International Journal for Quality Research 8(3) 399-416

ISSN 1800-6450

Yerriswamy Wooluru 1
Swamy D.R.
P. Nagesh

Article info:
Received 22.03.2014
Accepted 20.06.2014
UDC 54.061

THE PROCESS CAPABILITY ANALYSIS - A

TOOL FOR PROCESS PERFORMANCE
MEASURES AND METRICS - A CASE STUDY
Abstract: Process Capability can be evaluated through the
computations of various process capability ratios and indices.
The basic three capability indices commonly used in
manufacturing industries are Cp, Cpk,Cpm and Cpmk .Process
capability indices are intended to provide single number
assessment of the ability of a process to meet specification
limits on quality characteristics of interest. Thus, it identifies
the opportunities for improving quality and productivity. The
level of significance on process capability analysis has been
increased considerably over last decade, but the literature
findings reveal the importance of understanding the concepts,
methodologies and critical assumptions while its
implementation in manufacturing process. The objective of this
paper is to conduct process capability analysis for boring
operation by understanding the concepts, methodologies and
making critical assumptions.
Keywords: Process Capability Index, Normal Distribution,
Statistical Process Control, Run chart

1. Introduction1
Process capability study is a method of
combining the statistical tools developed
from the normal curve and control charts
with good engineering judgment to interpret
and analyze the data representing a process.
The purpose of the process capability study
is to determine the variation spread and to
find the effect of time on both the average
and the spread. The administration, analysis
and use of the process capability study
should be an integral part of the quality
engineering function. The results could be
used for new design applications, inspection
planning and evaluation techniques. It is a
1

Corresponding author: Yerriswamy Wooluru

email: [email protected]

type of tool that can be used to prevent

defects during the production cycle through
better designs, through factual knowledge of
machine or process limitations and through
knowledge of process factors that can or
cannot be controlled. In any manufacturing
operation there is a variability which is
manifested in the product made by the
operations .Quantifying the variability with
objectives and advantages of reducing it in
the manufacturing process is the prime
activity of the process management.
Process Capability refers to the evaluation of
how well a process meets specifications or
the ability of the process to produce parts
that conform to engineering specifications,
Process Control refers to the evaluation of
process stability over time or the ability of
the process to maintain a state of good
statistical control. These are two separate but

399

vitally important issues that we must address

when considering the performance of a
process and so the assessment of process
capability is inappropriate and statistically
invalid to assess with respect to conformance
to specifications without being reasonably
assured of having good statistical control.
Before evaluating the process capability, the
process must be shown under statistical
control i.e. the process must be operating
under the influence of only chance causes of
variation and also ensure that the process
data is normally distributed and observations
are independent.
A process may produce a large number of
pieces that do not meet the specifications,
even though the process itself is in a state of
statistical control (i.e., all the points on the
X-bar and R charts are within the 3 sigma
limits and vary in random manner).This may
be due to the lack of centering of the process
mean in other words, the actual mean value
of the parts being produced may be
significantly different from the specified
nominal value of the part. If this is the case,
an adjustment of the machine to move the
mean closer to the nominal value may solve
the problem. Another possible reason for
lack of conformance to specifications is that
a statistically stable process may be
producing parts with an unacceptably high
level of common-cause variation, even
though the process is centered at the nominal
value.

2. The basic capability indices

commonly used in
manufacturing industries are
Cp, Cpk Cpm and Cpmk.
Cp: It simply relates the Process Capability
to the Specification Range and it does not
relate the location of the process with respect
to the specifications. Values of Cp exceeding
1.33 indicate that the process is adequate to
meet the specifications. Values of Cp
between 1.33 and 1.00 indicate that the
process is adequate to meet specifications

