Chapter 7

Statistical Quality Control

 Quality Control Approaches
 Statistical process control (SPC)
Monitors the production process to prevent
poor quality
 Statistical Process Control
 Take periodic samples from a process
 Plot the sample points on a control chart
 Determine if the process is within limits
 Correct the process before defects occur
 Types Of Data
 Attribute data
 Product characteristic evaluated with a
discrete choice
– Good/bad, yes/no
 Variable data
 Product characteristic that can be
measured
– Length, size, weight, height, time, velocity
 SPC Applied To Services
 Nature of defect is different in services

 Service defect is a failure to meet customer

requirements

 Monitor times, customer satisfaction

 Service Quality Examples
 Hospitals
timeliness, responsiveness, accuracy
 Grocery Stores
Check-out time, stocking, cleanliness
 Airlines
luggage handling, waiting times, courtesy
 Fast food restaurants
waiting times, food quality, cleanliness
 Process Control Chart
Upper
control
limit

Process
average

Lower
control
limit

1 2 3 4 5 6 7 8 9 10

Sample number
 Constructing a Control Chart
 Decide what to measure or count
 Collect the sample data
 Plot the samples on a control chart
 Calculate and plot the control limits on the control
chart
 Determine if the data is in-control
 If non-random variation is present, discard the data
(fix the problem) and recalculate the control limits
 A Process Is In Control If
 No sample points are outside control limits

 Most points are near the process average

 About an equal # points are above & below
the centerline
 Points appear randomly distributed
The Normal Distribution

95 %
99.74 %

-3 -2 -1 = 0 1 2 3

Area under the curve = 1.0

Control Charts and the Normal
Distribution
UCL

+3
Mean

-3
LCL
 Types Of Data
 Attribute data (p-charts, c-charts)
Product characteristics evaluated with a
discrete choice (Good/bad, yes/no, count)

 Variable data (X-bar and R charts)

Product characteristics that can be measured
(Length, size, weight, height, time, velocity)
Control Charts For Attributes

 p Charts
Calculate percent defectives in a sample;
an item is either good or bad

 c Charts
Count number of defects in an item
 p - Charts
Based on the binomial distribution
p = number defective / sample size, n

p= total no. of defectives

total no. of sample observations

UCL = p + 3 p(1-p)/n

LCL = p - 3 p(1-p)/n
 p-Chart Example
The Western Jean Company produced denim jean. The
company wants to establish a p-chart to monitor the
production process and main high quality. Western
beliefs that approximately 99.74 percent of the
variability in the production process (corresponding to
3-sigma limits, or z = 3.00) is random and thus should
be within control limits, whereas 0.26 percent of the
process variability is not random and suggest that the
process is out of control.
 p-Chart Example
The company has taken 20 sample (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 follow.
Sample # Defects Sample # Defects
1 6 11 12
2 0 12 10
3 4 13 14
4 10 14 8
5 6 15 6
6 4 16 16
7 12 17 12
8 10 18 14
9 8 19 20
10 10 20 18
 p-Chart Calculations
Proportion
Sample Defect Defective
UCL = p + 3 p(1-p) /n
1 6 .06 = 0.10 + 3 0.10 (1-0.10) /100
2 0 .00
3 4 .04 = 0.190
. . .
20 18 .18
200
LCL = p - 3 p(1-p) /n
100 jeans in each sample = 0.10 + 3 0.10 (1-
0.10) /100
p = total defectives
total sample observations = 0.010

= 200 =
20 (100) 0.10
 0.2
0.18
0.16
0.14
Proportion defective

0.12
0.1
0.08
0.06
0.04
0.02
0
0

18
10

12

14

16

20
..

