1 Statistical Quality Control

STATISTICAL
QUALITY CONTROL

Quality Control for a

Fiber Rolls factory
Supervised by:
Prof. Tamer A. Mohamed
Eng. Kareem Safty
Prepared by:
Ahmed Abdelhamid 202475
Muhamed Barakat 199360
Ahmed Ihab 193948
Ahmed Elkaliouby 202196
Sara Atef 203362

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 2 Statistical Quality Control

i. Abstract.
This report aims to identify the importance of quality control. And how to apply quality
control techniques to produce product service with high quality and make the customer
satisfied.

ii. Acknowledgements
We would like to thank Prof. Tamer A. Mohamed and Eng. Kareem Safty for their great and
appreciated effort with us during the whole semester and helping us to understand the meaning
and the concept of statistics quality control.

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 3 Statistical Quality Control

Contents
i. Abstract .................................................................................................................... 2
ii. Acknowledgement ................................................................................................... 2
iii. List of figures ........................................................................................................... 4
iv. List of tables............................................................................................................. 4
v. Abbreviations .......................................................................................................... 5
1. Introduction .............................................................................................................. 6
2. History...................................................................................................................... 6
3. Experimental work ................................................................................................... 8
4. Data analysis and discussion ................................................................................... 9
4.1 Quality control ............................................................................................... 9
4.1.1 CTQ .......................................................................................................... 9
4.1.2 Sampling inspection technique .............................................................. 11
4.1.3 Measuring process .................................................................................. 12
4.1.4 Control charts ......................................................................................... 12
4.1.5 Process capability ................................................................................... 16
4.2 Quality Improvement................................................................................... 17
4.2.1 Histogram ............................................................................................... 17
4.2.2 Pareto diagram........................................................................................ 18
4.2.3 Cause & effect diagram .......................................................................... 19
5. Conclusion ............................................................................................................. 20
6. References ............................................................................................................. 21
7. Appendix ................................................................................................................ 22
vi.

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 4 Statistical Quality Control

iii. List of Figures

Figure 1 Fredric W. Taylor ..................................................................................... 7
Figure 2 Philip Crosby ............................................................................................ 7
Figure 3 Edwards Deming ...................................................................................... 8
Figure 4 Joseph M. Juran ........................................................................................ 8
Figure 5 Measuring process ..................................................................................... 8
Figure 6 CTQ tree .................................................................................................. 10
Figure 7 X-bar chart ............................................................................................... 13
Figure 8 R-chart ..................................................................................................... 13
Figure 9 X-bar & S-charts ..................................................................................... 14
Figure 10 X-bar Chart revised ............................................................................... 15
Figure 11 S-chart revised ...................................................................................... 15
Figure 12 X-bar Chart revised ............................................................................... 15
Figure 12 R-chart revised ...................................................................................... 15
Figure 13 Capability chart ..................................................................................... 16
Figure 14 Histogram .............................................................................................. 17
Figure 15 Pareto diagram....................................................................................... 18
Figure 16 Fishbone diagram .................................................................................. 19

iv. List of Tables

Table 1 Collected data ......................................................................................... 11
Table 2 Pareto data .............................................................................................. 18
Table 3 Factor for variable control charts ............................................................ 23

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 5 Statistical Quality Control

v. Abbreviations
CL Centre line

UCL Upper control limit

LCL Lower control limit

USL Upper specification limit

LSL Lower specification limit

CTQ Critical to quality

Cp Capability index

Cpk Actual capability

σ Standard deviation of the process

TQM Total quality management

SPC Statistical process control

R Range

S Standard deviation of the samples

̅
𝑿 Mean

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 6 Statistical Quality Control

1. Introduction.
Quality is the degree of excellence, is how a certain thing is good and meet the requirements.

Quality control is a process that used by any business such as; company, factory, restaurant,

etc. This process is used to check if the product(s) or the service(s) is controlled and meet the

specifications. Management and employees are the responsible for the high quality of any

business by planning and establishing a certain rules and controls to make the production

process under control and occur as required. Moreover, to achieve high quality of the products

or services, the ability of defining the problems that may obstruct the production process and

how to deal with them is important in the quality control process. In this report we will discuss

the quality control process for fiber roll.

2. History.
There are many stages that have effects on growing the quality control process that is used and

implemented in all modern companies and factories or any business to keep the quality of the

products as required. Due to quality control process, there are some benefits that are the

organization will grow and the customers will be satisfied as the product meet their

requirements. By implementing total quality management (TQM) which is depend on

quantitative methods by using statistical tools. Furthermore, using human resources to make

sure that the future needs are fulfilled not only the sufficiency of the customer satisfication

(Always room for improvement). The start of development of mass production industry is prior

1900. So, Fredric W. Taylor clarify some principle of how to manage the work and the

productivity. His work is composed of many tasks and this leads to a huge improvement in

productivity and the quality of his products is also improved. Beside his ways for work there

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 7 Statistical Quality Control

are two concepts for work which are certain time for the work to be accomplished, number of

units should be produced in a certain period.