400

but require close control. Values of Cp

below 1.00 indicate the process is not
capable of meeting specifications. If the
process is centered within the specifications
and is approximately normal then Cp =
1.00 results in a fraction nonconforming of
0.27%. It is also known as process potential.
Cpl: It estimates process capability for
specifications that consist of a lower limit
only (for example, strength) and it assumes
process output is approximately normally
distributed. Cpu: It estimates process
capability for specifications that consist of
an upper limit only (for example,
concentration). Assumes process output is
approximately normally distributed.
Cpk: It considers process average and
evaluates the process spread with respect to
where the process is actually located. The
magnitude of Cpk relative to Cp is a direct
measurement of how off-centre the process
is operating. It assumes process output is
approximately normally distributed. If the
characteristic or process variation is centered
between its specification limits, the
calculated value for CPK is equal to the
calculated value for CP. But as soon as the
process variation moves off the specification
center, it is penalized in proportion to how
far its offset. CPK is very useful and very
widely used. Generally, a CPK greater than
1.33 indicates that a process is capable in the
short term. Values less than 1.33 tells that
the variation is either too wide compared to
the specification or that the location of the
variation is offset from the center of the
specification. It may be a combination of
both width and location. Cpk measures how
far the process mean is from the nearer
specification limit in terms of 3 distances.
Cpk works well only for the bell-shaped
normal (Gaussian) distribution. For others
it is an approximation. Cpk = Cp only when
the process is perfectly centered. Cp
represents the highest possible value for
Cpk.
Cpm: It estimates process capability around
a target T, it is always greater than zero and

Y. Wooluru, D.R. Swamy, P. Nagesh

assumes process output is approximately

normally distributed. It is also known as the
Taguchi capability index, introduced in
1988. Cpk measures how well the process
mean is centered within the specification
limits, and what percentage of product will
be within specification limits. Instead of
focusing on specification limits, Cpm
focuses on how well the process mean
corresponds to the process target, which may
or may not be midway between the
specification limits. Cpm is motivated by
Taguchis
Loss
Function.
The
denominator of Cpm includes the Root Mean
Square deviation from the target. Cpk is
preferred to Cp because it measures both
process location and process standard
deviation. Cpm is often preferred to Cpk
because the variability term used in the index

is more consistent with Run to Target

Philosophy.
Cpmk: It estimates process capability
around a target (T), and accounts for an offcenter process mean and assumes process
output is approximately normally distributed.
The process capability index - Cpk considers
process average and evaluates half the
process spread with respect to where the
process average is actually located, though
Cpk takes the process mean into consideration
but it fails to differentiate an on-target
process from off-target process. The way to
address this difficulty is to use a process
capability index Cpm that is better indicator
of centering.
Summary of Process Capability Indices and
their usage is presented in the Table 1.

Table 1. List of process capability indices equations and their usage

Index
Estimation Equation
Usage
It estimates what the process is capable
of producing if the process mean were to

Cp
be centered between the specification
Cp =
limits. Assumes process output is
approximately normally distributed.
It estimates process capability for
specifications that consist of a lower

Cpl
Cpl =
It
limit only. Assumes process output is
approximately normally distributed.

Cpu

Cpu =

Cpk

Cpm

It estimates process capability for

specifications that consist of an upper
limit only. Assumes process output is
approximately normally distributed.
It estimates what the process is capable
of producing, considering that the
process mean may not be centered
between the specifications limits.
It estimates process capability around a
target T is always greater than zero. It
assumes process output is approximately
normally distributed. Cpm is also known
as the Taguchi capability index.

401

Cpmk

It estimates process capability around a

target T and accounts for an off-center
process mean. Assumes process output is
approximately normally distributed.

3. Methodology
Estimation of Process Capability for boring
operation involves the following steps:

Understanding the basic concepts of

process capability analysis and its
measures.
Process data collection.
Calculate required statistics
Validate the critical assumptions.
Estimation of Cp, Cpu, Cpl, Cpk,
Cpm and Cpmk.
Analysis of process capability
results.

If the process is not capable of

meeting the specification, find the
predominant factor that affecting
the process capability.
Take action to improve the process
performance.
Estimate the confidence intervals
and Carryout Hypothesis Testing.

4. Data collection
Critical quality characteristic of the gear i.e.
Bore diameter on the driver gear processed
by boring operation in an automotive
industry has been identified.The product
description is given in the Table 2 and the
measured values are presented in the Table
3.

