SlideShare a Scribd company logo
1 of 102
1 April 9, 2016 – v3.0
Six Sigma Process Capability Study (PCS)
by Operational Excellence Consulting LLC
2 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
3 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
4 April 9, 2016 – v3.0
How to get EVIDENCE to the following questions ?
 Case 1: RF Design specifies a critical capacitor in a design with (2 ± 0.1)F.
Is a particular supplier able to provide us with this type of component? How
well can the supplier meet the target/nominal value of 2F?
 Case 2: Mechanical Design specifies the distance between two holes by (32
± 0.5)mm. Is a particular supplier able to provide us with the required
accuracy? How well can the supplier meet the target/nominal value of 32mm?
 Case 3: The GSM Standard requires that the test “Peak TX Power” falls
between 28.3dB and 32.0dB. Is the new product able to fulfill this
requirement?
 Case 4: A fine pitch component requires a placement accuracy of ± 0.5mm. Is
the existing placement machine able to fulfill this requirement?
5 April 9, 2016 – v3.0
The process and quality control methods and techniques used today
got their start in the American Civil War at around 1789, when Eli
Whitney took a contract from the U.S. Army for the manufacture of
10,000 rifles at the unbelievably low price of $13.40 each.
At that time most of the products were handmade by small owner-
managed shops and product parts were thus not interchangeable.
The result of Whitney’s first mass production trail was that the rifles did
not work as well as the handmade rifles. In addition, the copied parts
did not fit as expected.
The History of Statistical and Process Thinking
6 April 9, 2016 – v3.0
GO - Test
NO-GO - Test
The first time that one presented machine produced parts was 1851 at the
industry exhibition in the Crystal Palace in London. An American gun smith took
10 working guns, took them apart, mixed all the parts in a box and re-assembled
them again. This was found a quite surprising “experiment”.
The History of Statistical and Process Thinking
7 April 9, 2016 – v3.0
Process Inspection
Good
Bad
Repair
Scrap
+
Monitor/Adjust
The Traditional Production Concept
The Detection Control Scheme
8 April 9, 2016 – v3.0
• The traditional production concept does not help us to
produce only good products.
• Every product has to be inspected.
• Products have to be repaired or even scraped.
• With respect to productivity and efficiency every activity
after the actual production process is a non-value added
activity.
The Traditional Production Concept
9 April 9, 2016 – v3.0
Prevention Control Scheme
Process Inspection
Good
Bad
Repair
Scrap
+
The Traditional Production Concept
10 April 9, 2016 – v3.0
Prevention Control Scheme
Process Inspection
Good
Bad
Repair
Scrap
+
An Advanced Production Concept
Monitor/Adjust
Learn/Improve
Selective measurement
• Product
• Process
11 April 9, 2016 – v3.0
Statistical Process Thinking - A Definition
All work is a series of
interconnected processes
All processes vary
Understanding, reducing and
controlling process variation
are keys to success
ASQ
12 April 9, 2016 – v3.0
Customer Satisfaction
or
Customer Dissatisfaction
Process/
System
Material
Machines Methods
Men
Environment
The Variation Management Approach
13 April 9, 2016 – v3.0
The Six Sigma Approach
Analytically speaking, this understanding may be expressed as
The process output “y” is a function of the process inputs “x1, x2, ... , xN”
where y is some product or process characteristic, also called the dependent
variable, and (x1, x2 ,..., xN) describes all the independent variables in the
cause system.
Thus, we may interpret this expression to mean the output variable (y) is a
function (f) of the input variables (x1, x2,..., xN) .
y = f(x1, x2, ... , xN)
14 April 9, 2016 – v3.0
Six Sigma - What is a Defect ?
A defect is any variation of a required
characteristic of the product or its part,
which is far enough removed from its
nominal value to prevent the product from
fulfilling the physical and functional
requirements of the customer.
15 April 9, 2016 – v3.0
The key to process control and continuous process
improvement is to understand the meaning and causes
of variation in the outcome of the process.
Variation Management – Continuous Improvement
16 April 9, 2016 – v3.0
Upper Specification Limit (USL)
Defect
nominal value
Process Capability and Process Control
Lower Specification Limit (LSL)
Defect
nominal value
Upper Specification Limit (USL)
Lower Specification Limit (LSL)
17 April 9, 2016 – v3.0
Upper Specification Limit (USL)
Defect
nominal value
Process Capability and Process Control
Lower Specification Limit (LSL)
Defect
nominal value
Upper Specification Limit (USL)
Lower Specification Limit (LSL)
process not capable
large variation & problem exist
root cause analysis
process improvement
18 April 9, 2016 – v3.0
Upper Specification Limit (USL)
process not capable
process out-of-control (trend)
root cause analysis
corrective action
Defect
nominal value
Process Capability and Process Control
Lower Specification Limit (LSL)
Defect
nominal value
Upper Specification Limit (USL)
Lower Specification Limit (LSL)
process not capable
large variation & problem exist
root cause analysis
process improvement
19 April 9, 2016 – v3.0
Product/Process Quality and Variation
“Traditional”
Attitude
Nominal
value
LSL USL
100 %
0 %
 Inspection/Yield and
re-active Problem Solving
Quality
“Six Sigma”
Attitude
Nominal
value
LSL USL
100 %
0 %
 Process Capability &
Process Control with
pro-active Improvements
Quality
20 April 9, 2016 – v3.0
Supplier Selection based on Yield or Process Capability
125.08125.06125.04125.02125.00124.98124.96124.94124.92
Upper SpecLower Spec
s
Mean-3s
Mean+3s
Mean
n
k
LSL
USL
Targ
Cpm
Ppk
PPL
PPU
Pp
Long-Term Capability
0
24969
0
32255
0.00
2.50
0.00
3.23
Obs
PPM<LSL Exp
Obs
PPM>USL Exp
Obs
%<LSL Exp
Obs
%>USL Exp
0.026
124.923
125.080
125.001
782.000
0.029
124.950
125.050
*
*
0.62
0.65
0.62
0.63
Process Capability Analysis for Supplier 1
125.04125.02125.00124.98124.96
Upper SpecLower Spec
s
Mean-3s
Mean+3s
Mean
n
k
LSL
USL
Targ
Cpm
Ppk
PPL
PPU
Pp
Long-Term Capability
0
1
0
1
0.00
0.00
0.00
0.00
Obs
PPM<LSL Exp
Obs
PPM>USL Exp
Obs
%<LSL Exp
Obs
%>USL Exp
0.010
124.969
125.031
125.000
1000.00
0.00
124.95
125.05
*
*
1.62
1.63
1.62
1.62
Process Capability Analysis for Supplier 2
100 % Yield due to 100 % Inspection 100 % Yield due to a capable Process
Which Supplier would you select ???
21 April 9, 2016 – v3.0
Remarks or Questions ?!?
22 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
23 April 9, 2016 – v3.0
A histogram provides graphical presentation and a first estimation
about the location, spread and shape of the distribution of the process.
0 10 20 30 40 50
The Histogram
24 April 9, 2016 – v3.0
Step 1: Collect at least 50 data points, but better 75 to 100 points, and organize
your data into a table. Sort the data points from smallest to largest and calculate
the range, means the difference between your largest and smallest data point, of
your data points.
The Histogram – How to create a Histogram?
Actual Measurements
Part Hole Size
1 2.6
2 2.3
3 3.1
4 2.7
5 2.1
6 2.5
7 2.4
8 2.5
9 2.8
10 2.6
Sorted Measurements
Part Hole Size
5 2.1
2 2.3
7 2.4
6 2.5
8 2.5
1 2.6
10 2.6
4 2.7
9 2.8
3 3.1
Minimum = 2.1
Maximum = 3.1
Range = 1.0
25 April 9, 2016 – v3.0
Step 2: Determine the number of bars to be used to create the histogram of the
data points. Calculate the width of one bar by dividing the range of your data by
the number of bars selected.
The Histogram – How to create a Histogram?
Number of Bars:
less than 50
50 - 100
100 - 250
over 250
5 or 7
5, 7, 9 or 11
7 - 15
11 - 19
Number of Data Points:
Minimum = 2.1
Maximum = 3.1
Range = 1.0
Bar Width = 0.2 (5 Bars)
26 April 9, 2016 – v3.0
Step 3: Calculate the “start” and “end” point of each bar and count how many
data points fall between “start” and “end” point of each bar.
The Histogram – How to create a Histogram?
Start End
Bar 1 2.1 2.1 + 0.2 = 2.3
Bar 2 2.3 2.5
Bar 3 2.5 2.7
Bar 4 2.7 2.9
Bar 5 2.9 3.1
Minimum = 2.1
Maximum = 3.1
Range = 1.0
Bar Width = 0.2 (5 Bars)
Sorted Measurements
Part Hole Size Bar
5 2.1 1
2 2.3 2
7 2.4 2
6 2.5 3
8 2.5 3
1 2.6 3
10 2.6 3
4 2.7 4
9 2.8 4
3 3.1 5
27 April 9, 2016 – v3.0
Step 4: Draw the histogram indicating by the height of each bar the number of
data points that fall between the “start” and “end” point of that bar.
The Histogram – How to create a Histogram?
Sorted Measurements
Part Hole Size Bar
5 2.1 1
2 2.3 2
7 2.4 2
6 2.5 3
8 2.5 3
1 2.6 3
10 2.6 3
4 2.7 4
9 2.8 4
3 3.1 5
0
1
2
3
4
5
NumberofDataPoints
2.1 2.3 2.5 2.7 2.9 3.1
28 April 9, 2016 – v3.0
1. The bell-shaped distribution:
Symmetrical shape with a peak in the
middle of the range of the data.
While deviation from a bell shape should
be investigated, such deviation is not
necessarily bad.
The Histogram – Typical Patterns of Variation
29 April 9, 2016 – v3.0
2. The double-peaked distribution:
A distinct valley in the middle of the range
of the data with peaks on either side.
This pattern is usually a combination of
two bell-shaped distributions and suggests
that two distinct processes are at work.
The Histogram – Typical Patterns of Variation
30 April 9, 2016 – v3.0
3. The plateau distribution:
A flat top with no distinct peak and slight
tails on either sides.
This pattern is likely to be the result of
many different bell-shaped distribution
with centers spread evenly throughout the
range of data.
The Histogram – Typical Patterns of Variation
31 April 9, 2016 – v3.0
4. The skewed distribution:
An asymmetrical shape in which the peak
is off-center in the range of the data and
the distribution tails off sharply on one
side and gently on the other.
This pattern typically occurs when a
practical limit, or a specification limit,
exists on one side and is relatively close
to the nominal value.
The Histogram – Typical Patterns of Variation
32 April 9, 2016 – v3.0
5. The truncated distribution:
An asymmetrical shape in which the peak
is at or near the edge of the range of the
data, and the distribution ends very
abruptly on one side and tails off gently on
the other.
This pattern often occurs if the process
includes a screening, 100 % inspection, or
a review process. Note that these
truncation efforts are an added cost and
are, therefore, good candidates for
removal.
The Histogram – Typical Patterns of Variation
33 April 9, 2016 – v3.0
The Histogram – The Bell-Shaped or Normal Distribution
34 April 9, 2016 – v3.0
The Histogram – Exercise 1
Distribution of Heights of U.S. Population:
Use the plot area below to construct a histogram from the
random sample of heights on the right:
59 66 63 70
60 66 69 70
65 62 71 72
68 65 67 69
65 66 70 68
64 64 73 73
63 67 71 68
63 68 70 68
65 67 64 71
61 64 70 72
70 63 68 68
68 63 66 66
64 63 67 74
63 62 66 68
62 62 67 70
35 April 9, 2016 – v3.0
Remarks or Questions ?!?
36 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
37 April 9, 2016 – v3.0
What are the Two Types of Product Characteristics ?
 Attribute: A characteristic that by comparison to some standard is judged
“Good” or “Bad” (free from scratches, fits, etc.)
 Variable: A characteristic measured in physical units (inches, volts, amps,
decibel, seconds, etc.)
 Variable characteristics are specified by designers as a nominal value
(target) and a tolerance around the target (variability).
 Manufacturing processes attempt to produce at the nominal value of the
characteristic.
 Since no process is perfect, variation of the characteristics will occur.
 Products with a product characteristic that falls inside the given
tolerances will be defined as “good” or “acceptable”.
 Products with a product characteristic that falls outside the given
tolerances will be defined as “bad” or “unacceptable”.
38 April 9, 2016 – v3.0
Relationship between Tolerance to Nominal Value
 The required value of a product or process characteristic is specified
as its nominal value.
 The maximum range of acceptable variation of the product or process
characteristic which will still work in the product determines the
tolerances about the nominal value.
Nominal Value
Specification
Upper
Spec. Limit (USL)
Lower
Spec. Limit (LSL) Tolerance
39 April 9, 2016 – v3.0
Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6
Measure of Location – The Sample Average
Definition:
N
xxx
x N

