Established
Performance
Time
18
Classifying Problems
19
Classifying Problems – Type I
Plant Floor Problems
Rapid response is needed
Usually have discernable root causes
Usually require less data collection and analysis
Usually can be solved by local experts
Usually gradual or sudden problem structures
Special causes
Specific problem requiring Problem Analysis
8D methodology applies
20
Classifying Problems – Type II
Technical Problems
Permanent corrective actions are needed
Usually have difficult to discern root causes
Usually require more data collection and analysis
Usually require some technical expertise to solve
May be any problem structure
Special Causes
Specific problem requiring Problem Analysis
8D methodology applies
21
Classifying Problems – Type III
Process Improvement
Major systemic fixes needed
Multiple causes and effects
May require data collection and analysis
May need “systems thinking” to solve
Usually requires process owner’s involvement
Common cause problem
Structures include startup and positive. Others may apply.
Broad problems requiring a Situation Analysis
Quality Improvement Projects, Continual Improvement
Projects or other methodologies apply.
22
Problem Solving Tools
Trend Chart Situation Analysis
Control Chart Flowchart
Pareto Chart Failure Analysis
Brainstorm Database
Checksheet Decision Analysis
Histogram Action Plan
Nominal Group Root Cause Analysis
Technique Cause & Effect Diagram
Five Why’s Scatter Diagram
Computer Aided Design of Experiments
Engineering Poka Yoke
APQP Preventive Action Matrix
23
Problem Solving Tools
Tool Purpose 8D Step
Indicator to track magnitude of D1 D2 D3 D4 D5 D6
Trend Chart
symptoms D7 D8
Quantifier to prioritize and
Pareto Chart D2 D8
subdivide the problems
Indicator to monitor and
Paynter Chart D2 D3 D6 D8
validate the problems
Method to move from symptom
Repeated Why D2
to problem description
Process to find root cause
Information Database using Is/Is Not, Differences, D2 D4 D5 D6
Changes
Method to choose best action
Decision Making D3 D5
from among alternatives
Record of assignments, D1 D2 D3 D4 D5 D6
Action Plan
responsibilities and timing D7 D8
Report of problem solving
D1 D2 D3 D4 D5 D6
EW8D process for management
D7 D8
review
24
Problem Solving Tools Quiz
Trend Charts
Pareto
Analysis
RAC-
Root Cause
Analysis
? Problem
Solving
Tools
?
RAC-
Pareto
Root Cause
Analysis
Analysis
?
25
D0: Recognize the Symptoms
Rework
Trend Charts
What others?
26
D0: Recognize the Symptoms
Here? Supplier
In
Tran
Or Here? sit Or Here? Or Here?
Company
Receiving / Inventory /
In-Process
Inventory Shipping
In
Or Here? Tran
Or Here? sit Or Here?
Or Here? Customer
27
D0: Recognize the Symptoms
Trend Chart
A line graph plotting data over time.
Use to observe behavior over time
Provides a baseline and visual examination of
trends
No statistical analysis
Look for trends and patterns
Ask “Why?”
Good for operations/processes where data for
control charts is not available
28
D0: Recognize the Symptoms
Trend Chart
Average Hours Worked Per Employee (3rd shift)
45
44
43
42
41
40
39
38
37
36
35
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Change shift starting times
New entrance opened.
Task group established.
29
D0: Recognize the Symptoms
Nonconforming Product
Out of Control Conditions on Charts
Rework
Trend Charts
What others?
30
D0: Recognize the Symptoms
Control Chart
A line graph of a quality characteristic
that has been measured over time
Based on sample averages or individual
samples
Includes statistically determined Control
Limits.
Requires certain assumptions and
interpretation
31
Interpreting Control Charts
Control Charts provide information as to whether a process is being influenced by
Chance causes or Special causes. A process is said to be in Statistical Control
when all Special causes of variation have been removed and only Common
causes remain. This is evidenced on a Control Chart by the absence of points
beyond the Control Limits and by the absence of Non-Random Patterns or Trends
within the Control Limits. A process in Statistical Control indicates that production is
representative of the best the process can achieve with the materials, tools and
equipment provided. Further process improvement can only be made by reducing
variation due to Common causes, which generally means management taking action
to improve the system.
Average
32
Interpreting Control Charts
When Special causes of variation are affecting a process and making it unstable
and unreliable, the process is said to be Out Of Control. Special causes of variation
can be identified and eliminated thus improving the capability of the process and
quality of the product. Generally, Special causes can be eliminated by action from
someone directly connected with the process.
