7 QC Tools for Effective Problem Solving
7 QC Tools for Effective Problem Solving
+ Plus
- minus
x multiply
divide
Customer is
demanding
Competition
This is a
sample
Increasing.
text.
Crisis
impending
Cost
rising
DEMING MODEL (PDCA)
[Link] [Link]
Implement Plan for
the plan improvement
[Link] [Link]
Test is the Implement as
plan per plan
working
DCA AND 8D PROBLEM SOLVING
Select Topic
P Understand Situation
L And Set Targets
A Containment Action
N
Analyse Causes
Standardise and
A Improve
C
T Congratulate Your Team
L S
OO
C T
Q
” 1 PARETO DIAGRAM
N T
I CE
N I F 2
G
CAUSE AND EFFECT
A
“M 3 CHECK SHEET
4 GRAPHS
5 HISTOGRAM
6 CONTROL CHART
7 FLOWCHART
7 QC TOOLS IN PDCA
Plan Do Check Action
No
P
Select Topic Propose
Pareto Data Collection YES
(Check Sheet) Causes Improvement Plan
Identified? (Flow chart)
Understand
situation and target Implement
(Process Flow) No Plan
• Histogram
• Graph D
Containment action Objectives Collect & Analyze
• Control Chart Met? Data
Requirements
Requirements
Inputs Activities Output
Feedback Feedback
Conformance
PROBLEM SOLVING
Present situation
FIRE-FIGHTING TO
IMPROVEMENT
100% PREVENT
PROBLEM IMPROVEMENT
PROJECTS
WORKING CONDUCTING
TIME ROUTINE
WORK PREVENT
PROBLEMS
50%
CONDUCTING
ATTENDING ROUTINE
BREAKDOWN WORK
EVENTS
ATTENDING
BREAKDOWNS
IDENTIFY PROBLEM
Management Brainstorming
decision
Customer
Suggestion
Complaint
from staff
HOW TO DESCRIBE TOPIC
Where?
(process,machine,operations
Product,region,work station, etc)
What?
(control of characteristics)
How?
(direction of improvement, level)
HOW TO SET TARGET
Target
Rateof
Rate ofOccurrence
Occurrence
Rate
Ofof Occurrence
Rim Warping
Of Rim Warping
What? Of problem
6.0%2.5%
6.0%2.5%
6.0%2.5%
By How Much
By When 3030
ththth Mar 2004
Mar
30 Mar2021
2004
Mark Twain
PARETO ANALYSIS
HISTORY
20% 80%
EFFORTS RESULTS
VITAL FEW TRIVIAL MANY
• In term of quality improvement, large majority of problem
(80%) are created by a few (20%) causes
WHAT IS PARETO CHART
• Pareto chart is a series of bars whose heights reflect the
frequency of impact of problem
• The bars are arranged in descending order of height from
left to right
• Bars on left are relatively more important that the bars on
the right
Pareto rule chart
Pareto chart
80% problem from
20% cause
Item No of Cases
• Account No 52
• Transaction No 23
• Spelling Name 38
• Telephone No 18
• Post code 9
• Address 11
• Date 87
• Basic data 4
• Others 6
TOTAL 248
STEP 1: ARRANGE DATA
Item No of cases
Date 87
Account No 52
Spelling Name 38
Transaction No 23
Telephone No 18
Address 11
Post code 9
Basic data 4
Others 6
TOTAL 248
STEP 2: CALCULATE THE
CUMULATIVE NUMBER
Item No of cases Cumulative
Number
Date 87 87
Account No 52 139
Spelling Name 38 177
Transaction No 23 200
Telephone No 18 218
Address 11 229
Post code 9 238
Basic data 4 242
Others 6 248
TOTAL 248
NO OF CASES
100
150
200
250
50
0
Date
Account No
Spelling Name
Transaction No
Telephone No
Address
Post Code
Basic Data
Others
STEP 3: DRAW THE BAR GRAPH
ITEM
NO OF CASES
100
150
200
250
50
0
Date
Account No
Spelling Name
Transaction No
Telephone No
Address
Post Code
Basic Data
Others
ITEM
STEP 4: DRAW THE CUMULATIVE CURVE
NO OF CASES
100
150
200
250
50
0
Date
Account No
Spelling Name
Transaction No
Telephone No
Address
Post Code
(0%-100%)
Basic Data
Others
STEP 5: DRAW VERTICAL AXIS
ITEM
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
STEP 6: DRAW 80% FROM RIGHT
VERTICAL AXIS TO FIND THE
CUMULATIVE NUMBER
250 100%
90%
200 80%
70%
NO OF CASES
150 60%
50%
100 40%
30%
50 20%
10%
0 0%
Telephone No
Post Code
Account No
Address
Spelling Name
Transaction No
Basic Data
Others
Date
ITEM
CONCLUSION
Problem
Root Cause
ANALYZE THE ROOT CAUSE
Completeness
Accuracy
Frequency
Content
Slow
Front Desk Computer
Check-In
Age
Capability
Maintenance
Machines Materials
CONSTRUCTING A CAUSE &
EFFECT DIAGRAM
• Identify the quality outcome to be discussed
• Agree on appropriate “headings” for each branch
• Brainstorm possible causes and classify
• Ask “why?” to define causes in more depth
• Use voting techniques to gain consensus on key
causes that contribute to the quality outcome
HOW SHOULD WE SOLVE A
PROBLEM?