Sample number
 c - Charts
 Count the number of defects in an item

 Based on the Poisson distribution

c = number of defects in an item
c = total number of defects
number of samples

UCL = c + 3 c

LCL = c - 3 c
 c-Chart Example
The Ritz Hotel has 240 rooms. The hotel’s housekeeping
department is responsible for maintaining the quality of
the room’s appearance and cleanliness. Each
individual housekeeper is responsible for an area
encompassing 20 rooms. Every room in use is
thoroughly clean and its supplies, toiletries, and so on
are restocked each day. Any defects that the
housekeeping staff notice that are not part the normal
housekeeping service are supposed to be reported
hotel maintenance.
 c-Chart Example
Every room is briefly inspected each day by a
housekeeping supervisor. However, hotel
management also conducts inspection for quality-
control purposes. The management inspector not only
check for normal housekeeping defects like clean
sheets, dust, room supplies, room literature, or towels,
but also for defects like an inoperative or missing TV
remote, poor TV picture quality or reception, defective
lamps, a malfunctioning clock, tears or stains in
bedcovers or curtain, or a malfunctioning curtain pull.
 c-Chart Example
An inspection sample include 12 rooms, i.e., one
room selected at random from each of the twelve
20-room blocks served by a housekeeper.
Following are the results from 15 inspection
samples conducted at random during a 1-month
period.
Sample # Defects Sample # Defects
1 12 11 12
2 8 12 10
3 16 13 14
4 14 14 17
5 10 15 15
6 11
7 9
8 14
9 13
10 15
 c - Chart Calculations
Count # of defects per roll in 15 rolls of denim fabric

Sample Defects
1 12 c = 190/15 = 12.67
2 8
3 16
UCL = c + 3 c
. .
= 12.67 + 3 12.67
. . = 23.35
15 15
190 LCL = c - 3 c
= 12.67 - 3 12.67
= 1.99
 Example c - Chart
24

21

18
Number of defects

.
15

12

0
0

12

14
10

Sample number
 Control Charts For Variables
 Mean chart (X-Bar Chart)
Measures central tendency of a sample
 Range chart (R-Chart)
Measures amount of dispersion in a sample

 Each chart measures the process differently. Both

the process average and process variability must
be in control for the process to be in control.
 Example: Control harts for
Variable Data
The Goliath Tool Company produces slip-ring
bearings, which look like flat doughnut or washer,
they fit around shafts or rods, such as drive shaft
in machinery or motor. In the production process
for a particular slip-ring bearing the employees
has taken 10 samples (during a 10 day period) of
5 slip-ring bearing (i.e., n = 5). The individual
observation from each sample are shown as
followed:
Example: Control Charts for Variable Data
Slip Ring Diameter (cm)
Sample 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
Constructing an Range Chart
UCLR = D4 R = (2.11) (.115) = 0.24

LCLR = D3 R = (0) (.115) = 0

where R = R / k = 1.15 / 10 = .115

k = number of samples = 10
R = range = (largest - smallest)
 Example R-Chart
0.25
UCL
0.2

0.15
Range

R
0.1

0.05

0 LCL
1 2 3 4 5 6 7 8 9 10
Sample number
 Constructing A Mean Chart
UCLX = X + A2 R = 5.01 + (0.58) (.115) = 5.08

LCLX = X - A2 R = 5.01 - (0.58) (.115) = 4.94

where X = average of sample means = X / n

= 50.09 / 10 = 5.01
R = average range = R / k = 1.15 / 10 = .115
 Example X-bar Chart
5.10
5.08
UCL
5.06
Sample average

5.04
5.02
5.00 X
4.98
4.96
4.94 LCL
4.92

10
1

9
Sample number
 Variation
 Common Causes
Variation inherent in a process
Can be eliminated only through improvements
in the system

 Special Causes
Variation due to identifiable factors
Can be modified through operator or
management action
 Control Chart Patterns
UCL UCL

LCL LCL

Sample observations Sample observations

consistently below the consistently above the
center line center line
 Control Chart Patterns

UCL UCL

LCL LCL

Sample observations Sample observations

consistently increasing consistently decreasing
Sample Size Determination
 Attribute control charts
50 to 100 parts in a sample

 Variable control charts

2 to 10 parts in a sample