MQC TechnicalPublication PDF

“Statistics in Metrology”
© RSB
Statistical Quality Control
1. It refers to statistical techniques employed for the
maintenance of uniform quality in a continuous flow of
manufactured product through statistical process control
2. SQC may be defined as the management technique by
means of which products of uniform acceptable quality
are manufactured
© RSB
Statistical Quality Control
• Objective
1. Effective and efficient utilization of available resources.
2. Standard inspection methods.
3. To take different measures to improve the standard of quality of
product.
● Advantages
1. Stable process which meets the product specifications, provides
evidence that the process has conditions which if maintained, will
result in an acceptable product.
2. Knowing that the process is in statistical control, is helpful to the
workers running a process.
3. Reduction in cost per piece and scrap.
© RSB
Three SQC Categories
• Statistical quality control (SQC) is the term used to describe the set of
statistical tools used by quality professionals
• SQC encompasses three broad categories of;

– Descriptive statistics
• e.g. the mean, standard deviation, and range
– Statistical process control (SPC)
• Involves inspecting the output from a process
• Quality characteristics are measured and charted
• Helpful in identifying in-process variation
– Acceptance sampling used to randomly inspect a batch of goods to
determine acceptance/rejection
• Does not help to catch in-process problems
© RSB
Statistical Process Control (SPC)
– The costs are low if the problems are detected early
– If a large number of defective products are produced, the

scrap or rework costs can be high
– Prevention of defects by applying statistical methods to

control the process is known as SPC
– Prevention refers to those activities designed to prevent

defects, defectives, and nonconformance in products and
services
© RSB
Statistical Process Control (SPC)
• Statistical process control (SPC) is a methodology focused
on quality control and improvement, using data analysis
• It consists of using valid statistical data analysis to determine

and eliminate variation due to assignable causes.
• It is based on the following principles:

- measuring the process
- identifying and eliminating unusual variation
- improving the process to its best target value
- monitoring the process performance over time
© RSB
Benefits of Statistical Process Control
• Process improvement
• In-depth management decision making
• Better understanding of process performance
• Establish process baselines
• Better control on variables that impact a process
• Gain More predictability
© RSB
Examples of Tools used in SPC
• Histograms
• Scatter Diagrams
• Run Charts
• Pareto Charts
• Control Charts
© RSB
Process & Variation
• A process is a set of activities operating on a system to
transform inputs to outputs (f(X) = Y)
INPUTS ACTIVITY OUTPUTS
X f(X) Y
• Example: making a cup of tea, baking a cake, getting to work, etc.
• Any process will have a certain degree of variation; some variation

will be inherent to the process, some will not.
• Variation in a Process = Common Cause Variation + Special Cause

variation
• Goal is to reduce process variation
© RSB
Sources of Variation
• Common causes of variation
– also called random or uncontrollable causes of variation
– It is the natural or expected variation in the process
– causes that are random in occurrence and are inherent in all
processes
– Unavoidable
• Special causes of variation

– also called assignable causes of variation
– It is unexpected variation that results from unusual occurrences
– the result of external sources outside the system
– these causes can and must be detected, and corrective action must
be taken to remove them from the process
– failing to do so will increase variation and lower quality
– Typical causes are poor employee training, worn tool, machine
needing repair, etc.
© RSB
Common causes of Variation
 Within any process there are many variable factors (such as wind and
muscle tremors in firing a gun) each of which may vary a small amount
and in a predictable way, but when taken together result in a degree of
randomness in the output, as indicated in the figure below. These
seemingly uncontrollable factors are called common causes of variation.
 Common causes of variation cannot be eliminated by reacting to

individual variations. The spread in results can only be reduced through a
deep understanding of how the process works.
 Control Charts can be used to differentiate between common and

special causes of variation.
Special Common
Cause Cause
© RSB
Process Variation
Total Process Common Assignable Cause

Variation = Cause Variation + Variation
 Variation is natural; inherent in the world

around us
 No two products or service experiences are
exactly the same
 With a fine enough gauge, all things can be
seen to differ
© RSB
Total Process Variation

Variation is often due to differences in:

 People
 Machines
 Materials
 Methods
 Measurement
 Environment
© RSB
Common Cause Variation

