Introduction of Quality Control

Statistical quality control (SQC)

 SQC is one of the most important applications of the statistical
techniques in industry.
 These techniques are based on the theory of probability and sampling
 It is extensively used in almost all industries.
 It covers almost all aspects of production i.e. quality of materials,
quality of human resources, quality of machines, quality of
management.
 Objectives of SQC
 To maintain the quality of manufactured products.
 To better use of available materials.
 To locate and identify the process faults in order to control the
scrap.
 To maintain standards of product.
 To achieve quality and consistency.
 To make adjustments in the process.
 Random variation (chance causes)
 Random variation or chance causes of variation are natural to
any production process.
 It is beyond the control of human hand.
 It can not be detected or prevented.
 It is also known as common or inherent variation.
 It can not be eliminated.
 E.g. Voltage fluctuations, earthquake.
 Systematic variation (assignable causes)
 They are known as non random causes.
 They are also called special causes of variation.
 They can be identified.
 They can be eliminated.
 E.g. negligence, improper handling of machines, faulty
equipment, unskilled or inexperienced technical staff etc.
 Uses Of SQC
 A production process is said to be in a state of statistical control
if it is operating in the presence of chance causes only and is
free from assignable causes of variation.
 It is used to asses the causes of variation in the quality of the
manufactured product.
 It enables us to determine whether the quality standards are
being met without inspecting every unit produced in the process.
 It is used to isolate the random variation and systematic
variation.
 It is used to detect , identify and eliminate the assignable causes
of erratic fluctuations whenever they are present.
Control Charts
 Control Charts show sample data plotted on a graph with CL, UCL, and
LCL, where
 CL = Central Line = Mean
 UCL = Upper Control Limit
 LCL = Lower Control Limit
 They are important tools of production management and the control of
quality.
 They are made by Walter A. Shewhart.
Contd..
 They help to detect the presence of assignable causes of erratic variations
in the process.
 It is an indispensable tool for bringing a process under statistical control.
 Its construction is based on plotting 3-σ control limits and any sample point
going 3-σ control limits is an indication of the lack of statistical control i.e.
presence of some assignable causes of variation which must be traced,
identified, and eliminated.
Contd..
 A typical control chart consists of
 Upper Control Limit (UCL)
 Lower Control Limit (LCL)
 Central Line (CL)
 For statistic (t) :
 UCL = E(t) + 3 S.E.(t)
 LCL = E(t) - 3 S.E.(t)
 CL = E(t)
 E(t) = Mean = Mathematical expectation.
Contd..
 If E(t) = µt
 Var(t) = σt2
 UCL = µt + 3 σt
 LCL = µt - 3 σt
 CL = µt
Control Chart
Control Chart
Control Chart
The 𝑋 Chart
Contd..
 When population mean (µ) and s.d. (σ) are given,
 UCL = µ + 3(σ/√n)
 LCL = µ - 3(σ/√n)
 CL = µ
Contd..
Contd..
Contd..
 Control Limits for 𝑋 chart
 UCL = 𝑋 + [ 3 𝑅 /(d2√n)]
 LCL = 𝑋 - [ 3 𝑅 /(d2√n)]
 CL =𝑋
where d2 = control chart factor.
Contd…
 Control Limits for 𝑋 chart
 UCL = 𝑋 + A2 𝑅
 LCL = 𝑋 - A2 𝑅
 where A2= 3 /(d2√n) another control chart
Example
 For each of following cases, find the UCL, LCL and CL for an 𝑋 chart
based on the given information.
 n = 9, 𝑋 = 26.7, 𝑅 = 5.3
 n = 22, 𝑋 = 8.1, 𝑅 = 7.4
 Also construct 𝑋 (x- bar) control chart.
Example
Altoona Tire Company sells its ATC-50 tires with a 50,000-mile tread –life
warranty. Lorrie Ackerman, a quality control engineer with the company/runs
simulated road tests to monitor the life of the output from the ATC-50
production process . From each of the last 12batches 0f 1,000 tires , she has
tested 5 tires and recorded the following results , with 𝑋 and R measured in
thousands of miles:
Contd..

Use the data above to help Lorrie construct an 𝑋 (x-bar) chart.

Is the production process in-control? Explain.
Control Chart for Range (R-Chart)
 It is constructed for controlling the variability (dispersion) of the product.
 Control Limits for R Chart:
 CL = Central Line = 𝑅
 UCL = 𝑅 + [ 3d3 𝑅 /d2]
 LCL = 𝑅 - [ 3d3 𝑅 /d2]
 where d2 and d3 are control chart factors.
Contd..
Control Limits for R – bar chart

UCL = 𝑅 D4,where D4 = 1+ 3d3 /d2

LCL = 𝑅 D3,where D3 = 1- 3d3 /d2

where D3 and D4 are control chart factors.