Table 2. Product description

Material: Cast steel

Part Name : Driver Gear

Operation: Boring

Specifications: 205.00 0.05

Instrument used : Dial Bore Gauge

All dimensions are in mm

Table 3.The measured values of Bore diameter

Sample
1
2
3
No.
1.
205.030
205.020
205.010
2.
205.010
205.020
205.025
3.
205.010
205.030
205.050
4.
205.030
205.020
205.030
5.
205.040
205.035
205.030
6.
205.030
205.030
205.025
7.
205.025
205.025
205.025
8.
205.015
205.020
205.025
9.
205.025
205.030
205.040
10.
205.025
205.025
205.020
11.
205.010
205.005
205.030
12.
205.030
205.020
205.030
13.
205.030
205.040
205.030
14.
205.030
205.040
205.030

402

205.045
205.030
205.030
205.040
205.030
205.030
205.025
205.010
205.010
205.010
205.040
205.030
205.040
205.030

205.010
205.010
205.020
205.035
205.035
205.035
205.025
205.020
205.010
205.020
205.040
205.030
205.030
205.025

205.023
205.019
205.028
205.031
205.034
205.030
205.025
205.018
205.023
205.020
205.025
205.028
205.034
205.031

0.035
0.020
0.04
0.02
0.01
0.01
0.00
0.015
0.03
0.015
0.035
0.01
0.01
0.015

Y. Wooluru, D.R. Swamy, P. Nagesh

15.
16.
17.
18.
19.
20.

205.010
205.035
205.045
205.040
205.030
205.040

205.010
205.040
205.038
205.030
205.025
205.035

205.020
205.037
205.045
205.025
205.030
205.030

5. Process Capability Analysis

According to Kotz and Montgomery (2000)
the following critical assumptions have been
made and validated before estimating the
process capability for boring operation.

The process must be in state of

statistical control.
The quality characteristic has a
normal distribution.
In the case of two sided
specifications, the process mean is
centered between the lower and
upper specification limits.
Observations must be random and
independent of each other.

205.040
205.042
205.033
205.025
205.030
205.030

205.050
205.040
205.030
205.020
205.035
205.020

205.026
205.038
205.038
205.028
205.030
205.031

0.04
0.007
0.015
0.02
0.01
0.02

All the above assumptions have been

verified as follows.
5.1 Construction of and R- Chart to
assess the statistical stability of the boring
operation
Control limits for - Chart

205.049 + [(0.577)
(0.029)] =205.06560,

= 205.049 - [(0.577)
(0.029)] =205.03219
Control limits for R-Chart
= 2.114 x0.029 =0.06124
UCL=

LCL=
= 0.00 x 0.029 =0.0000
From Table, for n=5
.
. =0.00.
=2.114

Figure 1. Control charts for case study data

403

It has been observed from the Figure 1 that

all plotted sample range and mean values are
within the control limits on both R-Chart as
well as X-Bar chart and no indication of
Trend, shift, run and clustering has been
noticed. Hence, it is concluded that the
process is under statistical control and
operating under the influence of only chance
causes of variation. i.e., the process is stable
over time.

5.2 Normal probability plot and

histogram for validating the Normality
assumption
Graphical methods including the histogram
and normal probability plot are used to check
the normality of the case study data. Figure 2
displays the histogram and Figure 3 display
the normal probability plot for the case study
data. The sample data appears to be normal.

Figure 2. Histogram for the case study data

Figure 3. Normal Probability Plot for the case study data

404

Y. Wooluru, D.R. Swamy, P. Nagesh

Test results for normal probability plot for

the data from MINITAB -14 statistical
software output shows Mean: 205.00,
Standard deviation: 0.0126, Anderson
Darling test statistic value: 0.215 and Pvalue: 0.844 is greater than the significance
level ( = 0.05) implies that the data is

distributed normally .Thus, it is concluded

that the sample data can be regarded as taken
from a normal process.
5.3 Construction of Run chart for
checking the assumption of Randomness
using MINITAB software

Run Chart of Bore diameter Dimension

205.09
205.08

Bore diameter

205.07
205.06
205.05
205.04
205.03
205.02
205.01
1

10

Number of runs about median:

Expected number of runs:
Longest run about median:
A pprox P-Value for C lustering:
A pprox P-Value for Mixtures:

20

30

51
51.00000
5
0.50000
0.50000

40
50
60
Observation

70

Number of runs up or down:

Expected number of runs:
Longest run up or down:
A pprox P-Value for Trends:
A pprox P-Value for O scillation:

80

90

100

56
66.33333
4
0.00669
0.99331

Figure 4. Run chart for case study data

Interpreting the Run Chart: p - values for
clustering, mixtures and oscillation are
greater than value of 0.05. In case of the
trend, P-value is smaller than the alpha value
and it warns that the process is about to go
out of control due to the factor like worn
tool. The actual numbers of runs are close to
the expected number of runs. Hence it is
concluded that the observations in the data
set are random. After validating the three
critical assumptions, the process capability
for the boring operation would be quantified.

6. Estimation of Process
Capability Indices for existing
process conditions
6.1Process Capability Index - Cp

X = ) and Average Range ( R ) value are

computed. The process standard deviation is
obtained with the help of formula, = R /
d2, value of d2 for a sample size of five is
noted as 2.326 from statistical tables for
control chart constants.
The process standard deviation () = R /
d2. ' = 0.029 / 2.326 = 0.01246.
Process capability index,
Cp =

= 1.34

Percentage of specification band used by the

current process.

The Average of Averages of the samples (

405

Cr = (1 / Cp) x 100 = (1/ 1.34) x 100 = 74.62

%.This means that the manufacturing
process uses 74.62 % of specification band.

6.4 Cpmk Process capability index-Cpmk

(a third- generation capability index that
incorporates the features of Cpk and
Cpm).

The capability Ratio

= 0.0063

(CR) =

= 0.744

At present the specification range used by

the process is 74.4%.
6.2 Process capability index Cpk
(Second- generation capability index,
developed from the original Cp)

Cpk = Min
(

Cpk = Min

Therefore, Cpk = 0.026

6.3. Process capability index- Cpm
(Second- generation capability index,
developed from the original Cp)
=

The case study analysis reveals that Cp is not

equal to Cpk which implies that the process is
not exactly centered. Also, Cp, Cpk Cpm and
Cpkm are not very nearer in their magnitude
and hence it can be stated that process under
study is not exactly centered. It is noticed
that even process is under statistical control
,stable over time and have potential to meet
the given specification limits, there has been
rejections as large as 4,64,626.00 products
out of 1 million products due to the shift of
the
process
mean
towards
upper
specification limit as shown in figure 9. In
order to reduce the scrap, it is necessary to
shift the process mean as close as possible to
the target value (i.e., 205.00 mm).

7. Process capability evaluation

after shifting the process mean
to the specification mean

0.33

Where, USL and LSL are upper and lower

specification limits. is process standard
deviation, is process mean, T is target
value.

After adjusting the process mean to the

target value of 205.00, data was collected
and presented in the Table 4.

Table 4. Measured values of bore dia after adjusting the process mean
Sample
1.
2.
3.
4.

406

1
205.007

2
205.006

3
204.982

4
205.005

5
Range
Mean
204.996 0.025
204.999

204.995

205.000

205.010

205.021

205.016 0.026

205.008

204.994

205.013

204.990

205.005

204.985 0.028

204.997

204.998

205.016

205.037

204.994

204.991 0.046

205.007

Y. Wooluru, D.R. Swamy, P. Nagesh

5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.