...21
8.4
5
24
5
62475


x
40 April 9, 2016 – v3.0
Example 1: x1 = 2 x2 = 5 x3 = 4
Definition: Order all data points from the smallest to largest. Then choose the
middle data point if the number of data points is odd, or the mean value of the
two middle data points if the number of data points is even.
Example 2: x1 = 5 x2 = 7 x3 = 4 x4 = 2
Example 3: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6 ?
Measure of Location – The Sample Median
2 – 5 – 4 2 – 4 – 5 median = 4
5 – 7 – 4 – 2 2 – 4 – 5 – 7 median = 4.5
41 April 9, 2016 – v3.0
Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6
Measure of Variability – The Sample Range
),...,,min(),...,,max( 2121 NN xxxxxxR 
Definition:
527)6,2,4,7,5min()6,2,4,7,5max( R
42 April 9, 2016 – v3.0
x3
x
average
_
x2
x1
x10
Measure of Variability – Sample Variance
     
9)110(
...
2
10
2
2
2
1
or
xxxxxx


Time
x6𝑥3 - 𝑥
𝑥2 - 𝑥
43 April 9, 2016 – v3.0
Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6
Measure of Variability – Sample Variance
     
)1(
...
22
2
2
12



N
xxxxxx
s N
          7.3
)15(
8.468.428.448.478.45
22222
2



s
Definition:
44 April 9, 2016 – v3.0
Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6
Measure of Variability – Sample Standard Deviation
     
)1(
...
22
2
2
1



N
xxxxxx
s N
LT
Definition:
          7.3
)15(
8.468.428.448.478.45
22222
2



LTs
92.17.3 LTs
45 April 9, 2016 – v3.0
Time t
Process
Characteristic
e.g. Hole Size
Process not in control
average
Subgroup size n = 5
Number of subgroups N = 7
Measure of Variability – The Principle of Subgrouping
46 April 9, 2016 – v3.0
Where
is the range of subgroup j, N the number of
subgroups, and d2 depends on the size n of a
subgroup (see handout).
sST , often notated as s or sigma, is another measure of
dispersion or variability and stands for “short-term
standard deviation”,
which measures the variability of a process or system
using “rational” subgrouping.
Measure of Variability – Standard Deviation sST
22
21 ...
dRd
N
RRR
s N
ST 


minmax XXRj 
n
2
3
4
5
6
7
8
9
10
d2
1.128
1.693
2.059
2.326
2.534
2.704
2.847
2.970
3.078
47 April 9, 2016 – v3.0
Time t
Process
Characteristic
e.g. Hole Size
Process not in control
average
Subgroup size n = 5
Number of subgroups N = 7
Measure of Variability – The Principle of Subgrouping
 sST stays the same, even if the process is not in control
 sLT increases over time because the process is not in control
 sST and sLT are identical if the process was in control
48 April 9, 2016 – v3.0
Long-term standard deviation:
Short-term standard deviation:
The difference between the standard deviations sLT and sST gives an
indication of how much better one can do when using appropriate
production control, like Statistical Process Control (SPC).
     
)1(
...
22
2
2
1



N
xxxxxx
s N
LT
Measure of Variability – Difference between sLT and sST
22
21 ...
dRd
N
RRR
s N
ST 


49 April 9, 2016 – v3.0
averageaverage
-1*s(igma)
average
-2*s(igma)
average
-3*s(igma)
average
+1*s(igma)
average
+2*s(igma)
average
+3*s(igma)
34.13 %34.13 %
13.60 % 13.60 %
2.14 %2.14 %
0.13 % 0.13 %
Measure of Variability – The Normal Distribution
If your process is under control, over 99.74% of your data points will fall between the
average ± 3s(igma) limits.
The distance between average ± 3s(igma) limits is called the width of the process or process capability.
Lower
Control Limit
Upper
Control Limit
50 April 9, 2016 – v3.0
Measure of Location and Variability – Exercise 2
Calculate the Mean Value or Average, Median, Range, and
long- and short-term Standard Deviation of the sample data.
You may copy the data into MS Excel and simplify the
calculations.
Group
1 59 66 63 62
2 60 66 69 65
3 65 62 71 72
4 68 65 67 69
5 65 66 70 68
6 64 64 73 73
7 63 67 71 68
8 63 68 65 68
MeasurementsOverall Mean Value =
Overall Median =
Subgroup Ranges =
Long-term Standard Deviation =
Short-term Standard Deviation =
Note: The Excel function for the Long-Term Standard Deviation is “stdev()”.
51 April 9, 2016 – v3.0
Measure of Location and Variability – Exercise 2 Results
Subgroup Median Range
1 59 66 63 62 63 7
2 60 66 69 65 66 9
3 65 62 71 72 68 10
4 68 65 67 69 68 4
5 65 66 70 68 67 5
6 64 64 73 73 69 9
7 63 67 71 68 68 8
8 63 68 65 68 67 5
Overall Range: 14
Overall Median: 66
Average Range: 7.1
Short-Term Standard Deviation: 3.46
Long-Term Standard Deviation: 3.55
Measurements
52 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
53 April 9, 2016 – v3.0
Attribute Data
(Count or Yes/No Data)
Variable Data
(Measurements)
Variable
subgroup
size
Subgroup
size
of 1
Fixed
subgroup
size
x chart
x-bar
R chart
x-bar
s chart
Count
Incidences or
nonconformities
Fixed
oppor-
tunity
Variable
oppor-
tunity
c - chart u - chart
Yes/No Data
Defectives or
nonconforming units
Fixed
subgroup
size
Variable
subgroup
size
np - chart p - chart
Process Control Charts – Types of Control Charts
Type of Data
54 April 9, 2016 – v3.0
 The x - chart is a method of looking at variation in a variable data
or measurement.
 One source is the variation in the individual data points over time.
This represents “long term” variation in the process.
 The second source of variation is the variation between
successive data points. This represents “short term” variation.
 Individual or x - charts should be used when there is only one data
point to represent a situation at a given time.
 To use the x - chart, the individual sample results should be
sufficient normally distributed. If not, the x - chart will give more
false signals.
Process Control Charts – The x - Chart
55 April 9, 2016 – v3.0
Upper control limit =
Lower control limit =
Upper control limit =
Lower control limit =
The x- chart
The R- chart
,
where x1, x2, ..., xN are the measurements, N the number of measurements,
, and .
Process Control Limit – The x - Chart
RxRxdRx  66.2128.133 2
RxRxdRx  66.2128.133 2
RRD  267.34
003  RRD
N
xxx
x N

...21
1
...32



N
RRR
R N
1 iii xxR
56 April 9, 2016 – v3.0
Process Control Charts – x - Chart Example
1 0 .0
C HA
No k
F ran
to
nt h
OBSERVATIONS7
9
1 1
1 3
1 5
1 7
1 9
2
1
1
2
EEEEEEEEEEEEEEEEEEEEEEEE
A
L
U
RANGES0
1
2
3
4
5
6
L
U
R
01.01.8701.02.8701.03.8701.04.8701.05.8701.06.8701.07.8701.08.8701.09.8701.10.8701.11.8701.12.8701.01.8801.02.8801.03.8801.04.8801.05.8801.06.8801.07.8801.08.8801.09.8801.10.8801.11.8801.12.88
1 00
G ro
A uto
C L O
C urv
K -S
A V E
P R O
UC L
L C L
re e n
128.12 RdRsST 
57 April 9, 2016 – v3.0
Process Control Charts – The Central Limit Theorem
Regardless of the shape of the distribution of a population, the
distribution of average values, x-bar’s, of subgroups of size n drawn
from that population will tend toward a normal distribution as the
subgroup size n becomes large.
Laplace and Gauss
The standard deviation sx of the subgroup averages is smaller than the
standard deviation s of the individual measurements. The relationship
between these two standard deviation s and sx as follows, where n is
the nuymber of measurements in each subgroup:
_
_
nssx

58 April 9, 2016 – v3.0
Process Control Charts – Exercise 3
Throw the Dice:
Step 1: Throw the dice 30 times and record the results in the table on the right.
Step 2: Draw a Histogram #1 of the 30 data points in one of the spreadsheets below.
Step 3: Calculate the average to 2 consecutive throws and draw the histogram #2 of the resulting 15 data
points.
What do you see and why?
AverageResults
Histogram #1 Histogram #2
59 April 9, 2016 – v3.0
 The (x-bar / R) - chart should be used if
 the individual measurements are not normally distributed,
 one can rationally subgroup the data and is interested in
detecting differences between the subgroups over time.
 The (x-bar / R) - chart is a method of looking at two different
sources of variation. One source is the variation in subgroup
averages. The other source is the variation within a subgroup.
 The x-bar - chart shows variation over time or long-term variation
and the R - chart is a measure of the short-term variation in the
process.
Process Control Charts – The (x-bar/R) - Chart
60 April 9, 2016 – v3.0
Upper control limit =
Lower control limit =
The R- chart
Upper control limit =
Lower control limit =
The x-bar - chart
where x-bar1, x-bar2, ..., x-barN are the averages of each subgroup, n the
number of items in a subgroup, N the number of subgroups,
., and
Process Control Limit – The x-bar/R - Chart
  RAxndRx  223
RD 4
RD 3
N
xxx
x
N

...21
N
RRR
R N

...21minmax
iii xxR 
  RAxndRx  223
61 April 9, 2016 – v3.0
n
2
3
4
5
6
A2
1.880
1.023
0.729
0.577
0.483
D3
0
0
0
0
0
D4
3.267
2.574
2.282
2.114
2.004
d2
1.128
1.693
2.059
2.326
2.534
Factors for x- and x-bar/R - Charts
62 April 9, 2016 – v3.0
1 0 .0
C HA
No k
F ran
5 -
o f
AVERAGES.0
.0
.0
.0
.0
.0
2
1
1
2
AA*
A
L
U
RANGES0
1 0
2 0
3 0
4 0
5 0
6 0
7 0
L
U
R
95.01.2095.01.2195.01.2395.01.2495.01.2595.01.2695.01.2795.01.2895.01.3095.01.3195.02.0195.02.0295.02.0395.02.0495.02.0695.02.0795.02.0895.02.0995.02.1095.02.1195.02.1395.02.2095.02.2195.02.22
INDIVIDUALS
6 .00 6 .0
G ro
A uto
C L O
C urv
K -S
A V E
P R O
UC L
L C L
e re e n
Process Control Charts – (x-bar/R) - Chart Example
63 April 9, 2016 – v3.0
 The (x-bar / s) - chart should be used instead the (x-bar / R) -
chart if the subgroup is larger than 10. In this case, the
standard deviation is a better measurement than the range for
the variation between individual measurements in a subgroup.
 The (x-bar / s) - chart can be used whenever one can use the
(x-bar / R) - chart.
 The (x-bar / s) - chart is a method of looking at sources of
variation. One chart looks at variation in the subgroup averages
x-bar. The other chart examines variation in the subgroups
standard deviation s.
Process Control Charts – The (x-bar/s) - Chart
64 April 9, 2016 – v3.0
Upper control limit =
Lower control limit =
Upper control limit =
Lower control limit =
The s- chart
The x-bar - chart
, and
where x-bar1, x-bar2, ..., x-barN are the averages of each subgroup, s1, s2, ...,
sN are the standard deviations of each subgroup, n the number of items in a
subgroup, N the number of subgroups,
.
Process Control Limit – The x-bar/s - Chart
sAx  3
sAx  3
sB 4
sB 3
N
xxx
x
N