The following are some of the more common Out of Control patterns:
Change To
Machine Made
Tool Broke
Tool Wear?
Average
33
Interpreting Control Charts
Points Outside of Limits
Average
Trends
34
Interpreting Control Charts
Run of 7 ABOVE the Line
A Run of 7 successive points above or below the center line is an out of control
condition.
35
Interpreting Control Charts
Systematic Variables
36
Interpreting Control Charts
Freaks
Sudden, Unpredictable
Instability
37
Interpreting Control Charts
Mixtures
38
Interpreting Control Charts
Stratification
39
Control Chart Analysis Reaction
1. CHECK
Check that all calculations and plots have been accurately completed,
including those for control limits and means. When using variable charts,
check that the pair (x bar, and R bar) are consistent. When satisfied that
the data is accurate, act immediately.
40
Control Chart Analysis Reaction
2. INVESTIGATE
Investigate the process operation to determine the cause.
Use tools such as:
Brainstorming
Cause and Effect
Pareto Analysis
Your investigation should cover issues such as:
The method and tools for measurement
The staff involved (to identify any training needs
Time series, such as staff changes on particular days of the week
Changes in material
Machine wear and maintenance
Mixed samples from different people or machines
Incorrect data, mistakenly or otherwise
Changes in the environment (humidity etc.)
41
Control Chart Analysis Reaction
3.ACT
Decide on appropriate action and implement it.
Identify on the control chart
The cause of the problem
The action taken
As far as possible,eliminate the possibility of the special cause happening
again.
42
Control Chart Analysis Reaction
4. CONTINUE MONITORING
Plotting should continue against the existing limits
The effects of the process intervention should become visible. If not, it should
be investigated.
Where control chart analysis highlights an improvement in performance, the
effect should be researched in order that:
Its operation can become integral to the process
Its application can be applied to other processes where appropriate
Control limits should be recalculated when out of control periods for which
special causes have been found have been eliminated from the process.
The control limits are recalculated excluding the data plotted for the out of
control period. A suitable sample size is also necessary.
On completion of the recalculation, you will need to check that all plots lie
within the new limits
43
D0: Recognize the Symptoms
Other Indicators
Customer Concerns & Issues
Warranty Data
Quality Reports
Product Quality Planning
44
D1: Establish the Team
45
D1: Establish the Team
The 8D Team Members
Cross Functional or Multi-Disciplinary
Process Owner
Technical Expert
Others involved in the containment,
analysis, correction and prevention of
the problem
46
D1: Establish the Team
Team Roles
Several roles need to be established for the team. These roles are: Leader,
Champion, Record Keeper (Recorder), Participants and (if needed)
Facilitator.
Leader
Group member who ensures the group performs its duties and
responsibilities. Spokesperson, calls meetings, establishes meeting
time/duration and sets/directs agenda. Day-to-day authority, responsible for
overall coordination and assists the team in setting goals and objectives.
47
D1: Establish the Team
Problem Solver Characteristics
Persistent
Intuitive (supported by mechanical aptitude)
Logic & discipline
Common sense
Ability to balance priorities
Ownership
Inquisitive and willing
Creative and open minded
Needs proof & facts
48
D1: Establish the Team
Effective Team Characteristics
Leadership
Clearly define goals
Clearly defined responsibilities
Trust & Respect
Authority
Positive Atmosphere
Good two way communication
Effective action plan with timing
49
D1: Establish the Team
Management Responsibility
Provide time and resouces
Provide mentoring
Understand need for change
Recognize accomplishments & team
process
50
D1: Establish the Team
Brainstorming
Generate a great number of possible
solutions to a problem
Use to avoid conventional or in-the-box
thinking
Overcome mental blocks, inspire
creativity
Take advantage of team synergy
Ideas from different perspectives
51
D1: Establish the Team
Team Check List
Team Check List Yes No
Has a champion accepted responsibility for monitoring the measurables?
52
D2: Problem Description
53
D2: Problem Description
Problem Statement
Problem statement =
Object + concern + quantification
Example:
20% of Tuesday’s first shift production of
end cap #3245A have a ¼” to ¾” crack at
the lower left corner of the strain relief hole.
Remember: A well defined problem is
half solved!
54
D2: Problem Description
Five Why’s
A technique for stepping through successive layers of
symptoms to find the root problem statement.
