A.Saravanan et al.

/ International Journal of Engineering Science and Technology (IJEST)

IMPLEMENTATION OF QUALITY
CONTROL CHARTS IN BOTTLE
MANUFACTURING INDUSTRY
A.SARAVANAN,
Assistant professor,
Department of Instrument ation Technology,
M.S.Ramaiah Institute of Technology, Bangalore-560054, India
E-mail id:[email protected]
DR P. NAGARAJAN
Associate professor
Department of Chemical Engineering,
Annamalai University,Annamalai Nagar-608002 India
Email id: [email protected]
Abstract:
Statistical process control is a powerful collection of problem solving tools useful in achieving process stability
and improving capability through the reduction of variability.SPC can be applied to any process. The major
objective of SPC is to quickly detect the occurrence of assignable causes of process shifts so that investigation
of the process and corrective action may be taken before many non conforming units are manufactured. The
eventual goal of SPC is the elimination of variability in the process. It may not be possible to completely
eliminate variability, but the control chart is an effective tool in reducing variability as much as possible. The
main objective of this paper is to implement SPC schemes, study of control charts and comparison performance
of various control charts for bottle manufacturing data.
Keywords: Statistical Process Control, Control Charts, Mean, Minitab
1. Introduction:
A control chart is a statistical tool used to distinguish between variations in a process resulting from common
causes and variation resulting from special causes. Every process has variation. Some variation may be the
result of causes which are not normally present in the process. This could be special cause variation. Some
variation is simply the result of numerous, ever-present differences in the process.
This is common cause variation. Control Charts differentiate between these two types of variation.[1]
A control chart consists of three horizontal lines called; Upper Control Limit (UCL), Center Line (CL) and
Lower Control Limit (LCL). The center line in a control chart denotes the average value of the quality
characteristic under study. If a point lies within UCL and LCL, then the process is deemed to be under control.
Otherwise, a point plotted outside the control limits can be regarded as evidence representing that the process is
out of control and, hence preventive or corrective actions are necessary in order to find and eliminate the
assignable cause or causes.[2]
2. What is a Control Chart?
A control chart presents a graphic display of process stability or instability over time. Every process has
variation. Some variation may be the result of causes which are not normally present in the process. This could
be special cause variation. Some variation is simply the result of numerous, ever-present differences in the
process. This is common cause variation. Control Charts differentiate between these two types of variation. One
goal of using a Control Chart is to achieve and maintain process stability. Process stability is defined as a state
in which a process has displayed a certain degree of consistency in the past and is expected to continue to do so
in the future. This consistency is characterized by a stream of data falling within control limits [3].
Types of control charts:
There are two main categories of Control Charts, those that display attribute data, and those that display
variables data [4]. The control chart decision tree is shown in the fig 2.1
Attribute Data: This category of Control Chart displays data that result from counting the number of
occurrences or items in a single category of similar items or occurrences. These “count” data may be expressed
as pass/fail, yes/no, or presence/absence of a defect.
Variables Data: This category of Control Chart displays values resulting from the measurement of a continuous
variable. Examples of variables data are elapsed time, temperature, and radiation dose.

ISSN : 0975-5462 Vol. 5 No.02 February 2013 335

 A.Saravanan et al. / International Journal of Engineering Science and Technology (IJEST)

Figure2.1 Control Chart Decision Tree

Are you Is Use XmR

Charting NO Data are sample YES
Chart for
Attribute Variables Size = 1?  Variables
data Data

YES
YES
NO

Use XmR chart for For sample size

attribute data
Attribute data Between 2 and
15, use X-Bar
And R Chart
3. Methodology:
In this study we have applied the control chart approach to the data collected from ACE glass containers pvt-ltd
Company. The variable under study is height of the specific type of bottle. In data set there were 16 batches of 5
observations each shown in Table-1..The following approaches based upon the principles of SQC were applied
i) The upper and lower control limits for Range(R) chart are
ௗଷ ௗଷ
LCL= D3ܴത where D3= 1- 3 Center line = ܴത UCL= D4ܴത where D4= 1 + 3
ௗଶ ௗଶ
ii) The Upper and lower control limits for Mean (ܺത) are
UCL= ܺധ +A2ܴത LCL= ܺധ -A2ܴത Center line = ܺധ
iii) The upper and lower control limits for S chart are
UCL= B4ܵҧ Center line = ܵҧ LCL= B3ܵҧ
The upper and lower control limits for EWMA (Exponentially weighted Moving Average) chart are


