e-ISSN: 2582-5208

International Research Journal of Modernization in Engineering Technology and Science

( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com

PROCESS CAPABILITY ANALYSIS IN A PLASTIC JAR MANUFACTURING

ORGANISATION: A CASE STUDY
Fahad Khan*1, Dr. Apratul Chandra Shukla*2
*1Post-Graduation Student, Dept. Of Mechanical Engineering, Ujjain Engineering College Ujjain, India.
*2Professor, Dept. Of Mechanical Engineering, Ujjain Engineering College Ujjain, India.
DOI : https://www.doi.org/10.56726/IRJMETS51535
ABSTRACT
It’s been a while since process capability and control charts have been widely used in manufacturing
organisation to monitor a process for quality improvement. Still these tools are considered to be extremely
effective in defect control and sustaining the quality production. This paper presents a case study using SPC
(Statistical Process Control) tool to find process capability index (Cpk) of 15 litre injection moulding jar
manufacturing in an edible oil extraction plant, where this jar is used as a safe packaging solution. By using data
of weight variation of 900 samples, ̅ and R charts are drawn. Also process capability indices are calculated to
analyse the process capability. The results are presented and inferences are drawn to assess the quality of jar as
per required standard.
Keywords: ̅ And R Chart, Process Capability Indices, Quality Control, Statistical Process Control (SPC).
I. INTRODUCTION
Statistical Process Control is a method of quality control that uses statistical methods to monitor and control a
process. The goal of SPC is to ensure that the process operates efficiently, producing products with minimal
defects. SPC involves using statistical tools to analyse the performance of a process, identify any variations, and
take corrective actions if necessary [1]. By collecting and analysing data, SPC helps organizations understand
the behaviour of their processes and make informed decisions to improve quality and reduce defects. Key
components of SPC include control charts, which visually display the performance of a process over time, and
various statistical techniques such as histograms, Pareto charts, and cause-and-effect diagrams [2]. The process
capability index, often denoted as Cp, Cpk, or Ppk, is a statistical measure used to assess the ability of a process
to produce products within specifications. It is a key component of Statistical Process Control (SPC) and is used
to determine if a manufacturing process is capable of producing products that meet customer requirements.
There are several variations of process capability indices, each with its own specific calculation and
interpretation. However, they all generally compare the spread of the process output to the tolerance limits of
the product specifications. Cp and Cpk are commonly used process capability indices [3]. Cp is a measure of the
capability of a process to meet the specifications, and it is calculated as the ratio of the specification width to the
process width. Cpk, on the other hand, also considers the process centring and measures the capability of a
process relative to the target value. Ppk is similar to Cpk but is calculated using the actual process centring,
rather than the target value. It takes into account the process mean and the distance of the process from the
specification limits. In general, a higher value of Cp, Cpk, or Ppk indicates that a process is more capable of
producing products within specification limits. Values above 1 generally indicate that the process is capable,
while values closer to 0 suggest that the process is not capable of meeting specifications [4]. By using process
capability indices, organizations can identify areas for improvement, set realistic quality targets, and make
informed decisions about process adjustments and improvements to ensure that products consistently meet
customer requirements.
II. LITERARY BACKGROUND
The control charts for variables are most commonly used to evaluate the stability of a process mean and
variance because of their simplicity, ease of implementation, and satisfactory performance in practice. To
implement the ̅ and R control charts, a variable inspection is performed to measure the actual values of x and
to calculate. The ̅ control charts are still a subject of research even though they have been proposed long ago,
in the 1920s, because of their practical use and simplicity. For instance, an economic ̅ control chart with
warning limits was introduced by Chung [5]. Original schemes to detect changes in the averages of the
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[5501]
 e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com
processes were proposed by Nenes and Panagiotidou [6], Aparisi and de Luna [7], Liu and Xue [8] Garza-
Venegas [9] Haq and Khoo [10].
Many articles have been published in the literature in which the authors have attempted to solve the above
practical issue found on a control chart. For instance, Wu and Jiao [11] proposed an attribute control chart to
monitor the mean of a variable in which the run length between two consecutive nonconforming samples is
checked, where samples are considered non-conforming if a pre specified number of non-conforming items is
found in them. Wu [12] Reynolds, Amin, Arnold and Nachlas [13] were the first scholars to develop an X-bar
chart with variable sampling interval (VSI) and this research inspired a number of researchers to design and
improve control charts Runger and Pignatiello [14], Chew [15] , Naderkhani and Makis [16] , Bai and Lee [17] ,
Chen [18]. Subsequently Prabhu, Runger and Keats [19] developed an X-bar chart with adaptive sample size
(VSS) and opened the way for research on VSS control charts Costa [20] Tagaras [21]. As a further
improvement, in VSSI control charts (variable sample size and sampling interval), the sample size and sampling
interval are modified simultaneously Costa [22], De Ma-galhães [23]. Numerous studies apply economic designs