Common cause variation

 naturally occurring and expected
 the result of normal variation in materials,
tools, machines, operators, and the
environment
© RSB
Special Cause Variation

Special cause variation

 abnormal or unexpected variation
 has an assignable cause
 variation beyond what is considered
inherent to the process
© RSB
Stable Process
• A process is stable (in-control) if
– all assignable causes are removed
– variation results only from common causes
© RSB
Statistics Revision
 Mean
 Median
 Mode
 Range
 Variance
 Standard Deviation
© RSB
Mean, Median, Mode
are all types of average.
An average summarises groups of

data.
© RSB
Mean
• Average often means the ‘mean’
• Mean = total of the numbers divided by how many
numbers.
Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6

Add up the numbers:
3 + 5 + 5 + 6 + 4 + 3 + 2 + 1 + 5 + 6 = 40
Divide by how many numbers:
40 ÷ 10 = 4
The class mean shoe size is 4
© RSB
Median
• Median is the middle value
• Put the numbers in order
• Choose the number in the middle of the list.
• If there are 2 numbers in the middle then it is
halfway between them.
Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6
Put in order: 1, 2, 3, 3, 4, 5, 5, 5, 6, 6
The class median shoe size is 4.5
© RSB
Mode
• Mode is the most common number
• Choose the number that appears the most
frequently.
• Sometimes there may be more than one mode.
Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6

Put in order: 1, 2, 3, 3, 4, 5, 5, 5, 6, 6
The class modal shoe size is 5.
© RSB
Range
• Range is how far from biggest to smallest.
• Take the smallest number from the largest.
Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6

Put in order: 1, 2, 3, 3, 4, 5, 5, 5, 6, 6
Subtract smallest from largest: 6 – 1 = 5
Range: 5
© RSB
Variance & Standard Deviation
a measure of how data points differ from the mean
• Data Set 1: 3, 5, 7, 10, 10

Data Set 2: 7, 7, 7, 7, 7
What is the mean and median of the above data set?

Data Set 1: mean = 7, median = 7
Data Set 2: mean = 7, median = 7
But we know that the two data sets are not identical!
The variance shows how they are different.
We want to find a way to represent
these two data set numerically.
© RSB
• The average of the squared deviations about the
mean is called the variance.
 Observed Value  mean

2
Variance 
number of observations
• Standard Deviation
 Observed Value  mean

2
Std. Deviation    Variance 

number of observations
© RSB
Variance : Example
Score (
X  X)
2
X X
X
1
3
2
5
3
7
4
10
5
10
Totals
35
The mean is 35/5=7.
© RSB
Variance : Example
Score (
X  X)
2
X X
X
1
3 3-7=-4
2
5 5-7=-2
3
7 7-7=0
4
10 10-7=3
5
10 10-7=3
Totals
35
© RSB
Variance : Example
Score (
X  X)
2
X X
X
1
3 3-7=-4 16
2
5 5-7=-2 4
3
7 7-7=0 0
4
10 10-7=3 9
5
10 10-7=3 9
Totals
35 38
© RSB
Variance : Example
Score (
X  X)
2
X X
X
1
3 3-7=-4 16
2
5 5-7=-2 4
3
7 7-7=0 0
4
10 10-7=3 9
5
10 10-7=3 9
Totals
35 38
x  X 
2
38
s 
2
  7.6
n 5
© RSB
Numerical
1
Consider the following distribution:
11, 17,25, 28, 34
Calculate the mean, variance and standard deviation.
© RSB
Numerical
1
• Mean = (11+17+25+28+34) = 23
5
• Variance = (11-23)2+(17-23)2+(25-23)2+(28-23)2+(34-23)2 = 66
• Standard Deviation = √66 = 8.12
© RSB
Numerical
2
Consider the following
distribution:
3, 5, 5, 6, 4, 3, 2, 1, 5, 6
Calculate
Mean , Median,
Mode, Range,
© RSB
Numerical 3
A batch of 100 test specimens made of grey cast iron FG 300 are tested on a
universal testing machine (UTM) to determine ultimate tensile strength of the
material. The results are tabulated as follows:
Class interval of ultimate Frequency

tensile strength (N/mm2)
261-280 4
281-300 32
301-320 50
321-340 12
341-360 2
Calculate :
(i) Mean
(ii) Standard Deviation
(iii) Variance
for the ultimate tensile strength of this sample
© RSB
Solution:
Given Data:
N = Total number of observations = 100
To find:
(i) Mean
(ii) Standard Deviation
(iii) Variance
STEP 1: Make Frequency Distribution Table