Measurement
Why? Why?
Why?
Why? Why?
Five Whys?
5 WHY CAUSE INVESTIGATION
Abnormal Occurrence
Why?
(cause/effect relationship)
C C C C Direct Cause
Why?
(cause/effect relationship)
C C C C
Cause
Why?
(cause/effect relationship)
C C C Cause
Why?
(cause/effect relationship)
C C C Cause
Why?
Therefore
C C C Root Cause Test
CHECK SHEET
CHECK SHEET
• A sheet designed to collect data easily and
systematically
• to allow efficient checking of all items for
inspection and verification without omission
WHAT IS DATA COLLECTION?
Data Collection
is
obtaining useful information.
• Attribute (categorical)
– Data obtained by counting observations which
occur in particular categories
VARIABLE AND ATTRIBUTE
VARIABLE AND ATTRIBUTE
ACCURATE AND PRECISE
Decide what
data is
needed
No Develop data
available?
collection plan
Yes
No No
Valid? Collect data
Yes
Usable?
Yes
Analyse
Data
METHODS OF DATA COLLECTION
• Historical
• Direct observation
• Interview
• Survey
• Experiments
Checksheet
Scratches 21
Crack 23
Incomplete 30
Material 4
Others 8
Total 86
CREATING CHECK SHEET
1. Determine the purpose of data collection
2. Consider the method of stratification, eg by week
etc.
3. Design the sheet according to the purpose
4. Analyze the data collected
GRAPH
GRAPH
• A drawing showing the status/relationship of data
in order to understand the fact easily
• Typical kind of graph
– Bar chart …. compare volume of quantity
– Line graph…. show the time series change
– Pie chart…..show the ratio of each composition
– Belt chart……show the ratio and its change of
each composition
Bar chart Line graph
Kadar Kecacatan Barangan Bagi Bulan Disember
J u m la h K eca c a ta n B a r a n g a n P e n g elu a r a n 1993
J u m la h 140
K e ca c ata n
300 284 120
250
100
200
162
80
Jumlah
150 133
100 60
71
50 23 40
9 4 4
0
Smooth Mill 20
Pre- Finish
Laminate
Finishing
Main Mill
Rough Mill
Assembly
Stor Siap
0
S eksy e n 9 11131415161718202223283031
Haribulan
LAMINATE 29 66,354
1985 60 11
1%
FINISHING
3%
36 40 24 43,274
ASSEMBLY 1990
23%
1995 27 41 32 40,471
STORE SIAP
Am Ucapan Tahniah Ucapan Takziah
SMOOTH 1%
ROUGH MILL
42% MILL Graf ‘Belt’Bagi Penggunaan
Telegram
1%
WHAT IS A LINE
GRAPH?
A line graph that shows results of a process over time.