205.019

204.994

205.022

204.983

205.011 0.039

205.006

204.993

205.038

205.028

205.007

204.985 0.053

205.010

205.005

205.015

205.019

205.010

205.026 0.021

205.015

204.985

204.983

205.027

205.005

205.019 0.044

205.004

204.982

204.985

204.982

204.990

205.024 0.042

204.993

204.997

204.987

204.997

204.994

204.973 0.024

204.990

204.986

205.020

204.983

204.993

205.000 0.037

204.996

204.998

205.004

205.007

205.011

205.001 0.013

205.004

204.983

205.017

205.001

204.995

204.985 0.034

204.996

204.986

205.012

205.006

205.014

204.996 0.028

205.003

205.014

204.982

204.998

205.020

205.007 0.038

205.004

205.001

205.013

204.994

205.001

204.980 0.033

204.998

205.020

205.009

204.993

204.995

205.010 0.027

205.005

205.000

205.013

205.001

205.003

205.005 0.013

205.004

204.991

205.017

204.996

204.963

204.992 0.054

204.992

205.015

204.988

205.006

204.980

204.992 0.035

204.996

R=0.66, =0.033, = 4100.03, =205.001

8. Process capability evaluation

after adjusting the process mean
8.1 Construction of and R- Chart to
assess the stability and uniformity of the
process
Control limits for Chart

205.001 + [(0.577)

(0.033)] =205.020

= 205.001 - [(0.577)
(0.033)] =204.982
Control limits for R-Chart
= 2.114 x0.033 =0.070,
UCL=
= 0.00 x 0.033 =0.0000
LCL=
From table, n=5
.
.

=0.00.

=2.114

407

Figure 5. Control charts for case study data

It has been observed that all the plotted
sample range and mean values are within the
control limits on both R-Chart as well as XBar chart and there are no indications of
Trend, shift, run and clustering. Hence it is
concluded that the process is under statistical
control and operating under the influence of
only chance causes of variation. i.e., the
process is stable over time.

8.2 Normal probability plot and

histogram for validating the Normality
assumption (After adjusting the process
mean).
Graphical methods including the histogram
and normal probability plot have been
constructed to check the normality of the
case study data. Figure 6 display the
histogram and Figure 7 shows normal
probability plot for the case study data. The
sample data appears to be normal.

Figure 6. Histogram for case study data

408

Y. Wooluru, D.R. Swamy, P. Nagesh

Figure 7. Normal Probabilities Plot for case study data

Test results for normal probability plot for
the data from MINITAB -14 statistical
software output shows Mean: 205.00,
Standard deviation: 0.0143, Anderson
Darling test statistic. 0.399 and P- value:
0.358 is greater than the significance level
(= 0.05).This implies that the data is
distributed normally .Thus, It has been

concluded that the sample data can be

regarded as taken from a normal process.
8.3 Construction of Run Chart for
checking the assumption of randomness of
the case study data

Figure 8. Run chart for case study data

409

Interpretation of Run chart: The P-values

for clustering, mixtures, trends and
oscillation are greater than alpha value of
0.05. The actual numbers of runs are close to
the expected number of runs. Hence, it is
concluded that the data is independent and
random.
After
validating
the
three
critical
assumptions, the process capability of the
boring operation would be quantified.

= 84.60%. This means that the

manufacturing process uses 84.60% of
specification band.

8.4 Process Capability Index - Cp

8.5 Process capability index - Cpk

(Second-generation
capability
index,
developed from the original Cp)

The mean of the sample means ( X = ) and

The capability Ratio,

(CR) =

The specification range used by the process

is 84.46%.

average range ( R ) value are computed. The

process standard deviation is obtained from
the formula, = R / d2. With the help of
statistical table the value of d2 for a subsample size of five is noted as 2.326. The
process standard deviation is estimated by:

= 0.846

Cpk = Min
(

Cpk = Min

= R /d2, ' = 0.033/2.326 = 0.0141

Therefore, Cpk = 1.158

Process capability index,
Cp =

8.6 Process capability index - Cpm

(Second-generation
capability
index,
developed from the original Cp)

= 1.182

Percentage of specification band used by the

process Cr = (1/Cp) x 100 = (1/1.182) x 100
= 1.18

Process capability index-Cpmk (a thirdgeneration capability index that incorporates

the features of Cpk and Cpm).

9. Estimation of non-conforming
Gears

9.1 Estimation of non- conforming gears

(Before shifting the process mean)

X
= 1.155

410

MINITAB-14 Statistical Software has been

used to perform process capability analysis
and found the number of gears fail to
conform to the specification limits per
million, as shown in the figure 9.