LCL  X  kˆ r 1  1  r   / n2  r 
2i

UCL  X  kˆ r 1  1  r   / n  2  r 
2i

Where r = EWMA weight parameter (0 to 1), σ = process standard deviation, k= default limits (3σ)

ISSN : 0975-5462 Vol. 5 No.02 February 2013 336

 A.Saravanan et al. / International Journal of Engineering Science and Technology (IJEST)

4. Main results
4.1 X bar, Range, S control charts for height of the bottle
Table -1 Data set for height of the specific type of Bottle.

Sample Observations
no ഥ
ࢄ R S

1 122.36 122.82 122.24 122.82 122.94 122.63 0.70 0.313

2 122.48 122.40 122.70 122.90 123.34 122.74 0.84 0.330
3 122.42 122.74 123.00 123.22 123.12 122.90 0.80 0.319
4 123.42 122.96 123.84 123.02 122.76 123.20 1.08 0.494
5 122.48 123.10 123.26 123.34 122.82 123.00 0.86 0.350
6 123.04 122.00 121.80 122.66 122.76 122.40 1.24 0.530
7 122.75 122.62 123.82 123.10 123.46 123.15 1.20 0.490
8 123.32 122.92 122.90 122.94 123.36 123.08 0.46 0.220
9 123.00 123.35 123.40 123.45 123.67 122.52 0.69 0.267
10 122.80 122.84 123.10 122.24 122.80 122.75 0.86 0.314
11 122.96 122.82 122.90 123.04 122.92 122.90 0.22 0.006
12 122.98 122.84 122.70 123.00 123.14 122.93 0.44 0.166
13 121.96 122.00 122.34 122.26 123.26 122.36 1.30 0.520
14 122.94 122.64 122.80 122.92 122.74 122.80 0.20 0.110
15 121.94 122.90 122.90 122.92 122.86 122.70 0.98 0.420
16 122.84 123.04 123.04 122.98 122.76 122.90 0.28 0.090
For the data set the X bar control chart is shown in figures 4.1(a)
For X bar chart the LCL=122.335 and UCL= 123.3, Center line=122.81

Xba r C ha r t
123.4

UC L=123.285
123.2

123.0
Sample Mean

_
_
122.8 X =122.81

122.6

122.4
LC L=122.335

122.2
1 2 3 4 5 6 7 8
Sa mple

Figure4.1 (a) X bar chart for height of the bottle

Any point falling outside the control limits indicates that assignable causes had affected the process and the
process is out of control. Looking into the X bar chart it is observed that all the points are falling with in the
control limits. It means that the assignable cause does not affect the process. (subgroup size is 2,so 8x2 = 16
samples)

ISSN : 0975-5462 Vol. 5 No.02 February 2013 337

 A.Saravanan et al. / International Journal of Engineering Science and Technology (IJEST)

R Chart
1.6
UC L=1.523
1.4

1.2

1.0
Sample Range
0.8

0.6
_
R=0.466
0.4

0.2

0.0 LCL=0

1 2 3 4 5 6 7 8
Sample

Figure4.1 (b) Range chart for height of the bottle

Looking into the R chart (fig 4.1(b)) it is observed that all the points are falling within the control limits. It
means that the assignable cause does not affect the process.(subgroup size is 2,so 8x2 = 16 samples)
Looking into the S chart (fig 4.1(c)) it is observed that all the points are falling within the control limits. It
means that the assignable cause does not affect the process.(subgroup size is 2,so 8x2 = 16 samples)
S C ha rt
0 .5