to determine the optimal parameter set for these adaptive control charts to minimize the average cost during
the control process Lin [24].
According to Montgomery [25] Statistical Process Control objects to detectnon conformities that can hinder a
process in order to prevent defectives units of a determined product to be done. For this purpose, he mentions
that control charts can be used to define a process capability by estimating parameters that act directly on the
analysed situation and to contribute on the process improvement by minimizing variability that is inherent to
the analysis. In addition, defines that the Statistical Process Control can be implemented by a control chart and
the following steps need to be followed:
a) Choose the characteristic to be measured;
b) Periodically measures the characteristic on sequential steps;
c) Calculate the mean and the range of the samples;
d) Post these values on the chart control;
e) Verify if the process is under control or not. Following these steps, it is possible to monitor the process
performance and notice its variability.
Several cases in which control charts can be applied are reported by the literature presents a study in which the
charts were useful to evaluate the operational quality of a mechanized and concluded that was possible to
assure a high variability of the process. Adaptive control charts have had increasing interest in the recent years.
Some of the more recent approaches are Safe [26] which developed control chart using Multi Objective
Optimization Yue and Liu [27].
Process Capability Index are frequently used by manufacturers and service companies to measure the
performance of process in generating outputs within the specification limits. Moreover, control charts for
variables may be developed using the PCI (Cp and Cpk) to establish and monitor the process through control
charts focused on two control mechanisms: the control limits and the capability indices define three PCIs
commonly used in the industry (Cp, Cpk and Cpm). Beside these other indices have been recently introduced in
SPC (CNp, CNpk, CNpm and CNpkm). All these indices have a degree of uncertainty because they are estimated.
Thus, it is necessary to present models that describe the probability distribution related of the estimates of
these indices. However, the practical implications of the output probability distribution in the estimation
process should also be considered. CPIs assume that the processes are an IC situation and the output
probability distribution is a normal distribution AIAG/ASQC [28]. However, in some cases, the output
probability distribution is not a normal distribution therefore, the capability indices generated with this
assumption are not accurate. To overcome this situation, several studies have proposed alternative methods,
such as the use of data transformation (non-normal distribution empirical probability distributions and mixed
distributions Ali and Riaz [29].
However, this paper will focus only on ̅, R chart, Cp and Cpk indices because they are the most common and
used by manufacturers. Cp is the Cpk that is defined as tolerance width divided by process capability,
irrespective of process centring.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[5502]
 e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com
III. METHODOLOGY
Firstly, a literature review was done to search for great information about the quality charts and its
requirements commonly used on the literature. Soon after, ̅ and Cpk was chosen to integrate this study and
provide information concerning how normal or not the process could be classified through this statistical
analysis.
Then, in order to arrange enough information for the ̅and R charts analysis, AV group of Jar Manufacturing
Ujjain, was considered in this study. The data collection was directly done with the collaboration of the AV
group oil packing administrators through data previously collected composed by a score which measures the
satisfaction of making 15 litres jars. Altogether, the information provided by the AV group were, respectively,
900 observations of randomly chosen jars weight taken in between a 3-month period. By using Statistical
process control (SPC) to find out the quality control which employs statistically method to monitor and control
a process. Process capability refers the evaluation of how well a process meets specification Cp is the ratio of
the process specification or tolerance to the natural variation of the process
Cp = Total specification or tolerance_______
Total effective range or natural variation
= (USL-LSL)
6σ
Where,
= Each observed value of jar in gram
USL = Upper specification limit
LSL = Lower specification limit
σ = Standard deviation
The process capability index (Cpk) is a measure of how process is centre within the tolerance range
Cpk = USL-̅
3σ
Cpk = ̅-LSL
3σ
If
Cp=Cpk means the process is correctly centroid
Cpk always equal to or less than Cp
Minimum requirement of Cpk value should be 1 or more
For finding ̅value
̅= Sum of all observed value in grams
Total number of Observation taken (n)
Formula for standard deviation S or (σ)
[ ̅]
S=√
For drawn ̅chart first find both limits
UCL = ̅ + 3σ
LCL = ̅ ˗ 3σ
UCL= Upper control limit for ̅ chart
LCL= Lower control limit for ̅ chart
For drawn R chart range will be
R= Highest observed value - Least observed value
̅=
For drawn R chart first find both limits
UCL= D4 x ̅
LCL= D3 x ̅
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[5503]
 e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com
Where D3 and D4 are constant value
UCL= Upper control limit for R chart
LCL= Lower control limit for R chart
IV. DATA COLLECTION AND ANALYSIS
In this section, the results of the data analysis obtained for each jar and the mathematical treatment they have
had in the SPC toolsoftware with their respective ̅charts will be exposed, as well as the calculation of their
limits of control and process capability.
Calculation
For finding ̅
̅=
̅ =501.58
For calculating standard deviation S or (σ)
[ ̅]
σ =√
Where n = 900