...21
N
sss
s N

...21
65 April 9, 2016 – v3.0
n
6
7
8
9
10
A3
1.287
1.182
1.099
1.032
0.975
B3
0.030
0.118
0.185
0.239
0.284
B4
1.970
1.882
1.815
1.761
1.716
c4
0.9515
0.9594
0.9650
0.9693
0.9727
Factors for x-bar/s - Charts
66 April 9, 2016 – v3.0
Common Causes: Causes that are implemented in the process due
to the design of the process, and affect all outcomes of the process.
Identifying these types of causes requires Design of Experiment
(DOE) methods.
Special Causes: Causes that are not present in the process all the
time and do not affect all outcomes, but arise because of specific
circumstances. Special causes can be identified using SPC.
Walter A. Shewhart (1931)
Out-of-Control Criteria – Two Causes of Variation
67 April 9, 2016 – v3.0
Unstable Process: A process in which variation is a result of both
common and special causes.
Stable Process: A process in which variation in outcomes arises
only from common causes.
Out-of-Control Criteria – Two Types of Processes
68 April 9, 2016 – v3.0
An out-of-control criteria is a signal of a special causes of
variation:
• Is a systematic pattern of the product or process
characteristic monitored and charted
• Has a low probability of occurring when the process is
stable and in control
SPC Out-of-Control Criteria – The Types of Signals
69 April 9, 2016 – v3.0
What is the “chance”
to loose the coin flip
11 times in a row?
1 =
2 =
…
…
…
11 =
What is the “chance”
to loose the coin flip
11 times in a row?
1 = 50% or 0.50
2 =
…
…
…
11 =
What is the “chance”
to loose the coin flip
11 times in a row?
1 = 50% or 0.50
2 = 25% or 0.50*0.50
…
…
…
11 =
What is the “chance”
to loose the coin flip
11 times in a row?
1 = 50% or 0.50
2 = 25% or 0.50*0.50
…
…
…
11 = 0.049% or 0.5011
70 April 9, 2016 – v3.0
Out-of-Control Criteria – The Basic Idea
average average
+1*s(igma)
average
-1*s(igma)
average
+2*s(igma)
average
-2*s(igma)
average
-3*s(igma)
average
+3*s(igma)
34.13 %34.13 %
13.60 % 13.60 %
2.14 %2.14 %
0.13 % 0.13 %
If your process is under control, over 99.74% of your data points will fall between the
average ± 3s(sigma) limits and there is only a 0.13% that a measurement point would fall
outside these limits.
Lower
Control Limit
Upper
Control Limit
71 April 9, 2016 – v3.0
Process Out-of-Control Criteria
Below is a list of the most commonly used out-of-control criteria included
in Minitab 17 and as defined by Walter Shewhart in the 1920s.
Criteria 1: Outlier
Criteria 2 & 5 & 6: Process Shift
Criteria 3: Process Trend
72 April 9, 2016 – v3.0
SPC Criteria #1 – 1 Point above or below 3 Sigma
All SPC Out-of-Control Criteria have
about a 1 in 1,000 chance to occur in
a process without a special cause.
Therefore, they are strong evidence
for the presence of a special cause.
73 April 9, 2016 – v3.0
SPC Criteria #2 – 9 Points on the same Side of the Average
9 consecutive points above or below
the process performance average
line often indicates a shift in process
performance.
74 April 9, 2016 – v3.0
SPC Criteria #3 – 6 Consecutive Points Increasing or Decreasing
6 consecutive points increasing or
decreasing often indicates a trend in
process performance due to a
special cause.
75 April 9, 2016 – v3.0
SPC Criteria #5 – 2 of 3 Points above or below 2 Sigma
2 of 3 consecutive points above or
below 2 Sigma line often indicates
a shift in process performance.
76 April 9, 2016 – v3.0
SPC Criteria #6 – 4 of 5 Points above or below 1 Sigma
4 of 5 consecutive points above
or below the 2 Sigma line often
indicates a shift in process
performance.
77 April 9, 2016 – v3.0
Special Causes showing in the MR, R, or s Chart
1 data point above or below the 3
Sigma line is often the only
criteria used to identify special
causes in process performance.
78 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
79 April 9, 2016 – v3.0
Relationship between Tolerance and Process Capability
For any critical characteristic the Capability Index cp, also known as
the design margin, is quantified by :
ST
p
s
LSLUSL
LCL-UCL
USL - LSL
c




6
-
Variation)Process(TotalCapabilityorWidthProcess
Tolerance)(Designion WidthSpecificat
3
)-(
ST
p
s
TUSL
c

or
3
)-(
ST
p
s
LSLT
c

and with T = Target
80 April 9, 2016 – v3.0
The Capability Index cp
1.00 < cp < 1.33
Average
= Nominal Value or Target
LSL USLLCL UCL
average
+ 3*s (or sigma)
average
- 3*s (or sigma)
x
81 April 9, 2016 – v3.0
Process Performance Study
In some situations (e.g. order of production unknown, only small
sample produced) the short-term standard deviation sST cannot be
calculated and therefore the long-term standard deviation sLT has to
used. This means the actual process performance is evaluated, rather
than the potential process capability. To address this difference the
Performance Index pp is used instead of the Capability Index cp.
LT
p
s
LSLUSL
p


6
-
or
3
)-(
LT
p
s
TUSL
p


3
)-(
LT
p
s
LSLT
p

and with T = Target
82 April 9, 2016 – v3.0
Actual Process Performance
Cpk is the Capability Index adjusted for k, which considers actual
operation of the process and takes into account any difference
between the design nominal and the actual process mean value x-
bar.
)
3
-
,
3
-
(min
STST
pk
s
LSLx
s
xUSL
c


where is the average of all measurements or data points.x
83 April 9, 2016 – v3.0
The Capability Index cpk
nominal
value
LSL USLLCL UCL
average
1.33 < cp < 2.00, but 1.00 < cpk < 1.33
average
- 3*s(igma)
average
+ 3*s(igma)
x
84 April 9, 2016 – v3.0
Process Performance Study
In some situations (e.g. order of production unknown, only small sample
produced) the short-term standard deviation sST can not be calculated
and therefore the long-term standard deviation sLT has to used. This
means the actual process performance is evaluated, rather than the
potential process capability. To address this difference the Performance
Index ppk is used instead of the Capability Index cpk.
)
3
-
,
3
-
(min
LTLT
pk
s
LSLx
s
xUSL
p


85 April 9, 2016 – v3.0
ppk ccTx 
The Capability Index cpk
 Note, the cpk as defined above can also be used in the case of
unilateral specification, means that there exist only an upper or
lower specification limit and no target value.
Continue:
 Note, if the process average (x-bar) is equal to the midpoint
(nominal value) of the specification interval, means if
86 April 9, 2016 – v3.0
Quantifying Actual Process Performance
In some cases, the target value (T) of the process is not the
midpoint of the specification interval. For example the lower
specification limit may be the best value for the quality
characteristic. The Taguchi Capability Index cpm is a measure for
the difference between the average of the process and its target
value.
 22
6 Txs
LSLUSL
c
ST
pm



Note, if
ppm ccTx 
87 April 9, 2016 – v3.0
The Capability Index cpm
Nominal Value
or
Target
LSL USLLCL UCL
average
1.50 < cp < 2.00, 1.33 < cpk < 1.5, but 0.50 < cpm < 1.00
average
- 3*s(igma)
average
+ 3*s(igma)
x
88 April 9, 2016 – v3.0
Process Performance Study
In some situations (e.g. order of production unknown, only small
sample produced) the short-term standard deviation sST can not be
calculated and therefore the long-term standard deviation sLT has to
used. This means the actual process performance is evaluated,
rather than the potential process capability. To address this
difference the Performance Index ppm is used instead of the
Capability Index cpm.
 22
6 Txs
LSLUSL
p
LT
pm



Note, if
ppm ppTx 
89 April 9, 2016 – v3.0
Why are cp and cpk useful and significant Measures?
 To maximize cp requires the joint and concurrent effort of both product and process
designers.
 Product Design has the goal of increasing the allowable tolerance to the maximum
which will still permit successful function of the product.
 Process Design has the goal of minimizing the variability of the process which
reproduces the characteristic required for successful function of the product, and for
centering the process on target (nominal) value of the characteristic.
 A high cp Index indicates that the process is capable of reproducing the
characteristic. (It makes no statement about the centering of the process.)
 A high cpk Index indicates that the process is actually reproducing the characteristic
within the desired limits. (It makes no statement about the inherent capability, other
than its minimum value.)
 cpk < 1.00 means “Process not capable” 1.00
< cpk < 1.33 means “Process is marginal” cpk >
1.33 means “Process is capable”
90 April 9, 2016 – v3.0
Confidence Limits for the Capability Indices
The confidence intervals for the capability indices can be defined as
follows.
where m= n·N is number of measurements and (1-100% the required
confidence (e.g. for a confidence of 95 % is z/2 = 1.96).
   12
ˆ
ˆ
12
ˆ
ˆ 22




m
c
zcc
m
c
zc
p
pp
p
p 
   
2
2
2
2
2
22
2
2
2
2
2
1
2
1
ˆ
ˆ
1
2
1
ˆ
ˆ





 




 







 




 


s
Tx
s
Tx
m
c
zcc
s
Tx
s
Tx
m
c
zc
pm
pmpm
pm
pm 
    mm
c
zcc
mm
c
zc
pk
pkpk
pk
pk








9
1
12
ˆ
ˆ
9
1
12
ˆ
ˆ
2
2
2
2 
91 April 9, 2016 – v3.0
Confidence Limits for the Capability Indices
92 April 9, 2016 – v3.0
Confidence Limits for the Capability Indices
93 April 9, 2016 – v3.0
Confidence Limits for the Capability Index cp
95% confidence interval
0.0000
0.2000
0.4000
0.6000
0.8000
1.0000
1.2000
1.4000
1.6000
0 100 200 300 400 500 600
m=N*n
delta
cp=0.5
cp=1.0
cp=1.5
cp=2.0
N = # of Subgroups
n = Subgroup Size
94 April 9, 2016 – v3.0
Confidence Limits for the Capability Index cpk
95% confidence interval
0.0000
0.2000
0.4000
0.6000
0.8000
1.0000
1.2000
1.4000
1.6000
0 100 200 300 400 500 600
m=N*n
delta
cpk=0.5
cpk=1.0
cpk=1.5
cpk=2.0
N = # of Subgroups
n = Subgroup Size
95 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
96 April 9, 2016 – v3.0
Treatment of Non-Normality - Skewness
Transformation of the original measurements using
“ The Power Ladder ”.

originaltransform xx 
This method works best when the ratio of the largest to smallest
measurement value is greater than zero. If this is not the case, a simple
transformation like
with a appropriate constant “c” should work.
cxx originaltransform 
97 April 9, 2016 – v3.0
Treatment of Non-Normality - Skewness
6543210-1
Upper SpecLower Spec
s
Mean-3s
Mean+3s
Mean
n
k
LSL
USL
Targ
Cpm
Cpk
CPL
CPU
Cp
Short-Term Capability
0
94876
30000
7429
0.00
9.49
3.00
0.74
Obs
PPM<LSL Exp
Obs
PPM>U SL Exp
Obs
%<LSL Exp
Obs
%>USL Exp
0.80061
-1.35198
3.45169
1.04985
100.000
0.300
0.000
3.000
*
*
0.44
0.44
0.81
0.62
Process Capability Analysis for Original
20100
2.4
1.6
0.8
0.0
Xbar and R Chart
Subgr
Means
X=1.050
3.0SL=2.124
-3.0SL=-0.02428
6
4
2
0
Ranges
1
R=1.862
3.0SL=3.938
-3.0SL=0.000
20100
Last 20 Subgroups
6
4
2
0
SubgroupNumber
Values
30
3.45169-1.35198
Cp: 0.62
CPU: 0.81
CPL: 0.44
Cpk: 0.44
Capability Plot
Process Tolerance
Specifications
StDev: 0.800612
III
III
5.02.50.0
Normal Prob Plot
5.02.50.0
Capability Histogram
Process Capability Sixpack for Original
“The Original Measurements“
The Capability Study indicates
almost 100 000 PPM below the
lower specification limit.
98 April 9, 2016 – v3.0
Treatment of Non-Normality - Skewness
3.02.52.01.51.00.50.0-0.5-1.0
8
7
6
5
4
3
2
1
0
95% Confidence Interval
StDev
Lambda
Last IterationInfo
0.693
0.692
0.695
0.393
0.337
0.281
StDevLambda
Up
Est
Low
Box-CoxPlot for Original

33.0
33.0
33.0
originaltransform
originaltransform
originaltransform
USLUSL
LSLLSL
xx



“The Power Ladder“
Which function would transform
the original data to a normal
distribution ?
99 April 9, 2016 – v3.0
Treatment of Non-Normality - Skewness
2.001.751.501.251.000.750.500.250.00
UpperSpecLowerSpec
s
Mean-3s
Mean+3s
Mean
n
k
LSL
USL
Targ
Cpm
Cpk
CPL
CPU
Cp
Short-TermCapability
0
485
30000
41141
0.00
0.05
3.00
4.11
Obs
PPM<LSL Exp
Obs
PPM>USL Exp
Obs
%<LSL Exp
Obs
%>USL Exp
0.28635
0.08559
1.80369
0.94464
100.000
0.310
0.000
1.442
*
*
0.58
1.10
0.58
0.84
Process CapabilityAnalysis for Transfor
20100
1.25
1.00
0.75
0.50
Xbar and R Chart
Subgr
Means
X=0.9446
3.0SL=1.329
-3.0SL=0.5605
1.5
1.0
0.5
0.0
Ranges
R=0.6660
3.0SL=1.408
-3.0SL=0.000
20100
Last 20 Subgroups
1.7
1.2
0.7
0.2
SubgroupNumber
Values
1.44220.0000
1.803690.08559
Cp: 0.84
CPU: 0.58
CPL: 1.10
Cpk: 0.58
Capability Plot
Process Tolerance
Specifications
StDev: 0.286349
III
III
1.51.00.5
Normal Prob Plot
1.51.00.5
Capability Histogram
Process Capability Sixpack for Transfor
“The Transformed Measurements“
The Capability Study for the transformed
data indicates “only” 500 PPM below the
lower specification limit !
100 April 9, 2016 – v3.0
Section 1: Introduction
Section 2: The Histogram
Section 3: Basic Statistics and Process Capability
Section 4: Introduction to Statistical Process Control
Section 5: Definitions of Process Capability Indices
Section 6: Non-Normal Distributed Processes
Process Capability Study – Table of Contents
101 April 9, 2016 – v3.0
The End …
“Perfection is not attainable, but if we chase perfection we can catch
excellence.” - Vince Lombardi
102 April 9, 2016 – v3.0
Terms & Conditions
After you have downloaded the training material to your own
computer, you can change any part of the course material and
remove all logos and references to Operational Excellence
Consulting. You can share the material with your colleagues and
re-use it as you need. The main restriction is that you cannot
distribute, sell, rent or license the material as though it is your
own. These training course materials are for your — and
your organization's — usage only. Thank you.