U C L= 0 .4 4 6 4
0 .4
Sample StDev

0 .3

0 .2
_
S = 0 .1 3 6 7
0 .1

0 .0 LC L= 0

1 2 3 4 5 6 7 8
S a m p le

Figure4.1(c) S chart for height of the bottle

The exponentially weight moving average (EWMA) technique has been widely used in order to monitor the
process mean, since it weighs the average of all past and present observations.EWMA is defined as
Zi= λXi + (1- λ)Zi-1
Where λ is the constant factor (0< λ< 1), Zi is the predicted value of process mean at time i, Xi is the observed
value of process mean at time i.

ISSN : 0975-5462 Vol. 5 No.02 February 2013 338

 A.Saravanan et al. / International Journal of Engineering Science and Technology (IJEST)

EWMA Chart of X
123.1
UCL=123.0741

123.0

122.9
_
_
EWMA

122.8 X=122.81

122.7

122.6
LCL=122.5459
122.5
1 3 5 7 9 11 13 15
Sample

Figure4.1 (d) EWMA chart for height of the bottle

For the same data set Exponentially weighted moving average control chart is shown in figure 4-1(d).
UCL= 123.074, LCL = 122.545, Center line = 122.81
In the EWMA control chart of X it is observed that all the points falling inside the control limits and there is no
sharp shift is observed so the process is IN- control.

EWMA Chart of R
1.2
UCL=1.1457
1.1

1.0

0.9

0.8 _
_
EWMA

X=0.7594
0.7

0.6

0.5

0.4
LCL=0.3730
0.3
1 3 5 7 9 11 13 15
Sample

Figure4.1 (e) EWMA chart for Range

For the same data set Exponentially weighted moving average control chart for Range is shown in figure 4-1(e).
UCL= 1.1457, LCL = 0.7594, Center line =0.373.

ISSN : 0975-5462 Vol. 5 No.02 February 2013 339

 A.Saravanan et al. / International Journal of Engineering Science and Technology (IJEST)

EWMA Chart of S
0.5
UCL=0.4743

0.4

_
_
EWMA

0.3 X=0.3087

0.2

LCL=0.1431

0.1
1 3 5 7 9 11 13 15
Sample

Figure4.1 (f) EWMA chart for Standard Deviation

In the EWMA control chart for Range it is observed that all the points falling inside the control limits and there
is no sharp shift is observed so the Process is in control. For the same data set Exponentially weighted moving
average control chart for S is shown in figure 3-e). UCL= 0.4743, LCL = 0.1431, Center line =0.3087.In the
EWMA control chart for S it is observed that all the points falling inside the control limits and there is no sharp
shift is observed so the process is IN-Control.
5. Conclusion
In any manufacturing industry monitoring of their key characteristics is most important. The absence of a
quality control section or the inefficient quality control section may be one cause of poor quality. The X bar
control chart and range control chart are the well known and the most popular tools for detecting in control or
out of control signals in the statistical quality control (SQC).The Control Charts has shown worth in the
production process. In this study we have applied the control chart approach to the data collected from ACE
glass containers pvt-ltd Company. The variable under study is height of the specific type of bottle. In all figures
the points are inside the control limits. This indicates that the process is
IN control and the assignable cause does not affect the process. All figures are computer generated charts
From Minitab, popular PC based statistics package.
REFERENCES
[1] D.C.Montgometry,[2000] ‘Introduction To Statistical Quality Control’, 2nd Edition,Willey,New York,
[2] A.J.Duncan, [1986] Quality control and Industrial statistics,5th edition,Irwin,Homewood,Illinois .
[3] W.A.Shewhart, Statistical Method From The Viewpoint Of Quality Control,Dover Publications,Inc,New York
[4] Raz,T & Wang J.H(1990), Probabilistic & Membership approaches in the construction of control charts for linguistic.
“Data.Production Planning And Control”

ISSN : 0975-5462 Vol. 5 No.02 February 2013 340