σ =√
σ = 6.47
Finding ̅ chart calculating limits
For upper limit UCL = ̅ + 3σ
= 501.58 + 3x6.47
= 520.99
For lower limit LCL = ̅ - 3σ
=501.58 – 3x6.47
=482.20
Therefore,
Cp = (USL-LSL)
6σ
= (520-480)
6x6.47
= 1.03
Calculation for finding ̅
̅=
= 11817
900
= 13.13
Finding R chart calculating limits
For upper limit UCL = D4 x ̅
= 2.114x13.13
= 27.75
For lower limit LCL = D3 x ̅
= 0x13.13
=0
Hence,
The value is above 1 generally indicate that the process is capable.

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

[5504]
 e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com
V. RESULTS AND DISCUSSION
In this research work Capability analysis of 15 litre jar with using SPC tool to shown in figure-1 The result show
first process capability (cp) which value is 1.03 after that process capability index value is 1.01 which greater
than 1 means process is capable. Another result shown in figure-1 that is standard deviation which we called a
sigma(σ) and the value is 6.47. The value of Lower standard limit (LSL) is 480 grams and the value of Upper
standard limit (USL) is 520 grams which means ±20 tolerance of weight is acceptable by the organisation and
normalisation weight of jar is 500 grams which indicate the centre line in figure-1.

Figure 1: Show a result by using SPC tool to find the process capability analysis and weight variation of 900 jars
Drawn X-bar chart in figure-2 shows the result of average value 501.58 which is also mention in fig-1 and
centre line of x-bar chart by taking that value calculating the Upper control limit (UCL) which is 520.99 and also
calculating the Lower control limit (LCL) which is 482.20 after that plot all 900 data point between UCL & LCL
check if there any of observation are out from both limit or not as we seen in figure-2 there is no such
observation found that go beyond the limits such as we called process is capable.
For drawn R chart in figure-3 shown the result of average value 13.13 which is also mention in fig-3 and centre
line of R chart by taking that value calculating the Upper control limit (UCL) which is 27.75 and calculating the
Lower control limit (LCL) which is 0 after that plot all data point between UCL & LCL check if there any of
observation are out from both limit or not as we seen in figure-2 there is no such observation found that go
beyond the limits.

Figure 2: Shows X-bar chart of all 900 observations which are lies between UCL& LCL centre line refers the
average value.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science
[5505]
 e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com

Figure 3: Shows R- chart of all observations which are lies between UCL & LCL centre line refers the average
value of range.
VI. LIMITATION AND FUTURE SCOPE
 Implementation of Statistical process Control is a costly endeavour. Companies need to spend acquiring the
necessary resources, material and recruiting personnel who are skilled in the quality function.
 For quality measurement it does not provide an integer number or by how much the rejected products are
defective, which would require precise measurements.
 Statistical Process Control emphasizes early detection and prevention of problems, but however it takes a
lot of time to apply successfully in a manufacturing setting, as it requires more observations.
 SPC focuses on a strict and a rigid process to follow and the goes against the new trends that favour
creativity and innovation.
VII. CONCLUSION
The value of process capability index is greater than 1 indicates that process is capable of maintaining quality
parameters but there is scope of further improvement in process capability. The organisation can work upon
better training of operators and strict compliance to maintenance schedule of various machines to ensure
sustenance of process. Similarly both X bar and R charts show that process is under control as all the points are
falling within control limits. However, it is suggested that the management can keep monitoring the process
and look for any trend in control charts which might be an indication of process going out of control in future.
Sustaining actions like strict quality check on input raw materials, close monitoring of in use failures, strict vigil
on workmanship, continuity of training of operators and compliance to maintenance schedule of various
machines will ensure long term conformance to quality parameters. Change of machines before the life expiry
could also be contemplated to ensure sustenance of quality. This study revels that control charts and process
capability analysis is highly pertinent and relevant in manufacturing units and can save large amount of quality
failures at a very reasonable cost.
VIII. REFERENCE
[1] DeVor, R.E., Chang, T. and Sutherland, J. W. (1992). Statistical Quality Design and Control. Macmillan,
New York.
[2] Domangue R. and Patch, S. C. (1991). Some omnibus exponentially weighted moving average statistical
process monitoring schemes. Techno metrics 33, 299-313.
[3] Duncan, A. J. (1986). Quality Control and Industrial Irwin, Homewood, IL.
[4] Montgomery, D. C. (1996). Introduction to Statistical Wiley,
[5] Chung KJ. An economic study of X-bar charts with warning limits. Computerin Eng.1993;24(1):1–7.
[6] Nenes G, Panagiotidou S. An adaptive Bayesian scheme for joint monitoring of process mean and
variance. Comput Oper Res. 2013;40(11):2801–2815.