Go to the point of occurrence of the problem (gemba)
Begin asking “Why?”
Using a flowchart, track back from symptom to
symptom until you find:
The root cause
A level where permanent corrective action can be
implemented
A point where “Why?” can no longer be answered
55
D2: Problem Description
Flowchart
Start or input at the beginning of
A picture of a
a process
sequence of the
steps. Yes Question or Decision? No
Completed process
56
D2: Problem Description
Situation Analysis
Tool used to break broad problems into
smaller prioritized pieces to attack one at
a time.
Many problem solving efforts start with
large, messy, poorly defined, unforcused
issues.
This method is detailed in the book The
New Rational Manager by Kepner &
Tregoe
57
D2: Problem Description
Pareto Analysis
A Pareto chart offers the following benefits:
Focuses on the problems or causes of
problems that have the greatest impact
Displays the relative significance of
problems or problem causes in a simple,
quick-to-interpret, visual format
Can be used repeatedly to produce
continuous improvements
58
D2: Problem Description
Pareto Analysis
Ball Lifting Percent Cum Percent
Frequency
Cause (%) (%)
Bonder Set-up
19 38% 38%
Issues
Unetched Glass
on Bond 11 22% 60%
Pad
Foreign Contam
on Bond 9 18% 78%
Pad
Excessive Probe
3 6% 84%
Damage
Silicon Dust on
2 4% 88%
Bond Pad
Corrosion 1 2% 90%
Cratering 1 2% 94%
Others 2 4% 100%
Total 50 100% -
59
D2: Problem Description
Paynter Chart
This chart is combination of Trend and
Pareto charts.
Provides information on actions taken
and shows effects.
Can be modified for Returns, Scrap,
Rework, etc.
60
D2: Problem Description
Paynter Chart
Number of 3rd shift workers affected
Problems: July Aug Sept Oct Nov Dec Total
Traffic jam on Hiway 90 84 4 3 0 90
Buses Late 30 30 9 8 30 30
% Late Employees Third Shift # Late Employees Not Enough Parking 17 16 17 8 0 17
100 60 Bad Weather 9 10 20 21 9 9
Road Construction 4 0 0 0 21 4
90 54 150 140 50 40 60 150
50 30
Average Hours Worked Per Employee (3rd shift)
40 24 45
44
30 18 43
42
20 12 41
40
10 6 39
38
0 0 37
1 2 3 36
Buses Late Bad Weather Not Enough Parking 35
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Change shift starting times
New entrance opened.
Task group established.
61
D2: Problem Description
Information Database
A tool for organizing all data about a
problem into four categories: What,
Where, When, Extent.
Used for Problem Analysis
Detailed in The New Rational Manager
by Kepner/Tregoe
62
D2: Problem Description
Information Database
Is Is Not
What: What is the object you are having a problem with? What could be happening but is not?
What is the problem concern? What could be the problem concern, but is not?
Where do you see the concern on the object? Be specific Where on the object is the problem NOT seen? Does the
Where:
in terms of inside to outside, end to end, etc. problem cover the entire object?
Where (geographically) can you take me to show me the Where else could you have observed the defective object,
problem? Where did you first see it? but did not?
When in time did you first notice the problem? Be as When in time could it have first been observed, buy was
When:
specific as you can about the day and time. not?
At what step in the process, life or operating cycle do you Where else in the process, life or operating cycle might
first see the problem? you have observed the problem, but did not?
Since you first saw it, what have you seen? Be specific
What other times could you have observed it but did not?
about minutes, hours, days, months. Can you plot trends?
How Big: How much of each object has the defect? How many objects could be defective, but aren't?
What is the trend? Has it leveled off? Has it gone away? Is What other trends could have been observed, but were
it getting worse? not?
How many objects have the defect? How many objects could have had the defect, but didn't?
How many defects do you see on each object? How many defects per object could be there, but are not?
How big is the defect in terms of people, time, $ and/or
How big could the defect be, but is not?
other resources?
What percent is the defect in relation to the problem?
63
D2: Problem Description
Checksheet
Checksheets are simple and effective
method of gathering information
on the job.
Ensures consistency of data collected.
Simplifies data collection and analysis.
Highlights trends.
Spots problems.
64
D2: Problem Description
Checksheet
Part Number 621532-B
Part Defect Checksheet
Date 12-16-04
Totals
31 18 5 54
65
D2: Problem Description
Histogram
Chart using bars of varying height to
show frequency distribution of some
characteristic.