More Related Content

What's hot

What's hot (20)

Basics of Process Capability
Basics of Process CapabilityBasics of Process Capability
Basics of Process Capability
 
Spaghetti Chart
Spaghetti ChartSpaghetti Chart
Spaghetti Chart
 
SMED - Single Minute Exchange of Dies
SMED - Single Minute Exchange of DiesSMED - Single Minute Exchange of Dies
SMED - Single Minute Exchange of Dies
 
The Basics 7 QC Tools - ADDVALUE - Nilesh Arora
The Basics 7 QC Tools - ADDVALUE - Nilesh AroraThe Basics 7 QC Tools - ADDVALUE - Nilesh Arora
The Basics 7 QC Tools - ADDVALUE - Nilesh Arora
 
Value Stream Mapping
Value Stream MappingValue Stream Mapping
Value Stream Mapping
 
Spc
SpcSpc
Spc
 
Value Stream Mapping
Value Stream MappingValue Stream Mapping
Value Stream Mapping
 
Measurement System Analysis (MSA)
Measurement System Analysis (MSA)Measurement System Analysis (MSA)
Measurement System Analysis (MSA)
 
Alternate Hourly Lean Introduction
Alternate Hourly Lean IntroductionAlternate Hourly Lean Introduction
Alternate Hourly Lean Introduction
 
Attribute MSA
Attribute MSAAttribute MSA
Attribute MSA
 
Training ppt for control plan
Training ppt for control plan   Training ppt for control plan
Training ppt for control plan
 
Kanban
KanbanKanban
Kanban
 
13. value stream mapping
13. value stream mapping13. value stream mapping
13. value stream mapping
 
APQP
APQPAPQP
APQP
 
7 QC Tools training presentation
7 QC Tools training presentation7 QC Tools training presentation
7 QC Tools training presentation
 
Vsm
VsmVsm
Vsm
 
TPM: Quality Maintenance (Hinshitsu Hozen)
TPM: Quality Maintenance (Hinshitsu Hozen)TPM: Quality Maintenance (Hinshitsu Hozen)
TPM: Quality Maintenance (Hinshitsu Hozen)
 
Value stream mapping - Future State
Value stream mapping - Future StateValue stream mapping - Future State
Value stream mapping - Future State
 
Six sigma
Six sigmaSix sigma
Six sigma
 
SMED-Observation Training
SMED-Observation TrainingSMED-Observation Training
SMED-Observation Training
 

Viewers also liked

Six Sigma : Process Capability
Six Sigma : Process CapabilitySix Sigma : Process Capability
Six Sigma : Process CapabilityLalit Padekar
 
Six Sigma the best ppt
Six Sigma the best pptSix Sigma the best ppt
Six Sigma the best pptRabia Sgh S
 
Miniature circuit breaker construction, can't be more specific any more!
Miniature circuit breaker construction, can't be more specific any more! Miniature circuit breaker construction, can't be more specific any more!
Miniature circuit breaker construction, can't be more specific any more! Guangzhou Shilin Electrical Co., Ltd.
 
EATON Miniature circuit breaker
EATON Miniature circuit breakerEATON Miniature circuit breaker
EATON Miniature circuit breakerMeena Tanwar
 
Total Quality Human Resource Management(2)
Total Quality Human Resource Management(2)Total Quality Human Resource Management(2)
Total Quality Human Resource Management(2)ahmad bassiouny
 
Ops A La Carte Statistical Process Control (SPC) Seminar
Ops A La Carte Statistical Process Control (SPC) SeminarOps A La Carte Statistical Process Control (SPC) Seminar
Ops A La Carte Statistical Process Control (SPC) SeminarJay Muns
 
Statistical process control (spc)
Statistical process control (spc)Statistical process control (spc)
Statistical process control (spc)Dinah Faye Indino
 
Process capability
Process capabilityProcess capability
Process capabilitypadam nagar
 
6. process capability analysis (variable data)
6. process capability analysis (variable data)6. process capability analysis (variable data)
6. process capability analysis (variable data)Hakeem-Ur- Rehman
 
Lecture 2: Creativity Development
Lecture 2: Creativity DevelopmentLecture 2: Creativity Development
Lecture 2: Creativity DevelopmentTathagat Varma
 
Implementing lean Six sigma
Implementing lean Six sigmaImplementing lean Six sigma
Implementing lean Six sigmasanobar77
 
Chap 9 A Process Capability & Spc Hk
Chap 9 A Process Capability & Spc HkChap 9 A Process Capability & Spc Hk
Chap 9 A Process Capability & Spc Hkajithsrc
 
Introduction To Statistical Process Control 20 Jun 2011
Introduction To Statistical Process Control 20 Jun  2011Introduction To Statistical Process Control 20 Jun  2011
Introduction To Statistical Process Control 20 Jun 2011Robert_Wade_Sherrill
 
Kanban Basics for Beginners
Kanban Basics for BeginnersKanban Basics for Beginners
Kanban Basics for BeginnersZsolt Fabok
 
Six Sigma & Process Capability
Six Sigma & Process CapabilitySix Sigma & Process Capability
Six Sigma & Process CapabilityEric Blumenfeld
 
Process capability analysis
Process capability analysisProcess capability analysis
Process capability analysisSurya Teja
 
Measuremen Systems Analysis Training Module
Measuremen Systems Analysis Training ModuleMeasuremen Systems Analysis Training Module
Measuremen Systems Analysis Training ModuleFrank-G. Adler
 
Six Sigma and Its Implementation
Six Sigma and Its ImplementationSix Sigma and Its Implementation
Six Sigma and Its ImplementationAnsar Lawi
 
10. measurement system analysis (msa)
10. measurement system analysis (msa)10. measurement system analysis (msa)
10. measurement system analysis (msa)Hakeem-Ur- Rehman
 

Viewers also liked (20)

Six Sigma : Process Capability
Six Sigma : Process CapabilitySix Sigma : Process Capability
Six Sigma : Process Capability
 
Six Sigma the best ppt
Six Sigma the best pptSix Sigma the best ppt
Six Sigma the best ppt
 
Miniature circuit breaker construction, can't be more specific any more!
Miniature circuit breaker construction, can't be more specific any more! Miniature circuit breaker construction, can't be more specific any more!
Miniature circuit breaker construction, can't be more specific any more!
 
EATON Miniature circuit breaker
EATON Miniature circuit breakerEATON Miniature circuit breaker
EATON Miniature circuit breaker
 
Total Quality Human Resource Management(2)
Total Quality Human Resource Management(2)Total Quality Human Resource Management(2)
Total Quality Human Resource Management(2)
 
Ops A La Carte Statistical Process Control (SPC) Seminar
Ops A La Carte Statistical Process Control (SPC) SeminarOps A La Carte Statistical Process Control (SPC) Seminar
Ops A La Carte Statistical Process Control (SPC) Seminar
 
Statistical process control (spc)
Statistical process control (spc)Statistical process control (spc)
Statistical process control (spc)
 
Process capability
Process capabilityProcess capability
Process capability
 
6. process capability analysis (variable data)
6. process capability analysis (variable data)6. process capability analysis (variable data)
6. process capability analysis (variable data)
 
Lecture 2: Creativity Development
Lecture 2: Creativity DevelopmentLecture 2: Creativity Development
Lecture 2: Creativity Development
 
Implementing lean Six sigma
Implementing lean Six sigmaImplementing lean Six sigma
Implementing lean Six sigma
 
Chap 9 A Process Capability & Spc Hk
Chap 9 A Process Capability & Spc HkChap 9 A Process Capability & Spc Hk
Chap 9 A Process Capability & Spc Hk
 
Introduction To Statistical Process Control 20 Jun 2011
Introduction To Statistical Process Control 20 Jun  2011Introduction To Statistical Process Control 20 Jun  2011
Introduction To Statistical Process Control 20 Jun 2011
 
Kanban Basics for Beginners
Kanban Basics for BeginnersKanban Basics for Beginners
Kanban Basics for Beginners
 
Six Sigma & Process Capability
Six Sigma & Process CapabilitySix Sigma & Process Capability
Six Sigma & Process Capability
 
Process capability analysis
Process capability analysisProcess capability analysis
Process capability analysis
 
Measuremen Systems Analysis Training Module
Measuremen Systems Analysis Training ModuleMeasuremen Systems Analysis Training Module
Measuremen Systems Analysis Training Module
 
Six Sigma and Its Implementation
Six Sigma and Its ImplementationSix Sigma and Its Implementation
Six Sigma and Its Implementation
 
Final Prresentation
Final PrresentationFinal Prresentation
Final Prresentation
 
10. measurement system analysis (msa)
10. measurement system analysis (msa)10. measurement system analysis (msa)
10. measurement system analysis (msa)
 

Similar to Six Sigma Process Capability Study (PCS) Training Module

ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...
ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...
ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...ijmvsc
 
Lean Six Sigma Mistake-Proofing Process Training Module
Lean Six Sigma Mistake-Proofing Process Training ModuleLean Six Sigma Mistake-Proofing Process Training Module
Lean Six Sigma Mistake-Proofing Process Training ModuleFrank-G. Adler
 
Missing Parts I don’t think you understood the assignment.docx
Missing Parts I don’t think you understood the assignment.docxMissing Parts I don’t think you understood the assignment.docx
Missing Parts I don’t think you understood the assignment.docxannandleola
 
7qctools173-100325112603-phpapp01.ppt
7qctools173-100325112603-phpapp01.ppt7qctools173-100325112603-phpapp01.ppt
7qctools173-100325112603-phpapp01.pptOswaldo Gonzales
 
Hızlı Ozet - Istatistiksel Proses Kontrol
Hızlı Ozet - Istatistiksel Proses KontrolHızlı Ozet - Istatistiksel Proses Kontrol
Hızlı Ozet - Istatistiksel Proses Kontrolmetallicaslayer
 
1. (25 points) Temperature, Pressure and yield on a chemical .docx
1. (25 points) Temperature, Pressure and yield on a chemical .docx1. (25 points) Temperature, Pressure and yield on a chemical .docx
1. (25 points) Temperature, Pressure and yield on a chemical .docxaulasnilda
 
7 QC Tools and Problem Solving Presentation.pdf
7 QC Tools and Problem Solving Presentation.pdf7 QC Tools and Problem Solving Presentation.pdf
7 QC Tools and Problem Solving Presentation.pdfAzizOUBBAD1
 
7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf
7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf
7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdfEngFaisalAlrai
 
6 sigma-process-mapping-1233778357447980-3
6 sigma-process-mapping-1233778357447980-36 sigma-process-mapping-1233778357447980-3
6 sigma-process-mapping-1233778357447980-3jseets
 
CHAPTER 4 SQC.pptx
CHAPTER 4 SQC.pptxCHAPTER 4 SQC.pptx
CHAPTER 4 SQC.pptxhinal10
 
qc-tools.ppt
qc-tools.pptqc-tools.ppt
qc-tools.pptAlpharoot
 
7qc_tools_173.ppt for 7 QC tools implementation of the quality
7qc_tools_173.ppt for 7 QC tools implementation of the quality7qc_tools_173.ppt for 7 QC tools implementation of the quality
7qc_tools_173.ppt for 7 QC tools implementation of the qualityMANISHDUBEY14500
 
Six Sigma - Statistical Process Control (SPC)
Six Sigma - Statistical Process Control (SPC)Six Sigma - Statistical Process Control (SPC)
Six Sigma - Statistical Process Control (SPC)Flevy.com Best Practices
 

Similar to Six Sigma Process Capability Study (PCS) Training Module (20)

ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...
ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...
ANALYZING THE PROCESS CAPABILITY FOR AN AUTO MANUAL TRANSMISSION BASE PLATE M...
 