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

[5506]
 e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science
( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:06/Issue:03/March-2024 Impact Factor- 7.868 www.irjmets.com
[7] Aparisi F, de Luna MA. Synthetic-.X bar control charts optimized for in-control and out-of-control
regions. Comput Oper Res. 2009;36(12):3204–3214.
[8] Liu Y, Xue L. The optimization design of EWMA charts for monitoring environmental performance. Ann
Oper Res. 2015;228(1):113–124.
[9] Garza-Venegas JA, Tercero-Gómez VG, Ho LL, et al. Effect of autocorrelation estimators on the
performance of the X bar control chart. J Stat Comput Simul. 2018;88(13):2612–2630.
[10] Haq A, Khoo BCK. A new synthetic control chart for monitoring process mean using auxiliary
information. J Stat Comput Simul. 2016;86(15):3068–3092.
[11] Wu Z, Jiao J. A control chart for monitoring process mean based on attribute inspection. Int J Prod Res.
2008;46(15):4331–4347.
[12] Wu Z, Khoo MB, Shu L, et al. An np control chart for monitoring the mean of a variable based on an
attribute inspection. Int J Prod Econ. 2009;121(1):141–147.
[13] Reynolds, M. R., Amin, R. W., Arnold, J. C., and Nachlas, J. A. (1988). X bar charts with variable sampling
intervals. Techno metrics, 30(2):181– 192.
[14] Runger, G. and Pignatiello, J. (1991). Adaptive sampling for process control. Journal of Quality
Technology, 23(2):135– 155.
[15] Chew, X. Y., Khoo, M. B. C., Teh, S. Y., and Castagliola, P. (2015). The variable sampling interval run sum
X bar control chart. Computers & Industrial Engineering, (90):25– 38.
[16] Naderkhani, F. and Makis, V. (2016). Economic design of multivariate bayesian control chart with two
sampling intervals. Int. J. Production Economics, (174):29– 42
[17] Bai, D. S. and Lee, K. T. (1998). An economic design of variable sampling interval control charts.
International Journal of Production Economics, 54(1):57– 64.
[18] Chen, Y.-K. (2004). Economic design of variable sampling interval t2 control charts-a hybrid markov
chain approach with genetic algorithms. Expert Systems with Applications, (33):683– 689.
[19] Prabhu, S. S., Runger, G. C., and Keats, J. B. (1993). X chart with adaptive sample sizes. International
Journal of Production Research, 31(12):2895– 2909.
[20] Costa, A. F. B. (1994). X bar chart with variable sample size. Journal of Quality Technology, 26:155– 163.
[21] Tagaras, G. (1998). A survey of recent developments in the design of adaptive control charts. Journal of
Quality Technology, 30:212– 231.
[22] Costa, A. F. B. (1997). X bar charts with variable sample size and sampling intervals. Journal of Quality
Technology, 29(2):197– 204.
[23] De Magalhães, M., Costa, A. F. B., and Moura Neto, F. (2009). A hierarchy of adaptive X control charts.
International Journal of Production Economics, 119(2):271– 283.
[24] Pearn, W.L. and Lin, P.C. (2004), “Testing process performance based on capability index Cpk with
critical values”, Computers & Industrial Engineering, Vol. 47 No. 4, pp. 351-369.
[25] Montgomery, D. C., and Runger, G. C. (1994). A combined adaptive sample size and sampling interval X
control scheme. Journal of Quality Technology, 27:74– 83.
[26] Safe, H., Kazemzadeh, R., and Gholipour Kanani, Y. (2018). A markov chainapproach for double-
objective economic statistical design of the variable sampling interval control chart. Communications
in Statistics-Theory and Methods, 47(2):277– 288.
[27] Yue, J. and Liu, L. (2017). Multivariate nonparametric control chart with variable sampling interval.
Applied Mathematical Modelling, 52:603– 612
[28] AIAG/ASQC (1991), Fundamental Statistical Process Control Reference Manual, AIAG, Troy, MI.
[29] Ali, S. and Riaz, M. (2014), “On the generalized process capability under simple and mixture models”,
Journal of Applied Statistics, Vol. 41 No. 4, pp. 832-852.

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

[5507]