STEP 2: Calculate Mean
STEP 3: Calculate Standard Deviation
STEP 4: Calculate Variance
© RSB
STEP 1: Frequency Distribution Table
Class Interval Class Frequency (fi) fixi fixi2

mid-point (xi)
261-280 4
281-300 32
301-320 50
321-340 12
341-360 2
Total ∑fi= ∑fixi= ∑fixi2=
© RSB

mid-point (xi)
261-280 270 4
281-300 290 32
301-320 310 50
321-340 330 12
341-360 350 2
Total ∑fi=100 ∑fixi= ∑fixi2=
© RSB

mid-point (xi)
261-280 270 4 1080 291600
281-300 290 32 9280 2691200
301-320 310 50 15500 4805000
321-340 330 12 3960 1306800
341-360 350 2 700 245000
Total ∑fi=100 ∑fixi= ∑fixi2=
© RSB

mid-point (xi)
261-280 270 4 1080 291600
281-300 290 32 9280 2691200
301-320 310 50 15500 4805000
321-340 330 12 3960 1306800
341-360 350 2 700 245000
Total ∑fi=100 ∑fixi=30520 ∑fixi2=9339600
© RSB
Mean 
 fx i i
f i
 fx
2
 fx 2
i i  i i
N

N 1
Variance S    
2
© RSB
Mean 
 fx i i

30520
 305.20 N / mm2
f i 100
 fx
2
30520
2
fx 2
i i  i i
N
9339600 
100
   15.86 N / mm2
N 1 100  1
Variance S      15.86  251.47 N / mm2

2 2
© RSB
Normal Distribution Curve
 Bell-shaped curve
 Symmetrical around the mean
 Defined by its mean and standard deviation
 68.3% of data found within one standard deviations away from

the mean
 95.5% of data found within two standard deviations away from

the mean
 99.7% of data found within three standard deviations away from

the mean
© RSB 40
Control Charts
 Statistical tool used to

monitor the stability
of a process over time
 Key features:
- UCL (Upper Control Limit) = mean + 3*sigma
- LCL (Lower Control Limit) = mean – 3*sigma
- central line (mean of data set)
 A process is said to be in control when data points fall within limits

of variation (i.e.: between Upper and Lower Control Limits)
© RSB 41
Control Charts & Data Types
 Control charts can measure two types of data:
- Continuous Data (can be measured)

E.g.: temperature, volume, weight, height, time
- Discrete Data (can be counted)

E.g.: How many people in this room?
How many defects in an inspected unit?
 There are different control charts to choose from depending on what data
is available.
© RSB 42
Control Chart Types
© RSB 43
Control Charts
• Definition:
- A statistical tool to determine if a process
is in control.
© RSB
History of Control Charts
• Developed in 1920’s
• By Dr. Walter A. Shewhart
• Shewhart worked for Bell Telephone Labs
© RSB
Two Types of Control Charts
• Variable Control Charts

• Attribute Control Charts
© RSB
Variable Control Charts
• Deal with items that can be measured .

• Examples
1) Weight
2) Height
3) Speed
4) Volume
© RSB
Types of Variable Control Charts
• X-Bar chart
• R chart
• MA chart
© RSB
Variable Control Charts
• X chart: deals with a average value in a

process
• R chart: takes into count the range of the
values
• MA chart: take into count the moving average
of a process
© RSB
Attribute Control Charts
• Control charts that factor in the quality

attributes of a process to determine if the
process is performing in or out of control.
© RSB
Types of Attribute Control Charts
• P chart
• C Chart
• U Chart
© RSB
Attribute Control Charts
• P Chart: a chart of the percent defective in

each sample set.
• C chart: a chart of the number of defects per
unit in each sample set.
• U chart: a chart of the average number of
defects in each sample set.
© RSB
Reasons for using Control Charts
• Improve productivity
• Make defects visible
• Determine what process adjustments need to
be made
• Determine if process is “in” or “out of control”
© RSB
Control Chart Key Terms
• Out of Control: the process may not