Y. Wooluru, D.R. Swamy, P. Nagesh

Figure 9. Process Capability Analysis before adjusting the process mean

9.2 Estimation of non-conforming Gears
(After adjusting the process Mean to
specification Mean)

Figure 10. Process Capability Analysis after adjusting the process mean
After adjusting the process mean, it has been
noticed that the process is under statistical
control, stable over time and capable of
meeting the given specification limits. Even
after shifting the process mean, it has been
noticed from the Figure.10, that rejections as
few as 159 gears as scrap and 337 gears as
rework out of 1 million gears. Still there is
an opportunity to reduce the scrap and
rework by identifying and reducing the

causes of variation and achieve Cp and Cpk

value is 1.33.

10. Estimation of non-conforming

Gears
10.1 Estimation of confidence interval for
Cp
The expressions for the above capability
indices involve population parameters. In

411

practice, they are replaced by their sample

estimations ( and ), leading to point

estimates
and mk .Here,
confidence intervals have been provided for

under the assumption of

normality of the distribution of the quality
characteristic. A 100 (1-)% confidence
interval for (Kushler and Hurley, 1992).

Where

and

are the lower

and upper percentage points on the chisquare distribution with (n-1) degree of
freedom.

1.023

1.547

10.2 Estimation of confidence interval for

Cpk
An approximate confidence interval at 95%
confidence level for Cpk has been estimated
under the assumption that the quality
characteristic is normally distributed.
(Kushler and Hurley, 1992).

Where n represents the sample size used to

calculate
and
represents the
standard normal value for a tail area of

0.984

11. Testing of Hypothesis

11.1 Testing of Hypothesis for Cp
Testing of the hypothesis has been done
whether the product would be accepted by
the customer, if the Cp Index for this
operation exceeds 1.00, at a significance

LCL

412

1.33
level of 0.05.
Let, Ho: Cp 1.00, H1 = Cp >1.00 with =
0.05
In this case, Cp=1.182. A one- sided
hypothesis test with = 0.05, at 95% lower
confidence limit of Cp is obtained as below.

= 1.048

Y. Wooluru, D.R. Swamy, P. Nagesh

The hypothesized value of

=1.0 < LCL.
It implies that the true value of
manufacturing capability Cp is not less than
1.048 with 0.95 level of confidence. Hence,
the null hypothesis (Ho) cant be accepted
and conclude that the process is capable of
meeting the given specification.
11.2 Testing of Hypothesis for Cpk
Testing of the hypothesis has been done
whether the product would be accepted by
the customer, if the Cpk Index for this
operation exceeds 1.00, at a significance
level of 0.05.
Let, Ho: Cpk 1.00, H1 = Cpk > 1.00 with
= 0.05.
In this case, Cpk=1.158 .A one - sided
hypothesis test with =0.05, at a 95% lower
confidence limit (LCL) of Cpk is obtained as
LCL=

Lower Confidence Limit (LCL) =1.012

The hypothesized value of
(1.0) < LCL
(1.012), It implies that the true value of
manufacturing capability Cpk is not less than
1.048 with 0.95 level of confidence. Hence,
the null hypothesis (Ho) can be rejected and
conclude that the process is capable of
meeting the given specification.

12. Results and discussion

and after adjusting the process mean) have

been quantified using Cp, Cpl, Cpu , Cpk.
Cpm and Cpmk indices and presented in the
Table-6 , Table-5 shows commonly used
capability
requirement
and
the
corresponding precision conditions. In this
case, before adjusting the processes mean,
Cpmk < 1.00. It implies that process was
inadequate. After adjusting the process mean
1.00< Cpmk < 1.33; this indicates that the
process is marginally capable and caution
needs to be taken regarding the process
consistency and rigid process control is
required using R-Chart. After adjusting the
process mean to the target value i.e., 205.00
mm, Process found to be potential as well as
capable of meeting the specification and it
requires close control as its Cpk value is less
than 1.33. The Cp Cpm Cpmk indices
show their sensitivity in exhibiting the
results. The difference in the Cpmk and the
Cp value indicates that the process mean is
still not exactly centered with the
specification limits .Table 5 shows some
commonly used capability requirement and
the corresponding precision conditions.
Table 5. Commonly used capability
requirement and the corresponding precision
conditions
Precision condition
Cpmk Values
Inadequate
Cpmk < 1.00
Marginally capable
1.00Cpmk< 1.33
Satisfactory
1.33Cpmk< 1.67
Excellent
1.67Cpmk< 2.00
Super
2.00 Cpmk