Lean Six Sigma Mistake-Proofing Process Training Module
Lean Six Sigma Mistake-Proofing Process Training ModuleLean Six Sigma Mistake-Proofing Process Training Module
Lean Six Sigma Mistake-Proofing Process Training Module
 
Training Module
Training ModuleTraining Module
Training Module
 
Missing Parts I don’t think you understood the assignment.docx
Missing Parts I don’t think you understood the assignment.docxMissing Parts I don’t think you understood the assignment.docx
Missing Parts I don’t think you understood the assignment.docx
 
7qc Tools 173
7qc Tools 1737qc Tools 173
7qc Tools 173
 
7qc Tools 173
7qc Tools 1737qc Tools 173
7qc Tools 173
 
ProjectReport_SPCinAM
ProjectReport_SPCinAMProjectReport_SPCinAM
ProjectReport_SPCinAM
 
7qctools173-100325112603-phpapp01.ppt
7qctools173-100325112603-phpapp01.ppt7qctools173-100325112603-phpapp01.ppt
7qctools173-100325112603-phpapp01.ppt
 
Hızlı Ozet - Istatistiksel Proses Kontrol
Hızlı Ozet - Istatistiksel Proses KontrolHızlı Ozet - Istatistiksel Proses Kontrol
Hızlı Ozet - Istatistiksel Proses Kontrol
 
1. (25 points) Temperature, Pressure and yield on a chemical .docx
1. (25 points) Temperature, Pressure and yield on a chemical .docx1. (25 points) Temperature, Pressure and yield on a chemical .docx
1. (25 points) Temperature, Pressure and yield on a chemical .docx
 
7 QC Tools and Problem Solving Presentation.pdf
7 QC Tools and Problem Solving Presentation.pdf7 QC Tools and Problem Solving Presentation.pdf
7 QC Tools and Problem Solving Presentation.pdf
 
7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf
7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf
7_QC_Tools_and_Problem_Solving_Presentation_1656881575.pdf
 
6 sigma-process-mapping-1233778357447980-3
6 sigma-process-mapping-1233778357447980-36 sigma-process-mapping-1233778357447980-3
6 sigma-process-mapping-1233778357447980-3
 
CHAPTER 4 SQC.pptx
CHAPTER 4 SQC.pptxCHAPTER 4 SQC.pptx
CHAPTER 4 SQC.pptx
 
qc-tools.ppt
qc-tools.pptqc-tools.ppt
qc-tools.ppt
 
7qc_tools_173.ppt for 7 QC tools implementation of the quality
7qc_tools_173.ppt for 7 QC tools implementation of the quality7qc_tools_173.ppt for 7 QC tools implementation of the quality
7qc_tools_173.ppt for 7 QC tools implementation of the quality
 
VSM
VSMVSM
VSM
 
Six Sigma - Statistical Process Control (SPC)
Six Sigma - Statistical Process Control (SPC)Six Sigma - Statistical Process Control (SPC)
Six Sigma - Statistical Process Control (SPC)
 
1b7 quality control
1b7 quality control1b7 quality control
1b7 quality control
 
What's new in Design-Expert version 9?
 What's new in  Design-Expert version 9? What's new in  Design-Expert version 9?
What's new in Design-Expert version 9?
 

More from Frank-G. Adler

Six Sigma Confidence Interval Analysis (CIA) Training Module
Six Sigma Confidence Interval Analysis (CIA) Training ModuleSix Sigma Confidence Interval Analysis (CIA) Training Module
Six Sigma Confidence Interval Analysis (CIA) Training ModuleFrank-G. Adler
 
Lean Quick Changeover (SMED) Training Module
Lean Quick Changeover (SMED) Training ModuleLean Quick Changeover (SMED) Training Module
Lean Quick Changeover (SMED) Training ModuleFrank-G. Adler
 
Lean Kanban Systems Training Module
Lean Kanban Systems Training ModuleLean Kanban Systems Training Module
Lean Kanban Systems Training ModuleFrank-G. Adler
 
How to achieve Operational Excellence?
How to achieve Operational Excellence?How to achieve Operational Excellence?
How to achieve Operational Excellence?Frank-G. Adler
 
Lean Standard or Standardized Work Training Module
Lean Standard or Standardized Work Training ModuleLean Standard or Standardized Work Training Module
Lean Standard or Standardized Work Training ModuleFrank-G. Adler
 
Global 8D Problem Solving Process Training Module
Global 8D Problem Solving Process Training ModuleGlobal 8D Problem Solving Process Training Module
Global 8D Problem Solving Process Training ModuleFrank-G. Adler
 
Strategy Planning and Deployment Process Training Module
Strategy Planning and Deployment Process Training ModuleStrategy Planning and Deployment Process Training Module
Strategy Planning and Deployment Process Training ModuleFrank-G. Adler
 
Business Process Improvement (BPI 7) Process Training Module
Business Process Improvement (BPI 7) Process Training ModuleBusiness Process Improvement (BPI 7) Process Training Module
Business Process Improvement (BPI 7) Process Training ModuleFrank-G. Adler
 
Lean 5S Visual Workplace Organization Training Module
Lean 5S Visual Workplace Organization Training ModuleLean 5S Visual Workplace Organization Training Module
Lean 5S Visual Workplace Organization Training ModuleFrank-G. Adler
 
Lean Value Stream Mapping (VSM) Training Module
Lean Value Stream Mapping (VSM) Training ModuleLean Value Stream Mapping (VSM) Training Module
Lean Value Stream Mapping (VSM) Training ModuleFrank-G. Adler
 
Balanced Scorecard Deployment Process Training Module
Balanced Scorecard Deployment Process Training ModuleBalanced Scorecard Deployment Process Training Module
Balanced Scorecard Deployment Process Training ModuleFrank-G. Adler
 

More from Frank-G. Adler (11)

Six Sigma Confidence Interval Analysis (CIA) Training Module
Six Sigma Confidence Interval Analysis (CIA) Training ModuleSix Sigma Confidence Interval Analysis (CIA) Training Module
Six Sigma Confidence Interval Analysis (CIA) Training Module
 
Lean Quick Changeover (SMED) Training Module
Lean Quick Changeover (SMED) Training ModuleLean Quick Changeover (SMED) Training Module
Lean Quick Changeover (SMED) Training Module
 
Lean Kanban Systems Training Module
Lean Kanban Systems Training ModuleLean Kanban Systems Training Module
Lean Kanban Systems Training Module
 
How to achieve Operational Excellence?
How to achieve Operational Excellence?How to achieve Operational Excellence?
How to achieve Operational Excellence?
 
Lean Standard or Standardized Work Training Module
Lean Standard or Standardized Work Training ModuleLean Standard or Standardized Work Training Module
Lean Standard or Standardized Work Training Module
 
Global 8D Problem Solving Process Training Module
Global 8D Problem Solving Process Training ModuleGlobal 8D Problem Solving Process Training Module
Global 8D Problem Solving Process Training Module
 
Strategy Planning and Deployment Process Training Module
Strategy Planning and Deployment Process Training ModuleStrategy Planning and Deployment Process Training Module
Strategy Planning and Deployment Process Training Module
 
Business Process Improvement (BPI 7) Process Training Module
Business Process Improvement (BPI 7) Process Training ModuleBusiness Process Improvement (BPI 7) Process Training Module
Business Process Improvement (BPI 7) Process Training Module
 
Lean 5S Visual Workplace Organization Training Module
Lean 5S Visual Workplace Organization Training ModuleLean 5S Visual Workplace Organization Training Module
Lean 5S Visual Workplace Organization Training Module
 
Lean Value Stream Mapping (VSM) Training Module
Lean Value Stream Mapping (VSM) Training ModuleLean Value Stream Mapping (VSM) Training Module
Lean Value Stream Mapping (VSM) Training Module
 
Balanced Scorecard Deployment Process Training Module
Balanced Scorecard Deployment Process Training ModuleBalanced Scorecard Deployment Process Training Module
Balanced Scorecard Deployment Process Training Module
 

Recently uploaded

Kenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith PereraKenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith Pereraictsugar
 
Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.Anamaria Contreras
 
Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Peter Ward
 
8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCR8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCRashishs7044
 
Market Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 EditionMarket Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 EditionMintel Group
 
NewBase 19 April 2024 Energy News issue - 1717 by Khaled Al Awadi.pdf
NewBase  19 April  2024  Energy News issue - 1717 by Khaled Al Awadi.pdfNewBase  19 April  2024  Energy News issue - 1717 by Khaled Al Awadi.pdf
NewBase 19 April 2024 Energy News issue - 1717 by Khaled Al Awadi.pdfKhaled Al Awadi
 
8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCR8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCRashishs7044
 
Financial-Statement-Analysis-of-Coca-cola-Company.pptx
Financial-Statement-Analysis-of-Coca-cola-Company.pptxFinancial-Statement-Analysis-of-Coca-cola-Company.pptx
Financial-Statement-Analysis-of-Coca-cola-Company.pptxsaniyaimamuddin
 
Investment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy CheruiyotInvestment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy Cheruiyotictsugar
 
Buy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail AccountsBuy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail AccountsBuy Verified Accounts
 
Digital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdfDigital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdfJos Voskuil
 
8447779800, Low rate Call girls in Rohini Delhi NCR
8447779800, Low rate Call girls in Rohini Delhi NCR8447779800, Low rate Call girls in Rohini Delhi NCR
8447779800, Low rate Call girls in Rohini Delhi NCRashishs7044
 
Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Seta Wicaksana
 
TriStar Gold Corporate Presentation - April 2024
TriStar Gold Corporate Presentation - April 2024TriStar Gold Corporate Presentation - April 2024
TriStar Gold Corporate Presentation - April 2024Adnet Communications
 
Annual General Meeting Presentation Slides
Annual General Meeting Presentation SlidesAnnual General Meeting Presentation Slides
Annual General Meeting Presentation SlidesKeppelCorporation
 
Call Us 📲8800102216📞 Call Girls In DLF City Gurgaon
Call Us 📲8800102216📞 Call Girls In DLF City GurgaonCall Us 📲8800102216📞 Call Girls In DLF City Gurgaon
Call Us 📲8800102216📞 Call Girls In DLF City Gurgaoncallgirls2057
 
Memorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQMMemorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQMVoces Mineras
 
Flow Your Strategy at Flight Levels Day 2024
Flow Your Strategy at Flight Levels Day 2024Flow Your Strategy at Flight Levels Day 2024
Flow Your Strategy at Flight Levels Day 2024Kirill Klimov
 

Recently uploaded (20)

Kenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith PereraKenya Coconut Production Presentation by Dr. Lalith Perera
Kenya Coconut Production Presentation by Dr. Lalith Perera
 
Corporate Profile 47Billion Information Technology
Corporate Profile 47Billion Information TechnologyCorporate Profile 47Billion Information Technology
Corporate Profile 47Billion Information Technology
 
Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.Traction part 2 - EOS Model JAX Bridges.
Traction part 2 - EOS Model JAX Bridges.
 
Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...Fordham -How effective decision-making is within the IT department - Analysis...
Fordham -How effective decision-making is within the IT department - Analysis...
 
8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCR8447779800, Low rate Call girls in Tughlakabad Delhi NCR
8447779800, Low rate Call girls in Tughlakabad Delhi NCR
 
Market Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 EditionMarket Sizes Sample Report - 2024 Edition
Market Sizes Sample Report - 2024 Edition
 
NewBase 19 April 2024 Energy News issue - 1717 by Khaled Al Awadi.pdf
NewBase  19 April  2024  Energy News issue - 1717 by Khaled Al Awadi.pdfNewBase  19 April  2024  Energy News issue - 1717 by Khaled Al Awadi.pdf
NewBase 19 April 2024 Energy News issue - 1717 by Khaled Al Awadi.pdf
 
8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCR8447779800, Low rate Call girls in Saket Delhi NCR
8447779800, Low rate Call girls in Saket Delhi NCR
 
Financial-Statement-Analysis-of-Coca-cola-Company.pptx
Financial-Statement-Analysis-of-Coca-cola-Company.pptxFinancial-Statement-Analysis-of-Coca-cola-Company.pptx
Financial-Statement-Analysis-of-Coca-cola-Company.pptx
 
Investment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy CheruiyotInvestment in The Coconut Industry by Nancy Cheruiyot
Investment in The Coconut Industry by Nancy Cheruiyot
 
Japan IT Week 2024 Brochure by 47Billion (English)
Japan IT Week 2024 Brochure by 47Billion (English)Japan IT Week 2024 Brochure by 47Billion (English)
Japan IT Week 2024 Brochure by 47Billion (English)
 
Buy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail AccountsBuy gmail accounts.pdf Buy Old Gmail Accounts
Buy gmail accounts.pdf Buy Old Gmail Accounts
 
Digital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdfDigital Transformation in the PLM domain - distrib.pdf
Digital Transformation in the PLM domain - distrib.pdf
 
8447779800, Low rate Call girls in Rohini Delhi NCR
8447779800, Low rate Call girls in Rohini Delhi NCR8447779800, Low rate Call girls in Rohini Delhi NCR
8447779800, Low rate Call girls in Rohini Delhi NCR
 
Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...Ten Organizational Design Models to align structure and operations to busines...
Ten Organizational Design Models to align structure and operations to busines...
 
