1.6.

3 Statistical Process
Control (SPC)
 Statistical Process Control, commonly
referred to as SPC
 is a method for controlling and ideally,
improving a process through statistical
analysis.
 A. Basics of Statistical
Process Control
 Statistical Process Control (SPC)
 is about quality control efforts that occur during
production/operation
 is the term used to describe the set of statistical tools
used by quality professionals
 is the application of statistical techniques to manage
the operation of processes
 is used to evaluate process output to decide if a
process is “in control” or if corrective action is needed
 involves monitoring production process to detect and
prevent poor quality during the production process
 Basics of Statistical Process
Control
 The main goals of SPC are:
 Provide practical tools for controlling quality
 Establish an ongoing measurement and
verification system
 Improve quality and reliability of products
and services without increased cost
 Increase productivity and reduce cost
 Improve customer satisfaction.
 B. SPC and Variability
 SPC
 Measures the value of a quality characteristic, and
 Helps to identify a change or variation in some quality
characteristic of the product or process.
 Variation in the production process leads to defects
and lack of product consistency.
 Variability is the cause of most quality problems:
 variation in products, in times of deliveries, in ways of
doing things, in materials, in people’s attitudes, in
equipment and its use, in maintenance practices
etc
 SPC and Variability

 The variability can be:

 Random:
 Variations due to system cause, or common
cause, or non-assignable cause
 Non-random:
 Variations due to specific causes, or
assignable causes, or special causes
 SPC and Variability
 Random variations
 Non-random variations
 are inherent to every
process  special causes
 are natural occurrences  identifiable to specific
cause and correctable
 are the result of
“chance” or “common”  include equipment out of
causes adjustment, defective
materials, changes in
 non assignable parts or materials, broken
 Example: machine part machinery or equipment,
clearances, bearing operator poor work
wear, lubrication, and methods, or errors due to
so on lack of training
 C. SPC and Control Charts

 Control Charts:
 Are important tools in statistical process
control
 Are time-ordered plot of sample statistics
 Used to distinguish between random
variability and nonrandom variability
 Have upper and lower limits, called control
limits, that define the range of acceptable
(i.e., random) variation for the sample
statistic
 SPC and Control Charts
Out of control
Upper
control
limit

Process
Average or
Center line

Lower
control
limit

1 2 3 4 5 6 7 8 9 10
Sample number or time
SPC and Control Charts
 SPC and Control Charts

 Steps Used in Developing Process Control

Charts
 Identify critical operations in the process where
inspection might be needed.
 Identify critical product characteristics.
 Determine whether the critical product characteristic
is a variable or an attribute.
 Select the appropriate process control chart.
 Establish the control limits and use the chart to
monitor and improve.
 Update the limits.
 SPC and Control Charts

 Two Types of Control Charts

 Attributes Control Charts
 Variables Control Charts
 I. Control Charts for Attributes

 Attribute control charts are developed for

 a product characteristic that can be
evaluated with a discrete response
 good – bad; yes – no; defective-non-
defective; acceptable –non-acceptable ; pass-
fail…
 Attribute data are counted (e.g., the
number of defective parts in a sample,
the number of calls per day);
 Control Charts for Attributes
Two Types of attribute control charts
 p-charts
 c-charts
 a) p-Chart

 P-Charts
 are statistical tools used to evaluate the
proportion defective, or proportion non-
conforming, produced by a process.
 are used to monitor the proportion of defective
items in a process.
 appropriate when the data consist of two
categories of items (defective or not
defective, good or bad, pass or fail etc)
 p-Chart (cont.)

 Control limits for P-chart:

 Centerline (CL)
 Upper control limit (UCL)
 Lower control limit (LCL)
p-Chart (cont.)

CL= P
UCL = p + zp
LCL = p - zp

z = number of standard deviations from

process average
p = average sample proportion defective
p = standard deviation of sample proportion
p-Chart (cont.)

total defectives
p = total sample observations

p = average sample proportion of defective

p= standard deviation of sample proportion

p(1 - p)
p = n
 p-Chart (cont.)