performing correctly
• In Control: the process may be performing
correctly
• UCL: upper control limit
• LCL: lower control limit
• Average value: average
© RSB
Process is OUT of control if:
• One or multiple points outside the control

limits
• Eight points in a row above the average value
• Multiple points in a row near the control limits
© RSB
Process is IN control if:
• The sample points fall between the control

limits
• There are no major trends forming, i.e.. The
points vary, both above and below the
average value.
© RSB
Examples of Control Charts
© RSB
Examples of Control Charts
© RSB
Control Charts
• The following control chart shows the

improvement of a process. The standard
deviation decreases as the process becomes
more capable.
© RSB
Example of Control Charts
© RSB
X - Control Chart
 It is the chart which is used to represent central
tendency.
 In this case most of observations tend to concentrate in

the centre f distribution.
 It represent any variation in the process average.
 It gives the information about size of the process in

order to control the process.
© RSB
Procedure to draw X Control Charts
Step-1 : Calculate the value of X
Step-3 : Calculate UCL and LCL for X control chart
Step-4: Draw X control chart
© RSB
Example:
If a batch contains ‘5’ number of samples and their sizes are
X1, X2, X3, X4, X5
X1  X 2  X 3  X 4  X 5
X
5
It can also be written as,
X 
 X
N
Where N = Total number of samples
© RSB
X is an average value of X with respect to number of batches or

subgroups to be inspected
Example:
Total ‘6’ numbers of batches are inspected with batch
contains ‘5’ number of samples each and X is given by
X1  X 2  X 3  X 4  X 5  X 6
X
6
It can also be written as,
X
X
Number of batches or subgroups
© RSB
UCLX  X  3 X
LCLX  X  3 X
where,

X 
n
  Std . deviation
n  Sample size
© RSB
OR using average range R Sample size A2

(n)
UCL  X  A2 R 2 1.88
3 1.02
LCL  X  A2 R
4 0.73
5 0.58
where, A2 is the factor determining 3 6 0.48
limits for X chart using value of R . 7 0.42
8 0.37
The values of A2 are standard for 9 0.34
specific size and are given 10 0.31
© RSB
Step-4: Draw X control chart
By using values of X , UCLX , LCLX

Mark batch number / subgroups numbers on X-axis and avg.
values ( X ) on Y-axis
UCLX
X X
LCLX
1 2 3 4 5 6 7 8
Batch number
© RSB
R - Control Chart
 It is the chart which is used to represent dispersion

tendency.
 In this case, the data is scattered about the zone of

central tendency.
 It represents the consistency of process.
 It gives the information about any change in dispersion

within the process.
© RSB
Procedure to draw R - Control Chart
Step-1 : Calculate the value of Range (R)
Step-2 : Calculate the average of range ( R )
Step-3 : Calculate UCL and LCL for R-control chart
Step-4: Draw R-control chart
© RSB
Step-1 : Calculate the value of Range (R)
The value of range (R) for each subgroup is calculated for

given data
Range R  X max  X min
© RSB
Step-2 : Calculate the average of range ( R )
The average value of range ( R ) is calculated by,
R
 R
…N= Total number of samples
N
Example:
Total ‘6’ numbers of batches are inspected with batch contains
‘5’ number of samples each then R is given by
R1  R2  R3  R4  R5  R6 R
R
R OR
Number of batches or subgroups
6
© RSB
Step-3 : Calculate UCL and LCL for R-control chart

Sample LCL UCL
UCLR  D4 R size (n) D3 D4
2 0 3.27
LCLR  D3 R 3 0 2.57
4 0 2.28
where, D3 and D4 are the factors 5 0 2.11
for LCL and UCL respectively for 6 0 2.00
R-Chart. 7 0.08 1.92
8 0.14 1.87
The value of D3 and D4 are 9 0.18 1.82
standard for specific sample size 10 0.22 1.78
© RSB
Step-4: Draw R-control chart