# Procedures Used
Number of Restrictive Procedures Used
Per Shift
10
8
Day
6
Evening
4 Night
2
0
M o nth
WHAT IS BAR GRAPH
Bar chart
J u m la h K eca c a ta n B a r a n g a n P e n g elu a r a n
J u m la h
K e ca c ata n
300 284
250
200
162
150 133
100 71
50 23
9 4 4
0
Pre- Finish
Laminate
Smooth Mill
Finishing
Main Mill
Rough Mill
Assembly
Stor Siap
S eksy e n
WHAT IS PIE CHART
Pie chart
MAIN MILL
19%
PRE-FINISH
10%
LAMINATE
1%
FINISHING
3%
ASSEMBLY
23%
STORE SIAP
SMOOTH 1%
ROUGH MILL
MILL
42%
1%
WHAT IS RADAR GRAPH
APPLICATION OF STATISTIC
COLLECTION
ANALYSIS DATA
INTERPRETATION
HISTOGRAM
WHAT ARE HISTOGRAMS
A simple, effective method of organizing an presenting
numerical data so that one can get an overall picture of where
measurements are concentrated and how spread out they are
25
20
Frequency
15
10
0
“GOAL POST” MENTALITY
LSL USL
A B C
LSL USL
TARGET
VALUE
A B C
WHAT ARE HISTOGRAMS
• dividing the existing range of data into certain
classes and then calculating the frequency of
occurrence data
• consist of bar chart that shows average, Upper
Specification Limit (USL) and Lower Specification
Limit (LSL)
COMPONENTS OF HISTOGRAM
25
20
Frequency
15
10
Class,k
Class width Class Boundary
ROLES OF HISTOGRAMS
• To make distributions easier to see and to allow
visual appreciation.
5
4
3
2
1
1.51 1.53 1.35 1.30 1.45 1.61 1.55 1.42 1.65 1.54
1.43 1.64 1.61 1.44 1.70 1.41 1.51 1.61 1.74 1.56
1.60 1.49 1.64 1.57 1.40 1.54 1.58 1.68 1.56 1.37
1.59 1.62 1.59 1.64 1.28 1.54 1.66 1.72 1.59 1.49
1.52 1.49 1.52 1.32 1.49 1.55 1.51 1.61 1.54 1.59
1.33 1.69 1.62 1.46 1.56 1.64 1.54 1.46 1.61 1.54
1.56 1.46 1.54 1.36 1.50 1.59 1.43 1.63 1.66 1.49
1.59 1.54 1.69 1.52 1.46 1.49 1.69 1.56 1.58 1.66
1.56 1.64 1.58 1.61 1.54 1.60 1.66 1.63 1.64 1.66
1.68 1.42 1.38 1.56 1.73 1.47 1.53 1.39 1.47 1.54
CONSTRUCTION OF HISTOGRAM
k = n
n = (number of data)
k =
= __________
CONSTRUCTION OF HISTOGRAM
Step 2: Find the largest and smallest value (L,S)
Data S L
1.51 1.53 1.35 1.30 1.45 1.61 1.55 1.42 1.65 1.54
1.43 1.64 1.61 1.44 1.70 1.41 1.51 1.61 1.74 1.56
1.60 1.49 1.64 1.57 1.40 1.54 1.58 1.68 1.56 1.37
1.59 1.62 1.59 1.64 1.28 1.54 1.66 1.72 1.59 1.49
1.52 1.49 1.52 1.32 1.49 1.55 1.51 1.61 1.54 1.59
1.33 1.69 1.62 1.46 1.56 1.64 1.54 1.46 1.61 1.54
1.56 1.46 1.54 1.36 1.50 1.59 1.43 1.63 1.66 1.49
1.59 1.54 1.69 1.52 1.46 1.49 1.69 1.56 1.58 1.66
1.56 1.64 1.58 1.61 1.54 1.60 1.66 1.63 1.64 1.66
1.68 1.42 1.38 1.56 1.73 1.47 1.53 1.39 1.47 1.54
CONSTRUCTION OF HISTOGRAM
Step 3: Determine the class width, w = (L-S)
k
w=( )
w = __________
L = Largest value
S = Smallest value
k = number of classes
CONSTRUCTION OF HISTOGRAM
Step 4: Prepare frequency table with the mark
No Class limit Frequency No of
freq
CONSTRUCTION OF HISTOGRAM
Step 5: Draw bars on graph.
