7 QC TOOLS
What is it?
! A tool for collecting data in a consistent form.
! Provides an easy, structured way of recording data as it is collected.
! Assures data will be recorded in similar manner.
What is it?
What is Check Sheet used for?
! Controlling and monitoring the production process.
! Analysis of cause and effect properly.
! Inspection for non-conformance.
! To make data-gathering easy.
! To arrange data automatically so that they can be used easily later
on.
Based on the purpose, customized Check Sheets are designed.
Decide the purpose: In the example Check Sheet is being used for
inspection
Design the Format: Based on the requirement, the fields are
customized
3 Cause and Effect Diagram
Pareto Diagrams
4 Graphs and Charts (including Control Chart)
! Developed by Dr. Kaoru Ishikawa in 1943.
! Also known asIshikawa DiagramandFish-bone Diagramdue to its shape/structure.
! Developed by Vilfredo Pareto (1897), an Italian economist.
! He observed that 80 % of Italys wealth lay in the hands of 20
% of the population.
! Designed for :
Stimulating thinking during a brainstorm of potential causes.
! Used for prioritization by 80-20 rule.
Providing a structure to understand the relationships between many possible causes
of a problem.
! Differentiates Vital Few from Useful Many (Juran).
Serving as a visual display of causes that have been studied.
! Left vertical axis depicts the actual frequency of items while
the right vertical axis denotes cumulative percentage.
Special Notes:
! Make the same number of cause & effect diagram as that of characteristics.
Types of Graphs and Charts
Benefits:
! Error in weight and length of the same product will have different cause & effect
structures and these should be analyzed in two separate diagrams.
! Write only the characteristic and the causal factors which are measurable.
! Useful in establishing priorities.
! Comparing Pareto charts of a given situation over time
determines whether an implemented solution reduced the
relative frequency or cost of that problem or cause.
The diagram shows that causes A, B & C together are
responsible for 80% of the defective items. Therefore, these
should be prioritized for problem solving. While comparing
Pareto Charts, the left vertical axis should have the same
scaling.
June 1 - July - 31
100
150
Useful Many
Number of defective items
80
Vital few
60
100
Total effect
Sept. 1 - Oct. 31
100
100
80
40
60
50
Effect
50
40
20
20
A
B C
D E Others
B A D E
Others
! After completing a cause & effect diagram, it is necessary to grasp the strength of the
cause & effect relationship objectively using data.
! Choose one quality characteristic and enclose in a box.
! Write the primary causes, which affect the quality characteristic as big bones also
enclosed by squares. Main branches must be independent.
! Write the causes (secondary causes) which affect the big bones (primary causes) and
write the causes
(tertiary causes)
w h i c h a f fe c t
the medium
sized bones as
small bones.
! Assign numbers
to indicate the
order in which
the important
fac tors seem
t o exe r t t h e
influence.
Line Graph
Pie Chart
Area Graph
Rader Chart
Bar Graph
Control Chart
Control Chart
Developed by Dr. W A Shewart of
Bell Telephone laboratories,
USA in 1924
Compound Graph (Line & Bar)
Run Chart / Trend Chart
Concept of Control Chart
Control Chart is a powerful
tool to differentiate
Chance vs. Assignable Causes.
! A time ordered plot of the data.
! Reflects of expected variation of the
data.
! Identifies when a special cause is acting
on the process.
Stratification
! Process of separation of data into categories.
! Developed by A M Guery, a French statistician.
! Normally done for identifying the categories contributing to the problem.
! A graphical representation of a frequency distribution, which is a
summary of variation in a product or process.
! Used in combination with other data analysis tools.
! Used when data from a variety of sources or categories have been collected
together. This tool separates the data into categories so that patterns can be
seen.
! Used for Defining Projects or Analyzing Symptoms or Testing theories
stages of problem solving.