Use for problem recognition, problem
definition, data analysis, and validation
of corrective actions.
Visually evaluate spread, centering,
capability.
66
D2: Problem Description
Histogram
23mm OD Histogram
P/N 543612 on Machine 6
6
5
Frequency
4
3
Frequency
2
1
0
19 21 23 25 27 or
e
M
Outside Diameter
67
D3: Containment
68
D3: Containment
Contain Symptom Flow
Immediate Containment Choose
with Current
Information and Verify Before
Problem Description Implement
Determine
Escape Point
Should an existing ‘check’ (control)
have caught the defect?
69
D3: Containment
Objectives
The objective of this step is to isolate the effects of the problem
by implementing containment actions.
Once a problem has been described, immediate actions are to be
taken to isolate the problem from the customer. In many cases
the customer must be notified of the problem.
These actions are typically ‘Band-aid’ fixes.
Common containment actions include:
100% sorting of components
Items inspected before shipment
Parts purchased from a supplier rather than manufactured in-
house
Tooling changed more frequently
Single source
70
D3: Containment
Containment Action Checksheet
Containment Action Checksheet Yes No
Has immediate containment action been taken to protect the customer?
Has the concern been stopped at each point in the process back to the source?
Have you verified that the action taken is FULLY effective?
Have you certified that parts no longer have the symptom?
Have you specially identified the 'certified' parts?
Have you validated the containment action?
Is data being collected in a form that will validate the effectiveness of the containment
action?
Has baseline data been collected for comparison?
Are responsibilities clear for all actions?
Have you ensured that implementation of the containment action will not create other
problems?
Have you coordinated the action plan with the customer?
71
D4: Determine Root Causes
72
D4: Determine Root Causes
Root Cause of Event (Occur or Occurrence)
What system allowed for the event to occur?
73
D4: Determine Root Causes
Root Cause Analysis – 5 Why’s
The 5 why's refers to the practice of
asking, five times, why a failure has
occurred in order to get to the root
cause/causes of the problem.
There can be more than one cause to a
problem as well.
This root cause analysis is often done by
a team with knowledge the problem
process or item.
74
D4: Determine Root Causes
Root Cause Analysis
Process of analyzing “is” & “is not” pairs
of information for differences and
changes that lead to root cause
75
D4: Determine Root Causes
Root Cause Analysis
Is Is Not
What:
Object Heavy traffic
Defect Late Employees
Where:
Seen on object I-70 Expressway
Seen geographically East bound I-70 near Main Street
When:
First seen July 7, 1996
When else seen Ever since
When seen in process (life cycle) Afternoon
How Big:
How many objects have the defect? Third shift (4:00PM)
How many defects per object? Once per day
What is the trend? Increasing --> SPECIAL CAUSE!
Enhanced Problem Description -->
76
D4: Determine Root Causes
Cause & Effect Diagram
Shows the relationship of causes and sub-causes to
an identified effect or problem. Clearly identify the
problem or effect to be diagrammed in the box at the
right
Draw the fishbone structure
Identify the major categories, factors, the causes
related to the effect.
Brainstorm, or note the causes of the problem that fall
within each of the major categories.
Each branch may have sub-branches, or sub-sub-
branches
As ideas are generated determine which branch of the
"bone" they should be placed
77
D4: Determine Root Causes
Cause & Effect Diagram
Effect
Method Materials
78
D4: Determine Root Causes
Scatter Diagram
Scatter diagrams are used to study possible
relationships between two variables. Although these
diagrams cannot prove that one variable causes the
other, they do indicate the existence of a relationship,
as well as the strength of that relationship.
A scatter diagram is composed of a horizontal axis
containing the measured values of one variable and a
vertical axis representing the measurements of the
other variable.
The purpose of the scatter diagram is to display what
happens to one variables when another variable is
changed. The diagram is used to test a theory that the
two variables are related. The type of relationship that
exits is indicated by the slope of the diagram.
79
D4: Determine Root Causes
Scatter Diagram
Strongly correlated
80
D4: Determine Root Causes
Scatter Diagram
Moderately correlated
81
D4: Determine Root Causes
Scatter Diagram
No Correlation
82
D4: Determine Root Causes
Design of Experiments - DOE
Shanin’s Red X Component Search
Taguchi’s Methods
Classical Design of Experiments
83
D5: Select & Verify CA
84
D5: Select & Verify CA
Poka Yoke
Poka Yoke Devices
Are Built within the Process
In General Have Low Cost
Have the Capacity for 100%
Inspection
Remember SQC is performed outside the
process which adds cost and allows defects
to escape the system.