TriStar Gold Corporate Presentation - April 2024
TriStar Gold Corporate Presentation - April 2024TriStar Gold Corporate Presentation - April 2024
TriStar Gold Corporate Presentation - April 2024
 
Annual General Meeting Presentation Slides
Annual General Meeting Presentation SlidesAnnual General Meeting Presentation Slides
Annual General Meeting Presentation Slides
 
Call Us 📲8800102216📞 Call Girls In DLF City Gurgaon
Call Us 📲8800102216📞 Call Girls In DLF City GurgaonCall Us 📲8800102216📞 Call Girls In DLF City Gurgaon
Call Us 📲8800102216📞 Call Girls In DLF City Gurgaon
 
Memorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQMMemorándum de Entendimiento (MoU) entre Codelco y SQM
Memorándum de Entendimiento (MoU) entre Codelco y SQM
 
Flow Your Strategy at Flight Levels Day 2024
Flow Your Strategy at Flight Levels Day 2024Flow Your Strategy at Flight Levels Day 2024
Flow Your Strategy at Flight Levels Day 2024
 

Six Sigma Process Capability Study (PCS) Training Module

  • 1. 1 April 9, 2016 – v3.0 Six Sigma Process Capability Study (PCS) by Operational Excellence Consulting LLC
  • 2. 2 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 3. 3 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 4. 4 April 9, 2016 – v3.0 How to get EVIDENCE to the following questions ?  Case 1: RF Design specifies a critical capacitor in a design with (2 ± 0.1)F. Is a particular supplier able to provide us with this type of component? How well can the supplier meet the target/nominal value of 2F?  Case 2: Mechanical Design specifies the distance between two holes by (32 ± 0.5)mm. Is a particular supplier able to provide us with the required accuracy? How well can the supplier meet the target/nominal value of 32mm?  Case 3: The GSM Standard requires that the test “Peak TX Power” falls between 28.3dB and 32.0dB. Is the new product able to fulfill this requirement?  Case 4: A fine pitch component requires a placement accuracy of ± 0.5mm. Is the existing placement machine able to fulfill this requirement?
  • 5. 5 April 9, 2016 – v3.0 The process and quality control methods and techniques used today got their start in the American Civil War at around 1789, when Eli Whitney took a contract from the U.S. Army for the manufacture of 10,000 rifles at the unbelievably low price of $13.40 each. At that time most of the products were handmade by small owner- managed shops and product parts were thus not interchangeable. The result of Whitney’s first mass production trail was that the rifles did not work as well as the handmade rifles. In addition, the copied parts did not fit as expected. The History of Statistical and Process Thinking
  • 6. 6 April 9, 2016 – v3.0 GO - Test NO-GO - Test The first time that one presented machine produced parts was 1851 at the industry exhibition in the Crystal Palace in London. An American gun smith took 10 working guns, took them apart, mixed all the parts in a box and re-assembled them again. This was found a quite surprising “experiment”. The History of Statistical and Process Thinking
  • 7. 7 April 9, 2016 – v3.0 Process Inspection Good Bad Repair Scrap + Monitor/Adjust The Traditional Production Concept The Detection Control Scheme
  • 8. 8 April 9, 2016 – v3.0 • The traditional production concept does not help us to produce only good products. • Every product has to be inspected. • Products have to be repaired or even scraped. • With respect to productivity and efficiency every activity after the actual production process is a non-value added activity. The Traditional Production Concept
  • 9. 9 April 9, 2016 – v3.0 Prevention Control Scheme Process Inspection Good Bad Repair Scrap + The Traditional Production Concept
  • 10. 10 April 9, 2016 – v3.0 Prevention Control Scheme Process Inspection Good Bad Repair Scrap + An Advanced Production Concept Monitor/Adjust Learn/Improve Selective measurement • Product • Process
  • 11. 11 April 9, 2016 – v3.0 Statistical Process Thinking - A Definition All work is a series of interconnected processes All processes vary Understanding, reducing and controlling process variation are keys to success ASQ
  • 12. 12 April 9, 2016 – v3.0 Customer Satisfaction or Customer Dissatisfaction Process/ System Material Machines Methods Men Environment The Variation Management Approach
  • 13. 13 April 9, 2016 – v3.0 The Six Sigma Approach Analytically speaking, this understanding may be expressed as The process output “y” is a function of the process inputs “x1, x2, ... , xN” where y is some product or process characteristic, also called the dependent variable, and (x1, x2 ,..., xN) describes all the independent variables in the cause system. Thus, we may interpret this expression to mean the output variable (y) is a function (f) of the input variables (x1, x2,..., xN) . y = f(x1, x2, ... , xN)
  • 14. 14 April 9, 2016 – v3.0 Six Sigma - What is a Defect ? A defect is any variation of a required characteristic of the product or its part, which is far enough removed from its nominal value to prevent the product from fulfilling the physical and functional requirements of the customer.
  • 15. 15 April 9, 2016 – v3.0 The key to process control and continuous process improvement is to understand the meaning and causes of variation in the outcome of the process. Variation Management – Continuous Improvement
  • 16. 16 April 9, 2016 – v3.0 Upper Specification Limit (USL) Defect nominal value Process Capability and Process Control Lower Specification Limit (LSL) Defect nominal value Upper Specification Limit (USL) Lower Specification Limit (LSL)
  • 17. 17 April 9, 2016 – v3.0 Upper Specification Limit (USL) Defect nominal value Process Capability and Process Control Lower Specification Limit (LSL) Defect nominal value Upper Specification Limit (USL) Lower Specification Limit (LSL) process not capable large variation & problem exist root cause analysis process improvement
  • 18. 18 April 9, 2016 – v3.0 Upper Specification Limit (USL) process not capable process out-of-control (trend) root cause analysis corrective action Defect nominal value Process Capability and Process Control Lower Specification Limit (LSL) Defect nominal value Upper Specification Limit (USL) Lower Specification Limit (LSL) process not capable large variation & problem exist root cause analysis process improvement
  • 19. 19 April 9, 2016 – v3.0 Product/Process Quality and Variation “Traditional” Attitude Nominal value LSL USL 100 % 0 %  Inspection/Yield and re-active Problem Solving Quality “Six Sigma” Attitude Nominal value LSL USL 100 % 0 %  Process Capability & Process Control with pro-active Improvements Quality
  • 20. 20 April 9, 2016 – v3.0 Supplier Selection based on Yield or Process Capability 125.08125.06125.04125.02125.00124.98124.96124.94124.92 Upper SpecLower Spec s Mean-3s Mean+3s Mean n k LSL USL Targ Cpm Ppk PPL PPU Pp Long-Term Capability 0 24969 0 32255 0.00 2.50 0.00 3.23 Obs PPM<LSL Exp Obs PPM>USL Exp Obs %<LSL Exp Obs %>USL Exp 0.026 124.923 125.080 125.001 782.000 0.029 124.950 125.050 * * 0.62 0.65 0.62 0.63 Process Capability Analysis for Supplier 1 125.04125.02125.00124.98124.96 Upper SpecLower Spec s Mean-3s Mean+3s Mean n k LSL USL Targ Cpm Ppk PPL PPU Pp Long-Term Capability 0 1 0 1 0.00 0.00 0.00 0.00 Obs PPM<LSL Exp Obs PPM>USL Exp Obs %<LSL Exp Obs %>USL Exp 0.010 124.969 125.031 125.000 1000.00 0.00 124.95 125.05 * * 1.62 1.63 1.62 1.62 Process Capability Analysis for Supplier 2 100 % Yield due to 100 % Inspection 100 % Yield due to a capable Process Which Supplier would you select ???
  • 21. 21 April 9, 2016 – v3.0 Remarks or Questions ?!?
  • 22. 22 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 23. 23 April 9, 2016 – v3.0 A histogram provides graphical presentation and a first estimation about the location, spread and shape of the distribution of the process. 0 10 20 30 40 50 The Histogram
  • 24. 24 April 9, 2016 – v3.0 Step 1: Collect at least 50 data points, but better 75 to 100 points, and organize your data into a table. Sort the data points from smallest to largest and calculate the range, means the difference between your largest and smallest data point, of your data points. The Histogram – How to create a Histogram? Actual Measurements Part Hole Size 1 2.6 2 2.3 3 3.1 4 2.7 5 2.1 6 2.5 7 2.4 8 2.5 9 2.8 10 2.6 Sorted Measurements Part Hole Size 5 2.1 2 2.3 7 2.4 6 2.5 8 2.5 1 2.6 10 2.6 4 2.7 9 2.8 3 3.1 Minimum = 2.1 Maximum = 3.1 Range = 1.0
  • 25. 25 April 9, 2016 – v3.0 Step 2: Determine the number of bars to be used to create the histogram of the data points. Calculate the width of one bar by dividing the range of your data by the number of bars selected. The Histogram – How to create a Histogram? Number of Bars: less than 50 50 - 100 100 - 250 over 250 5 or 7 5, 7, 9 or 11 7 - 15 11 - 19 Number of Data Points: Minimum = 2.1 Maximum = 3.1 Range = 1.0 Bar Width = 0.2 (5 Bars)
  • 26. 26 April 9, 2016 – v3.0 Step 3: Calculate the “start” and “end” point of each bar and count how many data points fall between “start” and “end” point of each bar. The Histogram – How to create a Histogram? Start End Bar 1 2.1 2.1 + 0.2 = 2.3 Bar 2 2.3 2.5 Bar 3 2.5 2.7 Bar 4 2.7 2.9 Bar 5 2.9 3.1 Minimum = 2.1 Maximum = 3.1 Range = 1.0 Bar Width = 0.2 (5 Bars) Sorted Measurements Part Hole Size Bar 5 2.1 1 2 2.3 2 7 2.4 2 6 2.5 3 8 2.5 3 1 2.6 3 10 2.6 3 4 2.7 4 9 2.8 4 3 3.1 5
  • 27. 27 April 9, 2016 – v3.0 Step 4: Draw the histogram indicating by the height of each bar the number of data points that fall between the “start” and “end” point of that bar. The Histogram – How to create a Histogram? Sorted Measurements Part Hole Size Bar 5 2.1 1 2 2.3 2 7 2.4 2 6 2.5 3 8 2.5 3 1 2.6 3 10 2.6 3 4 2.7 4 9 2.8 4 3 3.1 5 0 1 2 3 4 5 NumberofDataPoints 2.1 2.3 2.5 2.7 2.9 3.1
  • 28. 28 April 9, 2016 – v3.0 1. The bell-shaped distribution: Symmetrical shape with a peak in the middle of the range of the data. While deviation from a bell shape should be investigated, such deviation is not necessarily bad. The Histogram – Typical Patterns of Variation
  • 29. 29 April 9, 2016 – v3.0 2. The double-peaked distribution: A distinct valley in the middle of the range of the data with peaks on either side. This pattern is usually a combination of two bell-shaped distributions and suggests that two distinct processes are at work. The Histogram – Typical Patterns of Variation
  • 30. 30 April 9, 2016 – v3.0 3. The plateau distribution: A flat top with no distinct peak and slight tails on either sides. This pattern is likely to be the result of many different bell-shaped distribution with centers spread evenly throughout the range of data. The Histogram – Typical Patterns of Variation
  • 31. 31 April 9, 2016 – v3.0 4. The skewed distribution: An asymmetrical shape in which the peak is off-center in the range of the data and the distribution tails off sharply on one side and gently on the other. This pattern typically occurs when a practical limit, or a specification limit, exists on one side and is relatively close to the nominal value. The Histogram – Typical Patterns of Variation
  • 32. 32 April 9, 2016 – v3.0 5. The truncated distribution: An asymmetrical shape in which the peak is at or near the edge of the range of the data, and the distribution ends very abruptly on one side and tails off gently on the other. This pattern often occurs if the process includes a screening, 100 % inspection, or a review process. Note that these truncation efforts are an added cost and are, therefore, good candidates for removal. The Histogram – Typical Patterns of Variation
  • 33. 33 April 9, 2016 – v3.0 The Histogram – The Bell-Shaped or Normal Distribution
  • 34. 34 April 9, 2016 – v3.0 The Histogram – Exercise 1 Distribution of Heights of U.S. Population: Use the plot area below to construct a histogram from the random sample of heights on the right: 59 66 63 70 60 66 69 70 65 62 71 72 68 65 67 69 65 66 70 68 64 64 73 73 63 67 71 68 63 68 70 68 65 67 64 71 61 64 70 72 70 63 68 68 68 63 66 66 64 63 67 74 63 62 66 68 62 62 67 70
  • 35. 35 April 9, 2016 – v3.0 Remarks or Questions ?!?