Stratification An Example
Combined data from a chemical reaction process in two vessels, A and B
a) General type
b) Comb type
c) Positively skew type
f) Left-hand precipice type
Histogram An Example
! It displays variation over a period to time. Visual depiction of data
helps to infer about the population at a glance.
e) Plateau type
14
12
Type of Control Chart
Continuous value
X bar - R Chart (Average
Value and Range)
Discrete Value
pn Chart (No. of Defect
Units)
P Chart (Fraction of Defective)
C Chart (Number of Defects)
Assignable Cause
X Bar
10
15
20
25
30
35
40
45
U Chart (Number of Defects per
Unit)
50
6.3
6.2
UCL = 6.202
6.1
6.0
Mean = 6.020
5.9
LCL = 5.839
5.8
1
0.8
Shows explicit values for
CL (Center line) and Upper
& Lower Control Limits
(UCL & LCL)
UCL = 0.6664
0.6
0.4
Mean = 0.3152
0.2
0
5
10
15
20
25
30
35
Sub-group number
40
45
50
Criteria for Judging the
Un-control State :
1) When the points are
outside control limits
2) 7 continuous points
increasing or decreasing
3) 7 continuous points
lying on one side of the
average range
Note - For more details refer DM workbook
Scatter Diagram
! Developed by Dr. Buxton in 1794 and used by J F W Herschel in
1832 to fit a curve to scatter plots.
Patterns of Scatter Diagram
! Reveals the nature of relationship (cause-effect) between two
variables and their approximate strength in the diagnostic
journey to identify root cause of a problem. The better the
correlation, the tighter the points will hug the line.
f) Twin-peak type
g) Isolated-peak type
No. of data points =
100
16
Characteristic Value
! A graphical representation of relationship between two
variables. It can be between a cause and effect or between two
causes. If the variables are correlated, the points will fall among a
line or curve.
! It graphically shows shape of the distribution, average, spread and
level of conformance to specifications.
! Used to study process variation and assess process capability.
Type of Control Chart
! Popularized by Dr. Kaoru Ishikawa in 1950s.
! Enables to see patterns that are difficult to see in a table of numbers.
Histogram (A + B)
18
No. of observations
Histogram Various Shapes
Characteristic Value
Indicate the type of control chart at the centerline position
Histogram
Control Chart - Types and Example
X Chart (Measured Value)
Comparision of Pareto Diagrams Before and After Improvement
Some Patterns in Time Order Plots
! Developed by William Playfair (1759-1853), an English Economist and J H Lambert (1728
1777), a Swiss-German scientist and mathematician.
! A Pictorial representation of quantitative data.
! Represents large amounts of information / data comprehensively and in compact
manner.
! Communicates complex situations concisely and in clearly understandable manner for
anybody.
! Generally used to study the process / performance trend over time at a glance.
Ounces of Soup
HAR SHIKHAR
Range
1 Check Sheet
KAR VIJAY
Scatter Diagram An Example
Histogram with Specifications
Mean = 85.05
10
Stdev = 1.35
8
6
Max : 87.9
Min : 82.7
2
0
Cases in which the histogram
does not satisfy the specification
Cases in which the histogram satisfies
the specification
82.0 82.5 83.0 83.5 84.0 84.5 85.0 85.5 86.0 86.5 87.0 87.5 88.0 88.5
Yield
SL
After Stratification : Two vessels, A and B
Histogram (A)
Histogram (B)
18
14
Average
Stdev
Max
Min
No. of observations
14
12
10
8
84.07
0.805
86.3
82.7
6
4
10
8
6
Average
Stdev
Max
Min
86.03
1.036
87.9
83
Vessel A and
vessel B are
behaving differently,
therefore to be
dealt with
differently.
4
2
2
0
12
No. of observations
16
82.0
82.5 83.0
83.5
84.0
Yield
84.5
85.0 85.5 86.0
86.5
87.0
Now it is telling a
different story.
82.5 83.0 83.5 84.0 84.5 85.0 85.5 86.0 86.5 87.0 87.5 88.0 88.5
Yield
SL
! Plots the number of times an event has occurred.
! Shows the distribution of occurrence of event.
SL
a)
b)
c)
SU
SU
SU
SL
d)
SU
SL
e)
SU
SL : Upper Specification Limit; SU : Lower Specification Limit
Source
TOTAL QUALITY MANAGEMENT