Example
 For each of following cases, find the UCL, LCL and CL for an R chart
based on the given information.
 n = 9, 𝑋 = 26.7, 𝑅 = 5.3
 n = 22, 𝑋 = 8.1, 𝑅 = 7.4
 Construct R – bar chart.
Example
Construct R chart for the data given in Altoona Tire
Company. Is the variability in the tread life of the ATC-50
in control? Explain.
Control Chart for Attributes (P Chart)
 It is used when sample unit as a whole is classified as defective or non-
defective, winning or losing, head or tail etc.
 In many cases we merely want to assess whether or not a certain item is
defective.
 We can observe a number of defective items from a particular sample of
series of samples. This is referred as “sampling by attributes.”
Contd..
If p is Given,
If a standard value of p is given, then the control limits for p chart are
UCL = p + 3 √(p q /n), where, q = 1 – p

CL = p

LCL = p - 3 √(p q /n), where, q = 1 – p

Contd..
P is not Given
If no standard value of p is given, then the control limits for the fraction
nonconforming are
p (1  p )
UCL  p  3
n
CL  p
p (1  p )
where LCL  p  3
n
k k

 D  pˆ
i i
p i 1
 i 1
kn k
Contd..
 Where,
 D1, D2,…, Dk are k defective units.
 n = no of samples.

 Pi = Di /n

 Di = n Pi , i= 1,2,…,k.
Example
A process that produces bearing housings is investigated. Ten samples of
size 100 are selected.
Samp 1 2 3 4 5 6 7 8 9 10
le
5 2 3 8 4 1 2 6 3 4
Nonc
onf.

Is this process operating in statistical control?

Contd..
Here,
n = 100, k = 10
Sample 1 2 3 4 5 6 7 8 9 10
Nonconf. 5 2 3 8 4 1 2 6 3 4
Fraction
0.05 0.02 0.03 0.08 0.04 0.01 0.02 0.06 0.03 0.04
Nonconf.(ṗ)
k

 pˆ i
p i 1
 0.038
k
Contd..

Control Limits are:

0.038(1  0.038)
UCL  0.038  3  0.095
100
CL  0.038
0.038(1  0.038)
LCL  0.038  3  0.02  0
100
Contd..
Contd..
Conclusion: Since a number of sample points lie inside the control limits,
this process is operating in statistical control.
Example
For each of following cases, find the UCL, LCL and CL for a p chart based on
the given information.

n = 4, 𝑝 = 0.10.

n = 117, 0.42 is the target value of p.

Example
An inspection of 10 sample of size 400 each from 10 lots give the following
number of defective items:
17, 15, 14, 26, 9, 4, 19, 12, 9, 15
Calculate control limits for the number of defective items and state whether
the process is under control or not.
Control Charts for Number of Defects per Unit ( c-Chart)
 It is used when we count the number of defects per unit rather than
classifying a unit as defective or non defective.
 When actual occurrence of a defect in any unit are very large.
 When actual occurrence of a defect is very small.
 If Ci denotes the number of defects observed in the ith inspected item,
we can safely assume that Ci has a Poisson distribution.
 Let this Poisson distribution have a mean of λ for a process in control.
Condt..

Since E Ci    and VarCi    , the control limits are :

  3 .

If k items are inspected, then the unbiased estimator of  is :

k
1
C 
k
C .
i 1
i

The estimated control limits then become :

C 3 C .
Example
During an examination of equal length of cloth , the following
are the number of defects observed:
2, 3, 4, 0, 5, 6, 7, 4, 3, 2.
Draw a control chart for the number of defects and comment
whether the process is under control or not ?
Solution:
Let the number of defects per unit (equal length of cloth) be denoted by c.
Then the average number of defects in the 10 sample units is given by
c = (∑c) /k = 2+ 3+…+2 /10 =3.6

The 3-σ control limits for c-chart are

UCL= c + 3 × √(c) = 3.6 + 3× √(3.6) = 9.29
LCL= c – 3 × √(c) = 3.6 – 3 × √(3.6) = -2.09≈0
CL = 3.6
C-Chart
Control Chart
Contd..
Conclusion: Since all the sample points are within the control limits , the
process in statistical control.
Thank You