By using values of R , UCLR , LCL R
Mark batch number / subgroups numbers on X-axis and avg.
values ( R ) on Y-axis
UCLR
R R
LCL R
1 2 3 4 5 6 7 8
Batch number
© RSB
Numerical 4
Specification of item 500 ± 5,

n =5, N=20, A2=0.58, D3=0, D4=2.11,
Plot the chart and comment.
Batch 1 2 3 4 5 6 7 8 9 10
X 501 498 500 503 501 500 497 502 503 496
R 3 4 2 4 3 5 4 2 6 4
To find:
Draw X and R Chart
© RSB
Step 1: Calculate X and R
Step 2: Calculate UCL and LCL for X–chart
Step 3: Calculate UCL and LCL for R-Chart
Step-4: Plot X -chart
Step-5: Plot R-chart
© RSB
X R
R
X
Number of subgroups Number of subgroups
Step 2: Calculate UCL and LCL for X –chart

UCL  X  A2 R
LCL  X  A2 R
UCLR  D4 R
LCLR  D3 R
© RSB
X
X
Number of subgroups
501  498  500  503  501  500  497  502  503  496
X
10
X
R
R
Number of subgroups
3 4 2 435 4 2 6 4
R
10
R
© RSB
X
X
Number of subgroups
501  498  500  503  501  500  497  502  503  496
X
10
X  500.1
R
R
Number of subgroups
3 4 2 435 4 2 6 4
R
10
R  3.7
© RSB
UCL  X  A2 R
LCL  X  A2 R
UCL  500.1  0.58  3.7

UCL 
LCL  500.1  0.58  3.7

LCL 
© RSB
UCL  X  A2 R
LCL  X  A2 R
UCL  500.1  0.58  3.7

UCL  502.246
LCL  500.1  0.58  3.7

LCL  497.954
© RSB
Step 3: Calculate UCL and LCL for R–chart
UCLR  D4 R
LCLR  D3 R
UCLR  2.11 3.7

UCLR 
LCLR  0  3.7
LCLR  0
© RSB
UCLR  D4 R
LCLR  D3 R
UCLR  2.11 3.7

UCLR  7.807
LCLR  0  3.7
LCLR  0
© RSB
Step 4: Plot the X -chart
UCLX
X X
LCLX
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
UCLX  502.246
X X  500.1
LCLX  497.954
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
Step 5: Plot the R-chart
UCLR
R R
LCL R
1 2 3 4 5 6 7 8
Batch number
© RSB
UCLR  7.807
R R  3.7
LCLR  0
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
Numerical 5
Specification of item 40.37 ± 0.1,

A2=0.577, D3=0, D4=2.11,
Batch 1 2 3 4 5 6 7 8 9 10
X 31.8 34 30.8 35 33 33.8 35.8 34 33 33.8
R 4 2 5 5 19 4 14 7 9 5
To find:
Draw X and R Chart
© RSB
© RSB
X R
R
X
Number of subgroups Number of subgroups

UCL  X  A2 R
LCL  X  A2 R
UCLR  D4 R
LCLR  D3 R
© RSB
X
X
Number of subgroups
31.8  34  30.8  35  33  33.8  35.8  34  33  33.8
X
10
X
R
R
Number of subgroups
4  2  5  5  19  4  14  7  9  5
R
10
R
© RSB
X
X
Number of subgroups
31.8  34  30.8  35  33  33.8  35.8  34  33  33.8
X
10
X  33.5
R
R
Number of subgroups
4  2  5  5  19  4  14  7  9  5
R
10
R  7.4
© RSB
UCL  X  A2 R
LCL  X  A2 R
UCL  33.5  0.577  7.4

UCL 
LCL  33.5  0.577  7.4

LCL 
© RSB
UCL  X  A2 R
LCL  X  A2 R
UCL  33.5  0.577  7.4

UCL  37.7698
LCL  33.5  0.577  7.4

LCL  29.2302
© RSB
UCLR  D4 R
LCLR  D3 R
UCLR  2.11 7.4

UCLR 
LCLR  0  7.4
LCLR  0
© RSB
UCLR  D4 R
LCLR  D3 R
UCLR  2.11 7.4

UCLR  15.614
LCLR  0  7.4
LCLR  0
© RSB
UCLX
X X
LCLX
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
UCLX  37.7698
X X  33.5
LCLX  29.2302
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
UCLR
R R
LCL R
1 2 3 4 5 6 7 8
Batch number
© RSB
UCLR  7.807
R R  3.7
LCLR  0
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
Numerical 6
A2=0.577, D3=0, D4=2.115, N=8, n=4