Fig 1: Fredric W. Taylor

There are many contributions from different names in the development of the quality control

process. There is an author that has a very influential contribution in the development of the

quality control process. Which is Philip Crosby that have many achievements in quality control

field. “Quality is free”, this statement is said by Philip Crosby and then he converts this

statement to a book called “Quality is free” and released in 1979. “Quality is free” means that if

we pay for improve the quality and do the things perfect in the first time, this will lead to the

customer will know and note the difference between the good quality product and the bad

quality product. Another theory was established by Philip Crosby which is Zero Defect Theory.

In this theory, he stated if there is anything not cooperate in the project and not increase the

value of the project should be eliminated. That’s mean that everything being perfect in the first

time to avoid or decrease the money and time spent later.

Fig 2: Philip Crosby

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 8 Statistical Quality Control

Dr. W. Edwards Deming and Dr. Joseph M. Juran have an important role in developing the

methods of statistical quality control in japan since the second world war. Due to the effort by

Dr. W. Edwards Deming in reestablishing Japan as the pioneer of the industrial force. For that,

his great effort was appreciated by the government and they assign a special reward with his

name called “Deming prize” in 1951.

Fig 3: Edwards Deming Fig 4: Joseph M. Juran

3. Experimental work.
First, we started the project by visiting the Fiber Home factory to collect the data of the length
of the fiber rolls they produced. And then analyzing the data and determine the quality of the
production line by using quality control techniques.

Fig 5: measuring process

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 9 Statistical Quality Control

4. Data analysis and Discussion

Using statistical quality control tools, the quality control of the collected data will be discussed.

4.1 Quality Control

4.1.1 CTQ
CTQ: critical to quality it helps knowing and studying the customer’s needs and translate the
information into measurable product and process required.
Need: they need a product with their desired characteristics and good quality. Additionally, the
customer needs the length of fiber rolls to be suitable for his needs.
Drivers: customers want to check if all characteristic of good fibers are available so they will
check if product is temperature resistant, sustainable, in shape and having high tensile strength.
Requirements: for temperature resistance: product should resist the temperature effect as it has
high melting point and it consists of conjugated aromatic structures that produce a high level of
thermal resistance
-for sustainability: product should be made from natural materials or recycled ones and free of
chemicals.
- for being in shape: product should meet the standard specifications of the required shape.

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 10 Statistical Quality Control

Critical need Quality driver CTQ Requirments

Price Good price

for the
quality

Quality Strength

Good Being in Meet

customer shape standard
service specificatio
n

No
sustainability
chemicals

Temp. High
Fig 6: CTQ Tree resistance melting
point
DY2 Mechanical Engineering
[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 11 Statistical Quality Control

4.1.2 Sampling inspection technique

We took 20 samples and 5 subgroups, which means that we have 20 sample number.
Each sample number has 5 readings and sample numbers were calculated every 1
hour. So the total values of lengths of fiber rolls are 20 x 5 = 100 values of lengths.
The following table shows our data.

Sample Date Time Observations 𝑋̅ R S

number
1 10:00 268 260 255 258 264 261 13 5.099019514
2 11:00 260 265 270 255 258 261.6 15 5.941380311
3 24/4 12:00 267 263 257 259 260 261.2 11 4.207136794
4 13:00 260 255 265 262 258 260 10 3.807886553
5 12:00 260 266 259 255 260 260.4 11 4.037325848
6 13:00 260 255 263 268 262 261.6 13 4.722287581
7 25/4 14:00 265 260 255 259 270 261.8 15 5.805170109
8 15:00 257 260 260 258 262 259.4 5 1.949358869
9 9:00 262 260 250 259 265 260.2 10 3.701351105
10 10:00 260 255 256 254 259 256.8 5 2.588435821
11 26/4 11:00 266 259 260 255 253 258.6 13 5.029910536
12 12:00 267 263 255 270 260 263 15 5.873670062
13 10:00 260 259 260 257 262 259.6 5 1.816590212
14 11:00 258 255 264 260 259 259.2 9 3.271085447
15 27/4 12:00 266 260 265 260 259 262 7 3.240370349
16 13:00 260 258 262 260 255 259 7 2.645751311
17 12:00 262 257 260 251 261 259 7 2.915475947
18 13:00 255 257 265 262 260 259.8 10 3.962322551