  • 36. 36 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 37. 37 April 9, 2016 – v3.0 What are the Two Types of Product Characteristics ?  Attribute: A characteristic that by comparison to some standard is judged “Good” or “Bad” (free from scratches, fits, etc.)  Variable: A characteristic measured in physical units (inches, volts, amps, decibel, seconds, etc.)  Variable characteristics are specified by designers as a nominal value (target) and a tolerance around the target (variability).  Manufacturing processes attempt to produce at the nominal value of the characteristic.  Since no process is perfect, variation of the characteristics will occur.  Products with a product characteristic that falls inside the given tolerances will be defined as “good” or “acceptable”.  Products with a product characteristic that falls outside the given tolerances will be defined as “bad” or “unacceptable”.
  • 38. 38 April 9, 2016 – v3.0 Relationship between Tolerance to Nominal Value  The required value of a product or process characteristic is specified as its nominal value.  The maximum range of acceptable variation of the product or process characteristic which will still work in the product determines the tolerances about the nominal value. Nominal Value Specification Upper Spec. Limit (USL) Lower Spec. Limit (LSL) Tolerance
  • 39. 39 April 9, 2016 – v3.0 Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6 Measure of Location – The Sample Average Definition: N xxx x N  ...21 8.4 5 24 5 62475   x
  • 40. 40 April 9, 2016 – v3.0 Example 1: x1 = 2 x2 = 5 x3 = 4 Definition: Order all data points from the smallest to largest. Then choose the middle data point if the number of data points is odd, or the mean value of the two middle data points if the number of data points is even. Example 2: x1 = 5 x2 = 7 x3 = 4 x4 = 2 Example 3: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6 ? Measure of Location – The Sample Median 2 – 5 – 4 2 – 4 – 5 median = 4 5 – 7 – 4 – 2 2 – 4 – 5 – 7 median = 4.5
  • 41. 41 April 9, 2016 – v3.0 Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6 Measure of Variability – The Sample Range ),...,,min(),...,,max( 2121 NN xxxxxxR  Definition: 527)6,2,4,7,5min()6,2,4,7,5max( R
  • 42. 42 April 9, 2016 – v3.0 x3 x average _ x2 x1 x10 Measure of Variability – Sample Variance       9)110( ... 2 10 2 2 2 1 or xxxxxx   Time x6𝑥3 - 𝑥 𝑥2 - 𝑥
  • 43. 43 April 9, 2016 – v3.0 Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6 Measure of Variability – Sample Variance       )1( ... 22 2 2 12    N xxxxxx s N           7.3 )15( 8.468.428.448.478.45 22222 2    s Definition:
  • 44. 44 April 9, 2016 – v3.0 Example: x1 = 5 x2 = 7 x3 = 4 x4 = 2 x5 = 6 Measure of Variability – Sample Standard Deviation       )1( ... 22 2 2 1    N xxxxxx s N LT Definition:           7.3 )15( 8.468.428.448.478.45 22222 2    LTs 92.17.3 LTs
  • 45. 45 April 9, 2016 – v3.0 Time t Process Characteristic e.g. Hole Size Process not in control average Subgroup size n = 5 Number of subgroups N = 7 Measure of Variability – The Principle of Subgrouping
  • 46. 46 April 9, 2016 – v3.0 Where is the range of subgroup j, N the number of subgroups, and d2 depends on the size n of a subgroup (see handout). sST , often notated as s or sigma, is another measure of dispersion or variability and stands for “short-term standard deviation”, which measures the variability of a process or system using “rational” subgrouping. Measure of Variability – Standard Deviation sST 22 21 ... dRd N RRR s N ST    minmax XXRj  n 2 3 4 5 6 7 8 9 10 d2 1.128 1.693 2.059 2.326 2.534 2.704 2.847 2.970 3.078
  • 47. 47 April 9, 2016 – v3.0 Time t Process Characteristic e.g. Hole Size Process not in control average Subgroup size n = 5 Number of subgroups N = 7 Measure of Variability – The Principle of Subgrouping  sST stays the same, even if the process is not in control  sLT increases over time because the process is not in control  sST and sLT are identical if the process was in control
  • 48. 48 April 9, 2016 – v3.0 Long-term standard deviation: Short-term standard deviation: The difference between the standard deviations sLT and sST gives an indication of how much better one can do when using appropriate production control, like Statistical Process Control (SPC).       )1( ... 22 2 2 1    N xxxxxx s N LT Measure of Variability – Difference between sLT and sST 22 21 ... dRd N RRR s N ST   
  • 49. 49 April 9, 2016 – v3.0 averageaverage -1*s(igma) average -2*s(igma) average -3*s(igma) average +1*s(igma) average +2*s(igma) average +3*s(igma) 34.13 %34.13 % 13.60 % 13.60 % 2.14 %2.14 % 0.13 % 0.13 % Measure of Variability – The Normal Distribution If your process is under control, over 99.74% of your data points will fall between the average ± 3s(igma) limits. The distance between average ± 3s(igma) limits is called the width of the process or process capability. Lower Control Limit Upper Control Limit
  • 50. 50 April 9, 2016 – v3.0 Measure of Location and Variability – Exercise 2 Calculate the Mean Value or Average, Median, Range, and long- and short-term Standard Deviation of the sample data. You may copy the data into MS Excel and simplify the calculations. Group 1 59 66 63 62 2 60 66 69 65 3 65 62 71 72 4 68 65 67 69 5 65 66 70 68 6 64 64 73 73 7 63 67 71 68 8 63 68 65 68 MeasurementsOverall Mean Value = Overall Median = Subgroup Ranges = Long-term Standard Deviation = Short-term Standard Deviation = Note: The Excel function for the Long-Term Standard Deviation is “stdev()”.
  • 51. 51 April 9, 2016 – v3.0 Measure of Location and Variability – Exercise 2 Results Subgroup Median Range 1 59 66 63 62 63 7 2 60 66 69 65 66 9 3 65 62 71 72 68 10 4 68 65 67 69 68 4 5 65 66 70 68 67 5 6 64 64 73 73 69 9 7 63 67 71 68 68 8 8 63 68 65 68 67 5 Overall Range: 14 Overall Median: 66 Average Range: 7.1 Short-Term Standard Deviation: 3.46 Long-Term Standard Deviation: 3.55 Measurements
  • 52. 52 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 53. 53 April 9, 2016 – v3.0 Attribute Data (Count or Yes/No Data) Variable Data (Measurements) Variable subgroup size Subgroup size of 1 Fixed subgroup size x chart x-bar R chart x-bar s chart Count Incidences or nonconformities Fixed oppor- tunity Variable oppor- tunity c - chart u - chart Yes/No Data Defectives or nonconforming units Fixed subgroup size Variable subgroup size np - chart p - chart Process Control Charts – Types of Control Charts Type of Data
  • 54. 54 April 9, 2016 – v3.0  The x - chart is a method of looking at variation in a variable data or measurement.  One source is the variation in the individual data points over time. This represents “long term” variation in the process.  The second source of variation is the variation between successive data points. This represents “short term” variation.  Individual or x - charts should be used when there is only one data point to represent a situation at a given time.  To use the x - chart, the individual sample results should be sufficient normally distributed. If not, the x - chart will give more false signals. Process Control Charts – The x - Chart
  • 55. 55 April 9, 2016 – v3.0 Upper control limit = Lower control limit = Upper control limit = Lower control limit = The x- chart The R- chart , where x1, x2, ..., xN are the measurements, N the number of measurements, , and . Process Control Limit – The x - Chart RxRxdRx  66.2128.133 2 RxRxdRx  66.2128.133 2 RRD  267.34 003  RRD N xxx x N  ...21 1 ...32    N RRR R N 1 iii xxR
  • 56. 56 April 9, 2016 – v3.0 Process Control Charts – x - Chart Example 1 0 .0 C HA No k F ran to nt h OBSERVATIONS7 9 1 1 1 3 1 5 1 7 1 9 2 1 1 2 EEEEEEEEEEEEEEEEEEEEEEEE A L U RANGES0 1 2 3 4 5 6 L U R 01.01.8701.02.8701.03.8701.04.8701.05.8701.06.8701.07.8701.08.8701.09.8701.10.8701.11.8701.12.8701.01.8801.02.8801.03.8801.04.8801.05.8801.06.8801.07.8801.08.8801.09.8801.10.8801.11.8801.12.88 1 00 G ro A uto C L O C urv K -S A V E P R O UC L L C L re e n 128.12 RdRsST 
  • 57. 57 April 9, 2016 – v3.0 Process Control Charts – The Central Limit Theorem Regardless of the shape of the distribution of a population, the distribution of average values, x-bar’s, of subgroups of size n drawn from that population will tend toward a normal distribution as the subgroup size n becomes large. Laplace and Gauss The standard deviation sx of the subgroup averages is smaller than the standard deviation s of the individual measurements. The relationship between these two standard deviation s and sx as follows, where n is the nuymber of measurements in each subgroup: _ _ nssx 
  • 58. 58 April 9, 2016 – v3.0 Process Control Charts – Exercise 3 Throw the Dice: Step 1: Throw the dice 30 times and record the results in the table on the right. Step 2: Draw a Histogram #1 of the 30 data points in one of the spreadsheets below. Step 3: Calculate the average to 2 consecutive throws and draw the histogram #2 of the resulting 15 data points. What do you see and why? AverageResults Histogram #1 Histogram #2
  • 59. 59 April 9, 2016 – v3.0  The (x-bar / R) - chart should be used if  the individual measurements are not normally distributed,  one can rationally subgroup the data and is interested in detecting differences between the subgroups over time.  The (x-bar / R) - chart is a method of looking at two different sources of variation. One source is the variation in subgroup averages. The other source is the variation within a subgroup.  The x-bar - chart shows variation over time or long-term variation and the R - chart is a measure of the short-term variation in the process. Process Control Charts – The (x-bar/R) - Chart
  • 60. 60 April 9, 2016 – v3.0 Upper control limit = Lower control limit = The R- chart Upper control limit = Lower control limit = The x-bar - chart where x-bar1, x-bar2, ..., x-barN are the averages of each subgroup, n the number of items in a subgroup, N the number of subgroups, ., and Process Control Limit – The x-bar/R - Chart   RAxndRx  223 RD 4 RD 3 N xxx x N  ...21 N RRR R N  ...21minmax iii xxR    RAxndRx  223
  • 61. 61 April 9, 2016 – v3.0 n 2 3 4 5 6 A2 1.880 1.023 0.729 0.577 0.483 D3 0 0 0 0 0 D4 3.267 2.574 2.282 2.114 2.004 d2 1.128 1.693 2.059 2.326 2.534 Factors for x- and x-bar/R - Charts
  • 62. 62 April 9, 2016 – v3.0 1 0 .0 C HA No k F ran 5 - o f AVERAGES.0 .0 .0 .0 .0 .0 2 1 1 2 AA* A L U RANGES0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 L U R 95.01.2095.01.2195.01.2395.01.2495.01.2595.01.2695.01.2795.01.2895.01.3095.01.3195.02.0195.02.0295.02.0395.02.0495.02.0695.02.0795.02.0895.02.0995.02.1095.02.1195.02.1395.02.2095.02.2195.02.22 INDIVIDUALS 6 .00 6 .0 G ro A uto C L O C urv K -S A V E P R O UC L L C L e re e n Process Control Charts – (x-bar/R) - Chart Example
  • 63. 63 April 9, 2016 – v3.0  The (x-bar / s) - chart should be used instead the (x-bar / R) - chart if the subgroup is larger than 10. In this case, the standard deviation is a better measurement than the range for the variation between individual measurements in a subgroup.  The (x-bar / s) - chart can be used whenever one can use the (x-bar / R) - chart.  The (x-bar / s) - chart is a method of looking at sources of variation. One chart looks at variation in the subgroup averages x-bar. The other chart examines variation in the subgroups standard deviation s. Process Control Charts – The (x-bar/s) - Chart
  • 64. 64 April 9, 2016 – v3.0 Upper control limit = Lower control limit = Upper control limit = Lower control limit = The s- chart The x-bar - chart , and where x-bar1, x-bar2, ..., x-barN are the averages of each subgroup, s1, s2, ..., sN are the standard deviations of each subgroup, n the number of items in a subgroup, N the number of subgroups, . Process Control Limit – The x-bar/s - Chart sAx  3 sAx  3 sB 4 sB 3 N xxx x N  ...21 N sss s N  ...21
  • 65. 65 April 9, 2016 – v3.0 n 6 7 8 9 10 A3 1.287 1.182 1.099 1.032 0.975 B3 0.030 0.118 0.185 0.239 0.284 B4 1.970 1.882 1.815 1.761 1.716 c4 0.9515 0.9594 0.9650 0.9693 0.9727 Factors for x-bar/s - Charts
  • 66. 66 April 9, 2016 – v3.0 Common Causes: Causes that are implemented in the process due to the design of the process, and affect all outcomes of the process. Identifying these types of causes requires Design of Experiment (DOE) methods. Special Causes: Causes that are not present in the process all the time and do not affect all outcomes, but arise because of specific circumstances. Special causes can be identified using SPC. Walter A. Shewhart (1931) Out-of-Control Criteria – Two Causes of Variation
  • 67. 67 April 9, 2016 – v3.0 Unstable Process: A process in which variation is a result of both common and special causes. Stable Process: A process in which variation in outcomes arises only from common causes. Out-of-Control Criteria – Two Types of Processes