Sample X1 X2 X3 X4
No.
1 80.74 80.76 80.77 80.73
2 80.73 80.76 80.72 80.75
3 80.82 80.75 80.77 80.81
4 80.74 80.73 80.71 80.77
5 80.74 80.75 80.73 80.74
6 80.76 80.75 80.74 80.74
7 80.78 80.77 80.76 80.80
8 80.78 80.77 80.80 80.81
To find:
Draw X and R Chart
© RSB
Step 1: Calculate the mean for each sample
Step 2: Calculate the range for each sample
Step 3: Calculate the value of X and R
Step 6: Plot X -chart
Step 7: Plot R-chart

© RSB
For 1st sample,
80.74  80.76  80.77  80.73

X 
4
For 2nd sample,
80.73  80.76  80.72  80.75

X 
4
Similarly calculate for remaining X
© RSB
For 1st sample,
80.74  80.76  80.77  80.73

X  80.75
4
For 2nd sample,
80.73  80.76  80.72  80.75

X  80.74
4
Similarly calculate for remaining X
© RSB
Sample X1 X2 X3 X4
No. X
1 80.74 80.76 80.77 80.73 80.75
2 80.73 80.76 80.72 80.75 80.74
3 80.82 80.75 80.77 80.81
4 80.74 80.73 80.71 80.77
5 80.74 80.75 80.73 80.74
6 80.76 80.75 80.74 80.74
7 80.78 80.77 80.76 80.80
8 80.78 80.77 80.80 80.81
© RSB
Sample X1 X2 X3 X4
No. X
1 80.74 80.76 80.77 80.73 80.75
2 80.73 80.76 80.72 80.75 80.74
3 80.82 80.75 80.77 80.81 80.7875
4 80.74 80.73 80.71 80.77 80.7375
5 80.74 80.75 80.73 80.74 80.74
6 80.76 80.75 80.74 80.74 80.7475
7 80.78 80.77 80.76 80.80 80.7775
8 80.78 80.77 80.80 80.81 80.79
© RSB

For 1st sample,
X max  80.77
X min  80.73
R
For 2nd sample,

X max  80.76
X min  80.72
R
Similarly calculate for remaining R
© RSB

For 1st sample,
X max  80.77
X min  80.73
R  80.77  80.73  0.04
For 2nd sample,

X max  80.76
X min  80.72
R  80.76  80.72  0.04
Similarly calculate for remaining R
© RSB
Sample X1 X2 X3 X4 R
No. X
1 80.74 80.76 80.77 80.73 80.75 0.04
2 80.73 80.76 80.72 80.75 80.74 0.04
3 80.82 80.75 80.77 80.81 80.7875
4 80.74 80.73 80.71 80.77 80.7375
5 80.74 80.75 80.73 80.74 80.74
6 80.76 80.75 80.74 80.74 80.7475
7 80.78 80.77 80.76 80.80 80.7775
8 80.78 80.77 80.80 80.81 80.79
© RSB
Sample X1 X2 X3 X4 R
No. X
1 80.74 80.76 80.77 80.73 80.75 0.04
2 80.73 80.76 80.72 80.75 80.74 0.04
3 80.82 80.75 80.77 80.81 80.7875 0.07
4 80.74 80.73 80.71 80.77 80.7375 0.06
5 80.74 80.75 80.73 80.74 80.74 0.02
6 80.76 80.75 80.74 80.74 80.7475 0.02
7 80.78 80.77 80.76 80.80 80.7775 0.04
8 80.78 80.77 80.80 80.81 80.79 0.04
© RSB
X R
R
X
Number of samples Number of samples

UCL  X  A2 R
LCL  X  A2 R
UCLR  D4 R
LCLR  D3 R
Step 6: Plot X-chart
Step 7: Plot R-chart
© RSB
X
X
Number of subgroups
80.75  80.74  80.7875  80.7375  80.74  80.7475  80.7775  80.79
X
8
X
R
R
Number of subgroups
0.04  0.04  0.07  0.06  0.02  0.02  0.04  0.04
R
8
R
© RSB
X
X
Number of subgroups
80.75  80.74  80.7875  80.7375  80.74  80.7475  80.7775  80.79
X
8
X  80.7587
R
R
Number of subgroups
0.04  0.04  0.07  0.06  0.02  0.02  0.04  0.04
R
8
R  0.04125
© RSB
UCL  X  A2 R
LCL  X  A2 R
UCL  80.7587  0.577  0.04125