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 12 Statistical Quality Control

19 28/4 14:00 267 262 257 260 259 261 10 3.807886553

20 15:00 260 260 260 259 257 260.4 9 3.361547263
𝑋̿ 𝑅̅ 𝑆̅ = 3.8892
= 260.28 = 10

Table 1: Collected data

4.1.3 Measuring process.

It refers to the machines, methods, and people involved in determining the value of any
variable on a unit. The machines used in our fiber rolls production line factory consists of 8
machines inside of it each doing a different process. There are three fiber production line
machines in our factory and five people supervising on each one of the machines. In order to
measure the variable, we need (Length), we used two tools: 1. Digital measuring tool (Laser
meter) and 2. Simple ruler. The data measured is for lengths of fiber rolls.

4.1.4 Control Charts.

They are graphs used to study how a process changes over time, it has a central line for the
average, an upper line for the upper control limit, and a lower line for the lower control limit
and these lines are determined from historical data. The out-of-control signal is a single point
outside the control limits and this means that if this point existed in any data, the process is out
of control but if all the points are between the UCL and LCL so the process is controlled. We
will study X-bar R and X- bar S for our collected data. We can construct the control charts
using 4 important component which are, upper control limit (UCL), lower control limit (LCL),
center line (CL), and plotted points.
 X-bar R Charts
Firstly, we add column for X-bar and column for Range in table 1. We calculate X-
double bar by adding all X-bar and divided them by number of samples. R-bar
calculated by adding all ranges and divided them by number of samples.
For 𝑋̅ Chart: For R Chart:

̿ = 260.28
CL = 𝑋 CL = 𝑅̅ = 10
UCL = 𝑋̿ + (𝐴2 )(𝑅̅) = 266.05 UCL = 𝐷4 𝑅̅ = 21.14
LCL = 𝑋̿ − (𝐴2 )(𝑅̅) = 254.51 LCL = 𝐷3 𝑅̅ = 0
For n = 5 , 𝐴2 = 0.577 , 𝐷3 = 0 , 𝐷4 = 2.114
DY2 Mechanical Engineering
[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 13 Statistical Quality Control

x-bar chart
268

266
UCL, 266.05
264

262
Sample mean

260
CL, 260.28
258

256

254 LCL, 254.51

252

250
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample

Fig 7: X-bar Chart

R- chart
25
UCL, 21.14
20
 X bar S charts
Sample range

15
Firstly, we add column standard deviation in table 1. S-bar calculated by adding all
Standard deviations and divided them by number of samples. CL, 10
10

LCL, 0
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Sample

Fig 8: R Chart

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 14 Statistical Quality Control

For 𝑋̅ Chart: For S Chart:

̿ = 260.28
CL = 𝑋 CL = 𝑆̅ = 3.8892
UCL = 𝑋̿ + (𝐴3 )(𝑆̅) = 265.8 UCL = 𝐵4 𝑆̅ = 8.12
LCL = 𝑋̿ − (𝐴3 )(𝑆̅) = 254.7 LCL = 𝐵3 𝑆̅ = 0
For n = 5 , 𝐴3 = 1.427 , 𝐵3 = 0 , 𝐵4 = 2.089

Fig 9: X-bar and S Charts

We construct the Xbar-S chart using minitab, there is small error in the values. It may be due to that
the software take a certain number of decimals

 Comment: as shown in the X-bar R and X-bar S charts that all points all fall between
the control limits that indicate the process is in control (no assignable causes)
X-bar S is more accurate than X-bar R, because it depends on standard deviation and
analyze each point in the sample. while X-bar R depends on the range.
 to study the case of out-of-control process we will assume that there is two points are
out of control and recalculate the data after excluding these points. We will assume that
samples 10 and 12 are out of control. The following charts are after we remove samples

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 15 Statistical Quality Control

10 and 12. There is no indication for assignable causes so, the process is in control. The
following equations that are used to construct the revised X-bar R and X-bar S

x-bar chart s-chart

268 UCL, 9 UCL,
265.81 8.045
266 8
264 CL, 7
262 260.322 Sample S/dev
Sample mean

6
260
5 CL, 3.8512
258
4
256
3
254 LCL, 2
252 254.8
250 1 LCL, 0
248 0
1 2 3 4 5 6 7 8 9 101112131415161718 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sample Sample

Fig 10: X-bar Chart Revised Fig 11: S Chart Revised

X bar-chart R-bar chart

270
25 UCL,
UCL, 266.09
21.14
265 20
CL, 260.322
260
Axis Title

15
CL, 10
255 10
LCL, 254.55
250 5
LCL, 0
245 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 2 3 4 5 6 7 8 9 101112131415161718