  • 68. 68 April 9, 2016 – v3.0 An out-of-control criteria is a signal of a special causes of variation: • Is a systematic pattern of the product or process characteristic monitored and charted • Has a low probability of occurring when the process is stable and in control SPC Out-of-Control Criteria – The Types of Signals
  • 69. 69 April 9, 2016 – v3.0 What is the “chance” to loose the coin flip 11 times in a row? 1 = 2 = … … … 11 = What is the “chance” to loose the coin flip 11 times in a row? 1 = 50% or 0.50 2 = … … … 11 = What is the “chance” to loose the coin flip 11 times in a row? 1 = 50% or 0.50 2 = 25% or 0.50*0.50 … … … 11 = What is the “chance” to loose the coin flip 11 times in a row? 1 = 50% or 0.50 2 = 25% or 0.50*0.50 … … … 11 = 0.049% or 0.5011
  • 70. 70 April 9, 2016 – v3.0 Out-of-Control Criteria – The Basic Idea average average +1*s(igma) average -1*s(igma) average +2*s(igma) average -2*s(igma) average -3*s(igma) average +3*s(igma) 34.13 %34.13 % 13.60 % 13.60 % 2.14 %2.14 % 0.13 % 0.13 % If your process is under control, over 99.74% of your data points will fall between the average ± 3s(sigma) limits and there is only a 0.13% that a measurement point would fall outside these limits. Lower Control Limit Upper Control Limit
  • 71. 71 April 9, 2016 – v3.0 Process Out-of-Control Criteria Below is a list of the most commonly used out-of-control criteria included in Minitab 17 and as defined by Walter Shewhart in the 1920s. Criteria 1: Outlier Criteria 2 & 5 & 6: Process Shift Criteria 3: Process Trend
  • 72. 72 April 9, 2016 – v3.0 SPC Criteria #1 – 1 Point above or below 3 Sigma All SPC Out-of-Control Criteria have about a 1 in 1,000 chance to occur in a process without a special cause. Therefore, they are strong evidence for the presence of a special cause.
  • 73. 73 April 9, 2016 – v3.0 SPC Criteria #2 – 9 Points on the same Side of the Average 9 consecutive points above or below the process performance average line often indicates a shift in process performance.
  • 74. 74 April 9, 2016 – v3.0 SPC Criteria #3 – 6 Consecutive Points Increasing or Decreasing 6 consecutive points increasing or decreasing often indicates a trend in process performance due to a special cause.
  • 75. 75 April 9, 2016 – v3.0 SPC Criteria #5 – 2 of 3 Points above or below 2 Sigma 2 of 3 consecutive points above or below 2 Sigma line often indicates a shift in process performance.
  • 76. 76 April 9, 2016 – v3.0 SPC Criteria #6 – 4 of 5 Points above or below 1 Sigma 4 of 5 consecutive points above or below the 2 Sigma line often indicates a shift in process performance.
  • 77. 77 April 9, 2016 – v3.0 Special Causes showing in the MR, R, or s Chart 1 data point above or below the 3 Sigma line is often the only criteria used to identify special causes in process performance.
  • 78. 78 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 79. 79 April 9, 2016 – v3.0 Relationship between Tolerance and Process Capability For any critical characteristic the Capability Index cp, also known as the design margin, is quantified by : ST p s LSLUSL LCL-UCL USL - LSL c     6 - Variation)Process(TotalCapabilityorWidthProcess Tolerance)(Designion WidthSpecificat 3 )-( ST p s TUSL c  or 3 )-( ST p s LSLT c  and with T = Target
  • 80. 80 April 9, 2016 – v3.0 The Capability Index cp 1.00 < cp < 1.33 Average = Nominal Value or Target LSL USLLCL UCL average + 3*s (or sigma) average - 3*s (or sigma) x
  • 81. 81 April 9, 2016 – v3.0 Process Performance Study In some situations (e.g. order of production unknown, only small sample produced) the short-term standard deviation sST cannot be calculated and therefore the long-term standard deviation sLT has to used. This means the actual process performance is evaluated, rather than the potential process capability. To address this difference the Performance Index pp is used instead of the Capability Index cp. LT p s LSLUSL p   6 - or 3 )-( LT p s TUSL p   3 )-( LT p s LSLT p  and with T = Target
  • 82. 82 April 9, 2016 – v3.0 Actual Process Performance Cpk is the Capability Index adjusted for k, which considers actual operation of the process and takes into account any difference between the design nominal and the actual process mean value x- bar. ) 3 - , 3 - (min STST pk s LSLx s xUSL c   where is the average of all measurements or data points.x
  • 83. 83 April 9, 2016 – v3.0 The Capability Index cpk nominal value LSL USLLCL UCL average 1.33 < cp < 2.00, but 1.00 < cpk < 1.33 average - 3*s(igma) average + 3*s(igma) x
  • 84. 84 April 9, 2016 – v3.0 Process Performance Study In some situations (e.g. order of production unknown, only small sample produced) the short-term standard deviation sST can not be calculated and therefore the long-term standard deviation sLT has to used. This means the actual process performance is evaluated, rather than the potential process capability. To address this difference the Performance Index ppk is used instead of the Capability Index cpk. ) 3 - , 3 - (min LTLT pk s LSLx s xUSL p  
  • 85. 85 April 9, 2016 – v3.0 ppk ccTx  The Capability Index cpk  Note, the cpk as defined above can also be used in the case of unilateral specification, means that there exist only an upper or lower specification limit and no target value. Continue:  Note, if the process average (x-bar) is equal to the midpoint (nominal value) of the specification interval, means if
  • 86. 86 April 9, 2016 – v3.0 Quantifying Actual Process Performance In some cases, the target value (T) of the process is not the midpoint of the specification interval. For example the lower specification limit may be the best value for the quality characteristic. The Taguchi Capability Index cpm is a measure for the difference between the average of the process and its target value.  22 6 Txs LSLUSL c ST pm    Note, if ppm ccTx 
  • 87. 87 April 9, 2016 – v3.0 The Capability Index cpm Nominal Value or Target LSL USLLCL UCL average 1.50 < cp < 2.00, 1.33 < cpk < 1.5, but 0.50 < cpm < 1.00 average - 3*s(igma) average + 3*s(igma) x
  • 88. 88 April 9, 2016 – v3.0 Process Performance Study In some situations (e.g. order of production unknown, only small sample produced) the short-term standard deviation sST can not be calculated and therefore the long-term standard deviation sLT has to used. This means the actual process performance is evaluated, rather than the potential process capability. To address this difference the Performance Index ppm is used instead of the Capability Index cpm.  22 6 Txs LSLUSL p LT pm    Note, if ppm ppTx 
  • 89. 89 April 9, 2016 – v3.0 Why are cp and cpk useful and significant Measures?  To maximize cp requires the joint and concurrent effort of both product and process designers.  Product Design has the goal of increasing the allowable tolerance to the maximum which will still permit successful function of the product.  Process Design has the goal of minimizing the variability of the process which reproduces the characteristic required for successful function of the product, and for centering the process on target (nominal) value of the characteristic.  A high cp Index indicates that the process is capable of reproducing the characteristic. (It makes no statement about the centering of the process.)  A high cpk Index indicates that the process is actually reproducing the characteristic within the desired limits. (It makes no statement about the inherent capability, other than its minimum value.)  cpk < 1.00 means “Process not capable” 1.00 < cpk < 1.33 means “Process is marginal” cpk > 1.33 means “Process is capable”
  • 90. 90 April 9, 2016 – v3.0 Confidence Limits for the Capability Indices The confidence intervals for the capability indices can be defined as follows. where m= n·N is number of measurements and (1-100% the required confidence (e.g. for a confidence of 95 % is z/2 = 1.96).    12 ˆ ˆ 12 ˆ ˆ 22     m c zcc m c zc p pp p p      2 2 2 2 2 22 2 2 2 2 2 1 2 1 ˆ ˆ 1 2 1 ˆ ˆ                               s Tx s Tx m c zcc s Tx s Tx m c zc pm pmpm pm pm      mm c zcc mm c zc pk pkpk pk pk         9 1 12 ˆ ˆ 9 1 12 ˆ ˆ 2 2 2 2 
  • 91. 91 April 9, 2016 – v3.0 Confidence Limits for the Capability Indices
  • 92. 92 April 9, 2016 – v3.0 Confidence Limits for the Capability Indices
  • 93. 93 April 9, 2016 – v3.0 Confidence Limits for the Capability Index cp 95% confidence interval 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 1.2000 1.4000 1.6000 0 100 200 300 400 500 600 m=N*n delta cp=0.5 cp=1.0 cp=1.5 cp=2.0 N = # of Subgroups n = Subgroup Size
  • 94. 94 April 9, 2016 – v3.0 Confidence Limits for the Capability Index cpk 95% confidence interval 0.0000 0.2000 0.4000 0.6000 0.8000 1.0000 1.2000 1.4000 1.6000 0 100 200 300 400 500 600 m=N*n delta cpk=0.5 cpk=1.0 cpk=1.5 cpk=2.0 N = # of Subgroups n = Subgroup Size
  • 95. 95 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 96. 96 April 9, 2016 – v3.0 Treatment of Non-Normality - Skewness Transformation of the original measurements using “ The Power Ladder ”.  originaltransform xx  This method works best when the ratio of the largest to smallest measurement value is greater than zero. If this is not the case, a simple transformation like with a appropriate constant “c” should work. cxx originaltransform 
  • 97. 97 April 9, 2016 – v3.0 Treatment of Non-Normality - Skewness 6543210-1 Upper SpecLower Spec s Mean-3s Mean+3s Mean n k LSL USL Targ Cpm Cpk CPL CPU Cp Short-Term Capability 0 94876 30000 7429 0.00 9.49 3.00 0.74 Obs PPM<LSL Exp Obs PPM>U SL Exp Obs %<LSL Exp Obs %>USL Exp 0.80061 -1.35198 3.45169 1.04985 100.000 0.300 0.000 3.000 * * 0.44 0.44 0.81 0.62 Process Capability Analysis for Original 20100 2.4 1.6 0.8 0.0 Xbar and R Chart Subgr Means X=1.050 3.0SL=2.124 -3.0SL=-0.02428 6 4 2 0 Ranges 1 R=1.862 3.0SL=3.938 -3.0SL=0.000 20100 Last 20 Subgroups 6 4 2 0 SubgroupNumber Values 30 3.45169-1.35198 Cp: 0.62 CPU: 0.81 CPL: 0.44 Cpk: 0.44 Capability Plot Process Tolerance Specifications StDev: 0.800612 III III 5.02.50.0 Normal Prob Plot 5.02.50.0 Capability Histogram Process Capability Sixpack for Original “The Original Measurements“ The Capability Study indicates almost 100 000 PPM below the lower specification limit.
  • 98. 98 April 9, 2016 – v3.0 Treatment of Non-Normality - Skewness 3.02.52.01.51.00.50.0-0.5-1.0 8 7 6 5 4 3 2 1 0 95% Confidence Interval StDev Lambda Last IterationInfo 0.693 0.692 0.695 0.393 0.337 0.281 StDevLambda Up Est Low Box-CoxPlot for Original  33.0 33.0 33.0 originaltransform originaltransform originaltransform USLUSL LSLLSL xx    “The Power Ladder“ Which function would transform the original data to a normal distribution ?
  • 99. 99 April 9, 2016 – v3.0 Treatment of Non-Normality - Skewness 2.001.751.501.251.000.750.500.250.00 UpperSpecLowerSpec s Mean-3s Mean+3s Mean n k LSL USL Targ Cpm Cpk CPL CPU Cp Short-TermCapability 0 485 30000 41141 0.00 0.05 3.00 4.11 Obs PPM<LSL Exp Obs PPM>USL Exp Obs %<LSL Exp Obs %>USL Exp 0.28635 0.08559 1.80369 0.94464 100.000 0.310 0.000 1.442 * * 0.58 1.10 0.58 0.84 Process CapabilityAnalysis for Transfor 20100 1.25 1.00 0.75 0.50 Xbar and R Chart Subgr Means X=0.9446 3.0SL=1.329 -3.0SL=0.5605 1.5 1.0 0.5 0.0 Ranges R=0.6660 3.0SL=1.408 -3.0SL=0.000 20100 Last 20 Subgroups 1.7 1.2 0.7 0.2 SubgroupNumber Values 1.44220.0000 1.803690.08559 Cp: 0.84 CPU: 0.58 CPL: 1.10 Cpk: 0.58 Capability Plot Process Tolerance Specifications StDev: 0.286349 III III 1.51.00.5 Normal Prob Plot 1.51.00.5 Capability Histogram Process Capability Sixpack for Transfor “The Transformed Measurements“ The Capability Study for the transformed data indicates “only” 500 PPM below the lower specification limit !
  • 100. 100 April 9, 2016 – v3.0 Section 1: Introduction Section 2: The Histogram Section 3: Basic Statistics and Process Capability Section 4: Introduction to Statistical Process Control Section 5: Definitions of Process Capability Indices Section 6: Non-Normal Distributed Processes Process Capability Study – Table of Contents
  • 101. 101 April 9, 2016 – v3.0 The End … “Perfection is not attainable, but if we chase perfection we can catch excellence.” - Vince Lombardi
  • 102. 102 April 9, 2016 – v3.0 Terms & Conditions After you have downloaded the training material to your own computer, you can change any part of the course material and remove all logos and references to Operational Excellence Consulting. You can share the material with your colleagues and re-use it as you need. The main restriction is that you cannot distribute, sell, rent or license the material as though it is your own. These training course materials are for your — and your organization's — usage only. Thank you.