Step 6: Draw average line, X , Upper
Specification Limit (USL) and Lower
Specification Limit (LSL)
CONSTRUCTION OF HISTOGRAM
FREQUENCY
ANALYSIS OF HISTOGRAM
• Location (distance of average value to target)
• Center of distribution (span of values from largest to smallest)
• Spread of distribution (whether it is symmetrical, skewed, etc)
LOCATION SHAPE
SPREAD
Figure
2-11 (b)
Not normal,bimodal
Distribution
- Special cause present
- Two shifts, two operators,
two fixtures, etc, likely
ANALYSIS OF HISTOGRAM
LSL USL
Normal, Off
centered
Distribution with
defectives at
upper spec
LSL USL
Normal, Off
centered
Distribution with
defectives at
lower spec
ANALYSIS OF HISTOGRAM
LSL USL
LSL USL
• Actual
T
• Improvement
LSL USL
T
LSL’ USL’
CONTROL CHART
CONTROL CHART
A Run Chart with statistically determined upper and lower control limit
lines drawn on either side of the process average.
_ _ _ _ _ _ _ Upper
_ _ _ Control
_ _ _ _Limit
_ _[UCL]
_________
Measurement
Average
_ _ _ _ _ _ _ _Lower
_ _ _Control
_ _ _ Limit
_ _ [LCL]
_________
Time
Control charts are widely used to:
• analyze variation from a process
• to determine whether or not a process is stable
(statistical control)
MEASUREMENT
divided by the sum of the observations. The control
limits are the process average plus or minus three
standard deviations.
MEASUREMENT
process average
process average
Lower Control Limit [LCL]
TIME
TIME
CONTRIBUTORS TO VARIATION
09:10
09:20
09:30
09:40
09:50
10:00
10:10
CAUSE
COMMON
10:20
10:30
10:40
10:50
11:00
11:10
11:20
11:30
11:40
11:50
12:00
COMMON AND SPECIAL CAUSES
12:10
SPECIAL CAUSE
TYPE OF CONTROL CHART
• Variable Control Chart
– X-bar Chart
– R Chart
• Attribute Control Chart
– P Chart
– C Chart
INTERPRETING CONTROL
CHARTS
Upper Control Limit
Average
Change To Machine
Made
Tool Broke
Tool Wear?
Average
Average
Trends
A Run of 7 successive points above or below the center line is an out of control condition.
Sudden, Unpredictable
Instability
Stratification
Change
New operator
supplier Old machine need to
be replaced
3:21
2:52
2:24
Individual Value
1:55
1:26
0:57
0:28
0:00
09:00
09:20
09:40
10:00
10:20
10:40
12:00
12:20
12:40
13:00
13:20
13:40
14:00
14:20
14:40
15:00
11:00
11:20
11:40
UCL UCL
UCL
LCL
LCL
Additional improvements
LCL Process centered made to the process
Process not centered and stable
and not stable
PROCESS IMPOVEMENT
25
Journey Time (hours)
20 Second
Changed Process
15 Process Change
10 Third ProcessFourth
Change Process
5 Change
0
Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul
Month
FLOW CHART
FLOWCHART
• A diagram of the steps in process. A flowchart can
help identifying possible points in a process where
problems occur.
BENEFITS OF FLOW CHARTS
• Understanding of process steps
• Understanding the interdependence of process
steps
• Helps build a complete picture of the process
• Helps identify sources of variation
FLOW CHART STRUCTURE
• Beginning activity (start)
• Activity descriptions
• Process flow (from one activity to another)
• Decision points
• Ending activity (stop)
FLOW CHART GUIDELINES
• Don’t be concerned about defining the process
completely in the first draft
• Involve people who are familiar with the process
• Review the flow chart for accuracy and
completeness
• Create the “as is” as opposed to “as should be”
FLOW CHART SYMBOLS
Start/End
Activity
Decision
Report
FLOWCHART
A pictorial representation of a process.
Rectangle for processes or procedures, diamond for decisions
points, arrows for flow of information or work flow,
REMOVE NON-VALUE ADDED
What You Think It Is... Start
Yes
Re-submitting Order
Re-negotiation
Signing Letter of
Intent
PROCESS ANALYSIS
Value
Process Steps (Yes/No) Reason Time
Customer willing to pay for that
Submission of offer Yes step. 1 minutes
Leaving the showroom as a way
to obtain my car did not add
Leaving the showroom No value to me, the customer 10 minutes
Usually anything that begins with
prefix "re-" is being done for the
Resubmitting offer No second time 5 minutes
No change or transformation in
Waiting for decision No the product/service 45 minutes
Renegotiation of option
package No There's that "re-" again 30 minutes
The customer (me ) is important.
Signing letter of intent Yes It physically changed something 15 minutes
APPS QR
CODE
Thank You
017-9362125
arazak@[Link]