Fig 12: X-bar Chart Revised

Fig 12: R Chart Revised

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 16 Statistical Quality Control

4.1.5 Process capability

If the process is capable that means meeting specification limits and producing good parts and
products. Its central idea is to control variation to avoid product defects. There are minimum
and maximum acceptable values, specifications are not the same like control limits, as control
limits come from control charts and are based on the data while specifications are the numerical
requirements of the system.
𝑆̅ 3.889
𝑈𝑆𝐿 = 260.28 + 3 𝐶4 = 0.94 , 𝜎= = = 4.137
𝐶4 0.94

𝐿𝑆𝐿 = 260.28 − 3
𝑈𝑆𝐿−𝐿𝑆𝐿
Capability index (Cp) = = 0.24
6𝜎

Comment: there is waste in the process and the products not meet the specifications.
min(𝑈𝑆𝐿−𝑋̿)
Cpk = = 0.283
3𝜎

Cpk < 1 means that the process is producing product that does meet the specifications.

Fig 13: Capability chart

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 17 Statistical Quality Control

4.2 Quality Improvement

4.2.1 Histogram.
A graphical representation that divides a set of data points into an easily read visual
by collecting ranges and class intervals in addition to resembling a bar graph
clarifying the distribution of frequency of the data. For each class interval, there is
its bar. The height of each bar depends on the value of its frequency. Histogram can
be categorized into Symmetric, Bimodal, random, uniform, and probability
histogram

Fig 14: Histogram

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 18 Statistical Quality Control

4.2.2 Pareto diagram

They show the ordered frequency counts of data and they are frequently used to determine
which areas of process should be prioritized. Pareto graph is represented in bars, the bar length
is the frequency and they're organized with the longest on the left and the shortest on the right
(in descending order). In our case, we categorized the problems which lead to a delay in
production (hr). The following table illustrates it.
Category Delay (Hours) Cumulative frequency %CF

Over requests 35 35 44.87

Material out of stock 26 61 78.2

Lack in digital tracking 7 68 87.17

Packaging issue 5 73 93.5

Technical problems 3 76 97.4

Getting raw material from 100

storehouse 2 78

∑= 78

Table 2: Pareto data

Pareto Diagram
35
35 100%
30 90%
26
94% 97% 100% 80%
Delay Hours

25 87% 70%
78% 60%
20
50%
15 40%
10 45% 7 30%
5 20%
5 3 2
10%
0 0% Delay (Hours)
C frequency

Categories

Fig 15: Pareto diagram

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 19 Statistical Quality Control

 Comment:
From the pareto diagram, it is clear that “over requests” problem has the highest effect
on the stability of the process, and the delay in production. This should be solved
urgently, while the problem “getting raw material from storehouse” is a minor problem.
4.2.3 Cause & effect diagram
It is called fishbone diagram because it looks like one it is used to identify many possible
causes for an effect or problem and it can be used in brainstorming as it sorts the ideas into
useful information. In order to do a cause-and-effect diagram you should choose a problem and
right it on the right side of chart in a box, draw a center line from the box and draw arrows
running to it, then think about the major categories of causes of the problem. We use headings
as:
Methods
Machines (equipment)
People (manpower)
Materials
Measurement
Environment
Then write the causes of each category of these all heading to the main arrow and by this you
brainstormed all the causes that can happen for the chosen problem. The following is a cause &
effect diagram for the delay of our fiber rolls production.

Delay in
production

Fig 16: Fishbone diagram

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 20 Statistical Quality Control

5. Conclusion
In conclusion, quality control is one of the most important processes in the modern industry and
in all business. As it ensures that if the customers are satisfied. Quality control process
investigates the defects in the products and there are different statistical control processes to
determine if the manufacturing process is in control and to determine if there is any problem
during the production process.

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 21 Statistical Quality Control

6. References.
https://www.shmula.com/philip-crosby-contributions-to-the-theory-of-process-improvement-

and-six-sigma/27873/

https://www.qualitymag.com/articles/84673-probing-the-limits-the-claim-to-good-quality-is-

free

Montgomery, D. (2003). Applied Statistics and Probability for Engineers. 3rd edition, Wiley.

Besterfield, D. (2013). Quality Improvement. 9th edition, Prentice Hall.

Montgomery, D. (2012). Statistical Quality Control. (7th edition), Wiley

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 22 Statistical Quality Control

7. Appendix

d
dd

d
d
d
d
d
d
d
d
d
d
d
d
d

d
d
d
d
d

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]
 23 Statistical Quality Control

Table 3: Factors for variable control charts

DY2 Mechanical Engineering

[Type here] 21MECH41C
[Type here] Semester 2 (2021/2022)
]