85
D5: Select & Verify CA
Poka Yoke
Interference Fit
Poka Yoke
Orientation
Poka Yoke
86
D5: Select & Verify CA
Poka Yoke
Spring loaded
Floppy disks
have many 1 1 shutter
poka-yokes built
in. One example 6 4 6 4 6 mechanism - Do
you remember
is you cannot the old 5.25 inch
insert the disk floppies from the
into the drive early to mid-
completely if the 1980’s? Failsafe
disk is upside disk surface
down. This is 3 5 2 5 protection [#4].
because of the
corner notch Slide Tab to
[#1]. protect against
erasure.
720k disks have Mechanism
no hole [#2] senses [#5].
while HD disks
have hole ‘Precision’ alignment. Disk alignment holes and notches [#6] ensure the disk is
(mechanism properly aligned and also provides a ‘focus’ area for manufacturing.
senses)[#3].
87
D5: Select & Verify CA
Poka Yoke
Computer Files Computer Floppy
Drives
Microsoft: File type
identified by file Microsoft: Disk must
name suffix. If one be inserted and ejected
does not add the by hand. It is possible to
correct suffix, the Warning lights alert the
eject a disk while it is driver of potential
program the file is
being written to. problems. These
from will not
New lawn mowers are recognize it. devices employ a
required to have a Macintosh Poka Yoke warning method
safety bar on the Macintosh Poka (1984): Disk drive grabs instead of a control
handle that must be Yoke (1984): File disk as it is being method.
pulled back in order to type and creator inserted and draws it in
start the engine. If you application are and seats it. Disk
let go of the safety bar, identified and cannot be manually
the mower blade stops embedded in the first ejected. You must drag
in 3 seconds or less. part of every file. File the ‘desktop’ icon for the
This is an adaptation of name plays NO part disk to the ‘Trash’. The
the "dead man switch" in recognition by the drive then ejects the
from railroad originating program. disk as long as there
locomotives.
are no disk operations
taking place.
88
D5: Select & Verify CA
Corrective Action Check List
89
D6: Implement & Validate CA
Implementation can proceed when best
corrective action has been selected &
verified.
An effective implementation plan
reduces problems.
Validation is obtained by tracking
performance over time after
implementation to ensure the corrections
are permanent.
90
D6: Implement & Validate CA
Implementation Check List
Implement CA & Validate Over Time Yes No
Has the implementation plan been constructed to reflect Product
Development Process events and engineering change process?
Do the corrective actions make sense in relation to the cycle plan for the
products?
Have both Design and Process FMEAs been reviewed and revised as
required?
Have significant / safety / critical characteristics been reviewed and identified
for variable data analysis?
Do control plans include a reaction plan?
Is simultaneous engineering used to develop process sheets and
implement manufacturing change?
Is the Paynter Chart in place for validating data?
91
D7: Prevent System Recurrence
92
D7: Prevent System Recurrence
Prevent System Problems Check List
Prevent System Problems Check List Yes No
Have the system prevention practices, procedures & specification standards
that allowed the problem to occur and escape been identified?
Has a champion for system prevention practices been identified?
Does the team have the cross-functional expertise to implement the solution?
Has a person been identified who is responsible for implementing the system
preventive action?
Does the system preventive action address a large scale process in a
business, manufacturing or engineering system?
Does the system preventive action match root cause (occur & escape) of the
system failure?
Does the team utilize error proofing and successive checks on a proactive
on-going basis to eliminate the occurrence and escape of all defects?
Has a pieces over time (Paynter Chart) been used to indicate that the system
preventive actions are working?
Has the System Preventive Action been linked to the Product Development
phase?
93
D8: Congratulate the Team
94
D8: Congratulate the Team
Congratulate The Team Checksheet
95
References
http://elsmar.com/
http://www.isixsigma.com/spotlight/default.asp
http://www.isixsigma.com/dictionary/glossary.asp
http://www.asq.org/learn-about-quality/
Prince Corp, Corrective Action Manual
The New Rational Manager, Kepner & Tregoe
http://deming.eng.clemson.edu/pub/tutorials/
http://www.qualityspctools.com/menu.html
Ford Team Oriented Problem Solving
http://www.cjkurtz.com/qualitytools.htm
96