UCL 
LCL  80.7587  0.577  0.04125

LCL 
© RSB
UCL  X  A2 R
LCL  X  A2 R
UCL  80.7587  0.577  0.04125

UCL  80.7825
LCL  80.7587  0.577  0.04125

LCL  80.7348
© RSB
UCLR  D4 R
LCLR  D3 R
UCLR  2.115  0.04125

UCLR 
LCLR  0  0.04125
LCLR  0
© RSB
UCLR  D4 R
LCLR  D3 R
UCLR  2.115  0.04125

UCLR  0.08724
LCLR  0  0.04125
LCLR  0
© RSB
UCLX
X X
LCLX
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
UCLX  80.7825
X X  80.7587
LCLX  80.7348
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
UCLR
R R
LCL R
1 2 3 4 5 6 7 8
Batch number
© RSB
UCLR  0.08724
R R  0.04125
LCLR  0
1 2 3 4 5 6 7 8 9 10
Batch number
© RSB
Numerical 7
A2=0.577, D3=0, D4=2.115, N=5, n=5

Sample No. X1 X2 X3 X4 X5
1 12.62 12.63 12.62 12.61 12.59
2 12.60 12.56 12.56 12.66 12.58
3 12.62 12.60 12.57 12.62 12.57
4 12.61 12.59 12.58 12.61 12.59
5 12.65 12.60 12.63 12.60 12.56
To find:
Draw X and R Chart
© RSB
Control Chart Patterns
 When a control chart pattern is natural (follows normal curve pattern),
there are no abnormal extraneous conditions in the process.
 When the pattern is not natural, something is present in the process that
has an effect on the control chart pattern.
 All unnatural patterns need to be investigated to determine the cause.
 The chart itself will not determine the cause, it will just indicate whether
the pattern is natural or unnatural.
 A chart that is in control will show natural patterns of variation.
 Unnatural patterns tend to fluctuate too wide or they fail to balance
around the centerline.
 The portrayal of natural and unnatural patterns is what makes the control
chart a very useful tool for statistical process and quality control.
 When a chart is interpreted, we look for special patterns such as cycles,
trends, freaks, mixtures, groupings or bunching of measurements, and
sudden shifts in levels.
© RSB
Natural Pattern
The points follow a standard normal curve pattern.
The characteristics of natural patterns:
 Most of the points are near the average
 A few points will approach the control limits
 Very rarely will points exceed the control limits
© RSB
Mixtures
Most points fall near the control limits and few fall near the centerline.
 Possible causes of mixtures:

-Lots from different sources
-Differences in people performing checks
-Differences in test equipment
-Two or more different materials, machines, operators or gauges
© RSB
Sudden Shift in Level
The points show a marked shift in one direction.
 Possible causes of sudden shifts:

-New production equipment
-New people performing check
-New test sets
-New machine settings
-New supplier
-Inadequate or revised maintenance
-Change in equipment calibration
-Improved process capability
© RSB
Trends
The points show a gradual shift in one direction.
 Possible causes of trends:

-Equipment wear
-Inadequate or improved maintenance
-Gradual introduction of new material
-Process improvement or deterioration
© RSB
Grouping or Bunching
All or most points beyond the control limits appear quite close together.
 Possible causes of Grouping or Bunching:

-Different checkers or inspectors
-Plotting errors
-Change in test set calibration
© RSB
Cycles
Cycles are short term repeating trends in the data.
 Possible causes of cycles:

-Different equipment
-Different checkers or inspectors
-Differences between suppliers
-Seasonal effects (temperature, etc.)
© RSB
Freaks
Freaks are occasional occurrences that vary greatly

from the normal pattern.
 Possible causes of Freaks:

- Breakdown of equipment
- Accidental damage in handling
- Incomplete or omitted operation
- Wrong setting on machine
- Measurement error
© RSB
Instability
Instability is characterized by unnaturally large fluctuations. The
pattern exhibits erratic ups and downs and is frequently out of control
at both limits.
 Possible causes of Instability:

-Mixture of material
-Machine in need of repair (unstable)
-Unstable test equipment
-Non-random sampling
-Untrained checker or inspector
© RSB
Process Capability
 The statistical quality control analysis is used to determine whether the
given process is under control or not
 In practical applications, products are always designed with specific amount
of tolerance to compensate the effect of random variables in manufacturing
process.
 In other words, we can say that the process capability represents the
reliability of a process.
 Mathematically, the process capability is defined as 6σ. It is measure of the
spread of process. It is given by,
R

d2
  Std. deviation
R  Avg. range value
d 2  Factor for estimating  from R
© RSB
Process Capability Indices
We are often required to compare the output of a stable process with the
process specifications and make a statement about how well the process
meets specification. To do this we compare the natural variability of a
stable process with the process specification limits.
A process where almost all the measurements fall inside the

specification limits is a capable process. This can be represented
pictorially by the plot below:
Allowable process spread
USL
Actual process spread
LSL
© RSB
Process Capability Indices
There are several statistics that can be used to measure the capability of a
process:
Estimates what the process is capable of
producing if the process mean were to be
USL  LSL centered between the specification limits.
Cp  Assumes process output is approximately
6 normally distributed.
Estimates what the process is

USL     LSL  capable of producing, considering
C pk  min  ,
3 
that the process mean may not be
 3 centered between the specification
limits.
USL  LSL
C pm 
6  2

  T 2 Estimates process capability around a target, T.
© RSB
Numerical 8
For certain process,

USL  106
LSL  94
  1.03
  98.94
T  100
Find,
C p , C pk , C pm
© RSB

MQC TechnicalPublication PDF

Încărcat de

Informații document

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

MQC TechnicalPublication PDF

Încărcat de

Drepturi de autor:

Formate disponibile

“Statistics in Metrology”

• SQC encompasses three broad categories of;

– The costs are low if the problems are detected early

– If a large number of defective products are produced, the

– Prevention of defects by applying statistical methods to

– Prevention refers to those activities designed to prevent

• It consists of using valid statistical data analysis to determine

• It is based on the following principles:

• In-depth management decision making

• Better understanding of process performance

• Establish process baselines

• Better control on variables that impact a process

• Gain More predictability

• Any process will have a certain degree of variation; some variation

• Variation in a Process = Common Cause Variation + Special Cause

• Goal is to reduce process variation

• Special causes of variation

 Common causes of variation cannot be eliminated by reacting to

 Control Charts can be used to differentiate between common and

Total Process Common Assignable Cause

 Variation is natural; inherent in the world

Total Process Common Assignable Cause

Variation is often due to differences in:

Total Process Common Assignable Cause

Common cause variation

Total Process Common Assignable Cause

Special cause variation

An average summarises groups of

Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6

The class median shoe size is 4.5

Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6

The class modal shoe size is 5.

Class shoe sizes: 3, 5, 5, 6, 4, 3, 2, 1, 5, 6

• Data Set 1: 3, 5, 7, 10, 10

What is the mean and median of the above data set?

 Observed Value  mean

 Observed Value  mean

Std. Deviation    Variance 

The mean is 35/5=7.

Consider the following distribution:

11, 17,25, 28, 34

Calculate the mean, variance and standard deviation.

• Standard Deviation = √66 = 8.12

Class interval of ultimate Frequency

N = Total number of observations = 100

STEP 1: Make Frequency Distribution Table

Class Interval Class Frequency (fi) fixi fixi2

Class Interval Class Frequency (fi) fixi fixi2

Class Interval Class Frequency (fi) fixi fixi2

Class Interval Class Frequency (fi) fixi fixi2

STEP 3: Calculate Standard Deviation

STEP 4: Calculate Variance

STEP 3: Calculate Standard Deviation

STEP 4: Calculate Variance

Variance S      15.86  251.47 N / mm2

 Symmetrical around the mean

 Defined by its mean and standard deviation

 68.3% of data found within one standard deviations away from

 95.5% of data found within two standard deviations away from

 99.7% of data found within three standard deviations away from

 Statistical tool used to

 A process is said to be in control when data points fall within limits

- Continuous Data (can be measured)

- Discrete Data (can be counted)