Example on p-chart
 WJ jeans producing company wants to establish a p-
chart to monitor the production process and maintain
high quality. The company believes that approximately
99.7% of variability in a process (Z = 3) is random and
thus should be with in control limits. The company has
taken 20 samples (one per day for 20 days), each
containing 100 pairs of jeans (n=100), and inspected
them for defects, the results of which are as follows:
 p-Chart (cont.)
Sample Number of Proportion of Sample Number Proportion of
Number defectives Defectiveness Number of Defectiveness
(K) (K) defectives

1 6 0.06 11 12 0.12
2 0 0.00 12 10 0.10
3 4 0.04 13 14 0.14
4 10 0.10 14 8 0.08
5 6 0.06 15 6 0.06
6 4 0.04 16 16 0.16
7 12 0.12 17 12 0.12
8 10 0.10 18 14 0.14
9 8 0.08 19 20 0.2
10 10 0.10 20 18 0.18
 p-Chart (cont.)
 Construct a p-chart and determine if the process
is in control
total defectives
p = total sample observations = CL= 200 / 20(100) = 0.10

p(1 - p) 0.10(1 - 0.10)

UCL = p + z = 0.10 + 3
n 100
UCL = 0.190

p(1 - p) 0.10(1 - 0.10)

LCL = p - z = 0.10 - 3
n 100
LCL = 0.010
 0.20

p-Chart 0.18 UCL = 0.190

(cont.) 0.16

0.14

Proportion defective
0.12
p = 0.10
0.10

0.08
Observation:
0.06

 Out of control 0.04

 During day 2 0.02 LCL = 0.010

 During day 19 2 4 6 8 10 12 14 16 18 20
Sample number
 b) C-chart

 C-charts
 use the actual number of defects per item in a
sample
 focus on the actual number of defects per
sample (each item is a sample)
 C-chart

 The mean number of defects

represented by: c
total defects
c= = CL= Center line
total sample observations
 Construction of c-Chart

UCL = c + zc
c = c
LCL = c - zc
where
c = the average number of defects
z = number of standard deviations from
process average
C = number of defects per sample
c = standard deviation of sample defects
 c-Chart (cont.)

Example on C-chart
 The RH hotel has 240 rooms. The management
inspectors not only check for normal housekeeping
service defects like clean sheets, dust, room supplies,
or towels, but also for defects like missing TV remote,
poor TV picture quality, defective lamps, a
malfunctioning curtain pull, malfunctioning clock, etc.
The following table shows the result from 15 inspection
samples conducted at random during a one-month
period:
 Construct a 3 sigma C-chart and show if the process is
in control or out of control
 c-Chart (cont.)

Number of defects in 15 sample rooms

Sample Number of Sample Number of
Defects Defects
1 12 9 13
2 8 10 15
3 16 11 12
4 14 12 10
5 10 13 14
6 11 14 17
7 9 15 15
8 14 Total 190
c-Chart (cont.)

190
c= = 12.67
15

UCL = c + zc
= 12.67 + 3 12.67
= 23.35

LCL = c - zc
= 12.67 - 3 12.67
= 1.99
 24
UCL = 23.35
21

18

Number of defects
c = 12.67

15
c-Chart 12
(cont.) 9

3 LCL = 1.99

2 4 6 8 10 12 14 16
Sample number
 II: Control Charts for Variables

 A control chart for variables

 is used to monitor characteristics that can
be measured and have a continuum of
values
 Variable measure
 a product characteristic that is continuous
and can be measured
 E.g.. weight, speed, length, force,
diameter, time….
 Control Charts for Variables

 Two most commonly used control charts

for variable monitors :
 Mean ( x –bar ) Chart
 uses process average of a sample
 monitors the absolute value, or the location, of the
process average
 Range chart ( R-Chart )
 uses amount of dispersion in a sample
 monitors the spread (range) of the output
distribution, that is, the piece-to-piece variability.
 Control Charts for Variables

 The sample mean

 is the sum of the observations divided by the
total number of observations:
 Control Charts for Variables

 The range
 is the difference between the largest and the
smallest observations in a sample. An estimate of
the process standard deviation based on a sample
is given by
Control Charts for Variables
Control Charts for Variables
x-bar Chart Formula:

UCL = x=+ A2R LCL = x= - A2R

where
=
x = average of sample means = CL
A2 = a tabular value used to determine control
limits and comparable to 3 sigma limits
R = the average range value
 X-bar and R Charts Example:

 The quality controller of a company producing

and selling different types of bearings has
taken 10 samples (during a 10-day period) of
5 slip-ring bearings (i.e., n=5) and measured
the diameter of the bearings. The individual
observations from each sample (or subgroup)
are shown as follows:
 X-bar and R Charts Example:
OBSERVATIONS (SLIP- RING DIAMETER, CM)
SAMPLE (k) 1 2 3 4 5
1 5.02 5.01 4.94 4.99 4.96
2 5.01 5.03 5.07 4.95 4.96
3 4.99 5.00 4.93 4.92 4.99
4 5.03 4.91 5.01 4.98 4.89
5 4.95 4.92 5.03 5.05 5.01
6 4.97 5.06 5.06 4.96 5.03
7 5.05 5.01 5.10 4.96 4.99
8 5.09 5.10 5.00 4.99 5.08
9 5.14 5.10 4.99 5.08 5.09
10 5.01 4.98 5.08 5.07 4.99
 x-bar Chart Example:

 Construct X-bar chart and R chart and

show if the process is in control or out of
control
 Step 1: Compute Average and
Range of Each Sample Number
OBSERVATIONS (SLIP- RING DIAMETER, CM)
SAMPLE (k) 1 2 3 4 5 x R
1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
2 5.01 5.03 5.07 4.95 4.96 5.00 0.12
3 4.99 5.00 4.93 4.92 4.99 4.97 0.08
4 5.03 4.91 5.01 4.98 4.89 4.96 0.14
5 4.95 4.92 5.03 5.05 5.01 4.99 0.13
6 4.97 5.06 5.06 4.96 5.03 5.01 0.10
7 5.05 5.01 5.10 4.96 4.99 5.02 0.14
8 5.09 5.10 5.00 4.99 5.08 5.05 0.11
9 5.14 5.10 4.99 5.08 5.09 5.08 0.15
10 5.01 4.98 5.08 5.07 4.99 5.03 0.10
50.09 1.15
Step 2: Compute the Average of Sample
Averages and Average of Sample Ranges

= x 50.09
x= = = 5.01 cm
k 10

∑R 1.15
R= = = 0.115
k 10
Step 3: Compute the Upper and
Lower Control Limits for X-bar
Chart

UCL = x + A2R = 5.01 + (0.58)(0.115) = 5.08

LCL = x - A2R = 5.01 - (0.58)(0.115) = 4.94

Retrieve Factor Value A2

 5.10 –

5.08 –
UCL = 5.08
5.06 –

5.04 –

5.02 –
x= = CL= 5.01
Mean

5.00 –

4.98 –
Step 4:
4.96 –
Plot x-
bar 4.94 – LCL = 4.94

Chart 4.92 –
| | | | | | | | | |
1 2 3 4 5 6 7 8 9 10
Sample number
Step 5: Construct R- Chart

UCL = D4R LCL = D3R

R
R= k

where
R = range of each sample
k = number of samples
Construct R- Chart

CL = R
UCL = D4R = 2.11(0.115) = 0.243

LCL = D3R = 0(0.115) = 0

Retrieve Factor Values D3 and D4

Construct R- Chart
0.28 –
0.24 –
UCL = 0.243
0.20 –
0.16 –
Range

R = 0.115
0.12 –
0.08 –
0.04 – LCL = 0
0– | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10
Sample number
 Using x- bar and R-Charts
Together
 Process average and process variability must be
in control
 It is possible for samples to have very narrow
ranges, but their averages might be beyond
control limits
 It is possible for sample averages to be in
control, but ranges might be very large
 It is possible for an R-chart to exhibit a distinct
downward trend, suggesting some nonrandom
cause is reducing variation
 Sample Size Determination

 Attribute charts require larger sample sizes

 50 to 100 parts in a sample
 Variable charts require smaller samples
 2 to 10 parts in a sample