Beruflich Dokumente
Kultur Dokumente
net/publication/330798733
CITATIONS READS
0 236
1 author:
Tanmoy Das
Florida State University
3 PUBLICATIONS 1 CITATION
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Tanmoy Das on 01 February 2019.
a critical issue for the Multistage Manufacturing Generally, the choice of 3-sigma for control
System (MMS)5, in which each station may limits is determined by keeping a proper balance
inevitably shift to the out-of-control condition between control for Type I and Type II errors. For
resulting in higher nonconforming rate and manufacturing processes, Shewhart also found
larger quality loss. Ongoing inspection on real the coefficient 3 to be excellent in providing this
production time can tremendously help engineers balance12. As a result, the American Society for
to reduce defects in production line6. Testing and Materials printed
_ tables for control
On the basis of probability as defined by limits using 3-sigma for X and standard deviation
Tshebycheff ’s inequality, Shewhart7 originally (s) charts. Although the manufacturing industry
chose three standard errors (3-sigma units) as has widely adapted these 3-sigma limits for their
the marker for data points exceeding the limits processes, other coefficients may be reformed for
of statistical control. This inequality states that a particular application by carefully balancing the
no less than 89% of a sampling distribution, no effects of Type I and II.
matter the shape, lies within three standard errors Naturally, companies do not usually measure
of its mean. all products as this may be counterproductive or
In general, economically unfeasible. As a negotiation, most
P(μ–kσ≤X≤μ + kσ) ≥ 1 – 1/k2 companies select rational subgroups of products
for every k > 0. often using a random sampling scheme13. These
If the underlying distribution is normal and the subgroups are then laboriously checked for
process is in control, 99.73% of the monitored quality and their summary measurements have
means lie within these 3-sigma control limits. On been charted for quality control purposes since
the other hand, if the statistics to be monitored last century.
were not normal, at worst, 89% of these would With emphasis on early detection and prevention
be within the control limits. With a more refined of the problems, SPC has a discrete benefit over
version of Tshebycheff ’s inequality8,9 for strongly other quality approaches such as inspection of
unimodal statistical distributions (e.g., that of the end product14. SPC necessitates distinct processes
standard deviation estimator, s), at least 95% of and a discipline of following them. It requires an
the monitored statistics would
_ be within the limits. atmosphere where employees are not penalized
For the Shewhart Mean (X) chart, by virtue of the when problems are detected.15
central limit theorem, normality further increases
the percentage of means within 3-sigma control 3 METHODOLOGY
limits to around 99.73%. Researchers have found 3.1 Problem Statement
nonnormality effects on Shewhart control charts The problem to be solved in the research work
to be minimal10,11. is to collect data and identify the special causes of
variation in the industrial systems by control chart
and improve the sigma level.
Data collection methods used in this research
involves the use of observation, company
documents and structured experiments. After
gathering all necessary information, statistical
methods have been implemented to find the
defects. A statistical approach has been taken
to eliminate all nonconforming products by
full inspection and detect the out-of-control
occurrence using statistical process control
(SPC) techniques, in this research work, we
have applied Control Chart. introduction of a
system to eliminate all nonconforming products
Fig1: Control Chart with 3 sigma control limit by full inspection and detect the out-of-control
occurrences using statistical process control
MIST Journal of Science and Technology|Volume 5|Issue 1|December 2017
Control Chart for Continuous Quality Improvement.......... Tanmoy Das 67
(SPC) techniques is required. In this research contractual specifications on the products’ mean
work, Control Charts have been applied. and standard deviation are helpful to set up limits
for the process. In industrial SPC applications,
3.2 Choice of Control Chart these means and standard deviations are often
Depending on observation type, we decide what called “control” or “target” values. The formula
type of control charts best suited the scenario. For is,
example, for variable data, we may use Xbar-R
chart, for attribute data, we would adapt p-chart
or np-chart.
Control limits for R-chart:
3.3 Mathematical Model Calculate the control limits for the R chart. The
If the process is in control, almost all points will upper control limit is given by UCLr. The lower
lie within the upper control limit (UCL) and the control limit is given by LCLr.
lower control limit (LCL). The formula is,
–
These control limits are chosen so that if the UCLr = D4 R
process is in control, nearly all of the sample LCLr = D3 R
points will fall between them. If all the data
points are situated between the control limits, 4.0 EXPERIMENTAL DETAILS
the process is presumed to be in control, and no 4.1 Data Collection
action is necessary. However, a data point that SPC data has been collected in the form of
plots outer of the control limits is interpreted as measurements of a product feature or process
evidence that the process is out of control, and instrumentation readings. The data is then
investigation and corrective action are required to recorded and tracked on various types of control
find and remove the common and special causes charts, based on the type of data being collected.
responsible for this performance. It is customary It is imperative that the correct type of chart is
to connect the sample points on the control chart used gain value and obtain useful information.
with straight-line segments, so that it is easier to The data can be in the form of continuous
visualize how the sequence of points has evolved variable data or attribute data. The data can also
over time. be collected and recorded as individual values or
Even if all the points plot inside the control limits, an average of a group of readings.