After validating the critical assumptions on

the process, process capability for (before
Table 6. Quantified values of Cp, Cpl, Cpu, Cpk. Cpm and Cpmk indices
Index value before
Index value after
Fig.no.
Index
shifting
shifting
the Process Mean
the Process Mean
1
Cp
1.34
1.182
2
Cpk
0.026
1.158
3
Cpu
0.026
1.158
4
Cpl
2.66
1.205
5
Cpm
0.33
1.180
6
Cpmk
0.0063
1.155

413

13. Conclusion
The case study was conducted in an
automotive industry and examined using
Cp,Cpk, Cpm and Cpmk index, to show the
importance of process capability analysis for
monitoring and ensuring the products quality
to satisfy the customers requirements.
Before quantifying the indices, validation of
the three critical assumptions were tested
with the help of statistical tools like control
charts, histogram and normal probability plot
and run chart using the statistical softwareMinitb-14.The quantified values presented in
the Table 6. Shown their sensitiveness in
exhibiting the results. Among all the indices

Cpmk does provide more capability

assurance with respect to process yield and
process loss to the customers than the other
two indices Cpk and Cpm. This is a desired
goal according to todays modern quality
improvement theory, as reduction of process
loss (variation from the target) is as
important as increasing the process yield
(meeting the specification).The construction
of confidence interval for Cpk is not straight
forward as that of Cp. In this paper Bissels
approach has been used to construct the
confidence interval of Cpk, As it is
significantly
influenced
by
sample
size,sample size of 100 observations is used
for the process capability study.

References:
Bissel, A.F. (1990). How Reliable Is Your Capability Index? Journal of Applied Statistics, 39,
331-340.
Carot, M.T., Sabas, A., Sanz, J.M. (2013). A new approach for measurement of the efficiency
of Cpm and Cpmk control charts, International Journal for Quality Research, 7(4), 605-622.
Chen, K.S., Pearn,W.L., Lin, P.C. (2003). Capabilit measures for processes with multiple
characteristics, Quality and Reliability Engineering International, 19, 101-110.
Gildeh, B.S., & Asghari, S. (2011). A new method for constructing confidence interval for
Cpm based Fuzzy data, International Journal for Quality Research, 5(2), 67-73.
Gunter, B.H. (1989). The use and abuse of Cpk ,1-4, Quality Progress, 22(1), 72-73(3), 108109(5), 79-80(7).
Juran, J., Gryna, F., (1988). Juran's Quality Control Handbook, 4th edition., McGraw-Hill
,New York.
Kumar, S.G., (2010). A quantitative approach for detection of unstable Processes using a run
chart. Quality Technology and Quantitative Management, 7(3), 231-247.
Kushler, R.H. (1992). Confidence Bounds for Capability Indices. Journal of Quality
Technology, 24(4), 118-195.
Montgomery, D.C. (2000). Introduction to Statistical Quality Control, Fourth Edition, John
Wiley and Sons, Inc.
Prabhuswamy, M.S., & Nagesh, P. (2007). Process capability Analysis made simple through
graphical approach, Kathmandu University. Journal of science, Engineering and
Technology, 1(3).
Prabhuswamy, M.S., Nagesh, P. (2010-2011). Process capability validation and short - Long
term process Capability Analysis with case study, Proceedings of ETIMES-2006.
Ray, S., & Das, P. (2011). Improving machining process capability by using Six Sigma,
International Journal for Quality Research, 5(2),109 -121.

414

Yerriswamy Wooluru

Swamy D.R

P. Nagesh

JSS Academy of Technical

Education,
Bangalore-560060
India
[email protected]

JSS Academy of Technical

Education,
Bangalore-560060
India
[email protected]

Sri Jayachamarajendra
College of Engineering,
JSS Centre for Management
studies
Mysore-570006
India
[email protected]

415

416

Y. Wooluru, D.R. Swamy, P. Nagesh