if they behave in a systematic or nonrandom To conduct our research, and implement the
manner, then this could be an indication that the methodology proposed, we have consulted
process is out of control. For example, if 18 of with two companies in Bangladesh, namely
the last 20 points plotted above the center line DBL Group and Bangladesh Lamps Limited-
but below the upper control limit and only two Transcom. In this section, data collection method
of these points plotted below the center line but for these two companies will be described.
above the lower control limit, we would be very In DBL Group, data collection is exclusively based
suspicious that something was wrong. Once the on their Knitting division. Most observations
process is in control, all the plotted points would are recorded manually on forms customized
have a random pattern. Methods for looking to collect specific information then input into
for sequences or nonrandom patterns can be a computer database (Microsoft Excel). Data
applied to control charts as an aid in detecting is also taken automatically through the use of
out-of-control conditions. Usually, there is a automated devices that send information directly
reason why a particular nonrandom pattern to a database.
appears on a control chart, and if it can be found In particular, most of the data are related to the
and eliminated, process performance can be buyer wise order receiving and delivery Status in
improved. 2016. Additionally, the records related to wastage
from garments, and the reasons for revision
Control limits for
_ X-bar chart: for different products are collected. Moreover,
For a Shewhart X SPC chart, past information or the wastage info of Jute (fabric wastage, fabric
MIST Journal of Science and Technology|Volume 5|Issue 1|December 2017
Control Chart for Continuous Quality Improvement.......... Tanmoy Das 68
cutting, and jute Sewing) are being prepared. control specialist to embrace the idea of SPC to
We have examined the improvement of the improve their system efficiency.
quality of product in the samples of apparel 4.2.1 Scenario 1: DBL group (Knitting division)
products in Knitting division of DBL Group, Regression analysis is conducted to test for
Gazipur, Bangladesh during 2015 – 2017. The statistically significant slope ( α = 0.005).
main reasons for revision of the products are Figure 2 depicts the Xbar-R chart where in
tabulated in Table 1: Reason for revision. R chart, three data points are out of control.
One crucial question to be answered is how to Therefore, we would be going to drop them by
increase the number of client order at Knitting adjusting the system parameters. Technically, we
division at DBL? need to figure out the special causes of variation
for these 3 data points.
Table 1: Reason for revisions
4.2 Analytics
In this study, we explored the use of SPC charts
with a real data set for two industries in Bangladesh.
Because SPC charts have not previously been
used extensively in the manufacturing companies
in this country, the findings from the research
would, hopefully, give the engineers and quality Fig 3: Frequency of the client order received
MIST Journal of Science and Technology|Volume 5|Issue 1|December 2017
Control Chart for Continuous Quality Improvement.......... Tanmoy Das 69
Stem-and-Leaf Display: C1
The data summary of the control charts for the All control charts have used three sigma limits
data set of BLL is listed in Table 8. Here, the and have been generated using Minitab software
number of sample is 150, three sigma limit has version 17. Some of the statistical analyses have
been implemented as guided by researchers. been performed using Microsoft Excel 2016.
By moving the control limits farther from the with appropriate quality engineering tools.
center line, we decrease the risk of a type I
error - that is, the risk of a point falling beyond 6.0 CONCLUSION
the control limits, indicating an out-of-control In this study, we have applied control chart to
condition when no assignable cause is present. visually and quantitatively examine the variation
On the other hand, spreading the control limits in manufacturing process. The control chart in
will also surge the danger of a type II error - that Figure 8 clearly shows us the data points which
is, the jeopardy of a point falling between the are far away from the mean or expected value as
control limits when the process is really out of per engineering specification.
control. If we move the control limits closer to In summary, the practice of the SPC tools is very
the center line, the opposite effect is obtained: little because of the challenges systems are facing
The risk of type I error is increased, while the risk in Bangladesh or other developing countries.
of type II error is decreased. Even though other Hence, some serious motivation and inspiration
interpretations and creations of these data are are mandatory to put the manufacturing on a
possible, we believe that ours are probable and consciousness initiative to adapt SPC.
hope they are valuable. Although SPC tools are commonly implemented,
control chart in particular, in the manufacturing
5.1 Need for SPC in Bangladesh industries of most developed countries; we do
By visualizing data, we can easily detect defects not see any proper implementation of SPC in
in manufacturing process. In Bangladesh, Bangladeshi industries. Therefore, embracing the
most of the industries are still considered to be idea of SPC in the industries in Bangladesh is
lagging behind to apply statistical process control highly recommended.
methodologies to improve the quality of the
system. Acknowledgement
The introduction of the SPC system as a tool Special thanks to Mr. Sanaour Hossain, Industrial
would greatly enrich the quality aspects in the Engineer in Quality division at Bangladesh
system. The use of SPC is expected to assist in Lamps Limited - Transcom/ Philips for providing
solving common quality related problems and industry based data from manufacturing lines.
ought to be maintained by topper management I also show gratitude to Mr. Rakibul Islam,
levels in the industries in Bangladesh. Lecturer, Dept. of IPE, MIST, to assist me for
collecting data from Knitting division of DBL
5.2 Future Work group. I am appreciative to the department of
From this research endeavor, we know the exact IPE at MIST for continuous support during this
place where the defects would occur in the research work.
system. The future work will involve identifying
the reasons of the defects, and eliminating them