11

Control Charts

C
ontrol charts are decision-making tools that provide information for timely
decisions concerning recently produced products. Control charts are also
problem-solving tools that help locate and investigate the causes of poor or
marginal quality.
Control charts contain a centerline—usually the mathematical average of the samples
plotted—upper and lower statistical control limits that define the constraints of common
cause variation, and performance data plotted over time.

CONTROL CHART TYPES AND SELECTION

There are two general classifications of control charts: variables charts and attributes
charts (see Table 11.1). Variables are things that can be measured: length, temperature,
pressure, weight, and so on. Attributes are things that are counted: dents, scratches,
defects, days, cycles, yes/no decisions, and so on.

Table 11.1 Variables and attributes control charts selection.

Variables control charts
Type Distribution Sample Application
X and R Normal 2 ≤ 10 Measurement subgroups
X and s Normal > 10 Measurement subgroups
Attributes control charts
Type Distribution Sample Application
c Poisson Constant Count number of defects per item
u Poisson Varies Count number of defects per item
np Binomial Constant Count of defective items
p Binomial Varies Count of defective items
g Binomial Individual Interval between rare events
Variables or attributes control charts
Type Distribution Sample Application
X and mR Normal 1 Individual counts or measurements

117
118 Chapter Eleven

CONTROL CHART INTERPRETATION

A process is said to be in control when the control chart does not indicate any out-of-
control condition and contains only common causes of variation. If the common cause
variation is small, then a control chart can be used to monitor the process. See Figure
11.1 for a representation of stable (in control) and unstable (out of control) processes. If
the common cause variation is too large, the process will need to be modified.
When a control chart indicates an out-of-control condition (a point outside the control
limits or matching one or more of the criteria below), the assignable causes of variation
must be identified and eliminated.
Improper control chart interpretation can lead to several problems, including blaming
people for problems they cannot control, spending time and money looking for problems
that do not exist, spending time and money on process adjustments or new equipment
that are not necessary, taking action where no action is warranted, and asking for worker-
related improvements where process or equipment improvements need to be made first.
The following rules should be used to properly indicate that a process is out of control:
  2QHSRLQWEH\RQGWKHƱFRQWUROOLPLW
2. Eight or more points on one side of the centerline without crossing
3. Four out of five points in zone B or beyond
4. Six points or more in a row steadily increasing or decreasing
5. Two out of three points in zone A
6. Fourteen points in a row alternating up and down
7. Any noticeable/predictable pattern, cycle, or trend
Please note that depending on the source, these rules can vary. See Figure 11.2 for control
chart interpretation rules.

Stable process Unstable process

e

e
m

m
Ti

Ti

Figure 11.1 Stable and unstable variation.

Source: Adapted from Innovative Control Charting (Wise and Fair 1997). Used with permission.
 Rule 1 Rule 3 Rule 4 Rule 7
One point beyond Four out of five Six points or more in a Any noticeable/predictable
the control limit points in zone B row steadily increasing pattern, cycle, or trend
or beyond or decreasing

Upper control limit

+3σ
Zone A

+2σ
Zone B

+1σ
Zone C
Centerline

95%
68%

99.7%
Zone C

–1σ
Zone B

–2σ
Zone A

–3σ
Lower control limit
Rule 5 Rule 2 Rule 6
Two out of three Eight or more points on Fourteen points in a row
points in zone A one side of the centerline alternating up and down
without crossing

Figure 11.2 Control chart interpretation rules.

Source: Adapted from Innovative Control Charting (Wise and Fair 1997). Used with permission.
Control Charts
119
120 Chapter Eleven

Target
value

Accuracy

Precision

Figure 11.3 Control chart accuracy and precision.

Accurate Accurate
and but not
precise precise

Precise Neither
but not accurate
accurate or precise

Figure 11.4 Accuracy versus precision.

The target value (which is hopefully the control chart centerline) is closely related to
process accuracy. The range chart is closely associated with process precision (spread or
dispersion) (see Figures 11.3 and 11.4).
 Control Charts 121

X-BAR AND R CONTROL CHARTS

X-bar and R control charts assume a normal distribution and are usually used with a
subgroup size of less than 10 (typically 3–5). A minimum of 25 subgroups is necessary
to construct the chart.

X chart R chart
ΣX
X= (Subgroup) R = Xn – Xi (Subgroup)
n
ΣX ΣR
X= (Centerline) R= (Centerline)
k k

X UCL = X + A2 * R R UCL  D4 * R

X LCL = X − A2 * R R LCL  D3 * R

R
s (Estimate of sigma)
d2

where
n = Subgroup size
k = Number of subgroups
(See Appendix O, “Control Chart Constants.”)
Example: A process has an X-bar of 5.496 and an R-bar of 0.065. Twenty-five subgroups
of three samples are taken (Table 11.2). Calculate the centerlines and upper and lower
control limits, and construct an X-bar and R chart (Figure 11.5).

Σ X 137.390
X= = = 5.496
k 25
Σ R 1.620
R= = = 0.065
k 25

X UCL = X + A2 * R = 5.496 + 1.023 * 0.065 = 5.535

X UCL = X + A2 * R = 5.496 − 1.023 * 0.065 = 5.430

R UCL = D4 * R = 2.574 * 0.065 = 0.167

R LCL = D3 * R = 0 * 0.065 = 0
122 Chapter Eleven

Table 11.2 Data for X and R chart.

Subgroup X R
1 5.447 0.080
2 5.487 0.100
3 5.450 0.060
4 5.450 0.060
5 5.477 0.100
6 5.510 0.020
7 5.510 0.030
8 5.463 0.080
9 5.553 0.040
10 5.510 0.110
11 5.627 0.030
12 5.610 0.110
13 5.507 0.090
14 5.497 0.070
15 5.540 0.080
16 5.413 0.060
17 5.490 0.020
18 5.490 0.060
19 5.467 0.060
20 5.467 0.060
21 5.500 0.070
22 5.477 0.070
23 5.490 0.030
24 5.470 0.080
25 5.490 0.050
Sum 137.390 1.620
Average 5.496 0.065
 Control Charts 123

X chart
5.650

5.600 UCL

5.550

5.500 X

5.450
LCL

5.400
1 4 7 10 13 16 19 22 25

R chart
0.180
UCL
0.160

0.140

0.120

0.100

0.080

0.060 R

0.040

0.020

0.000
1 4 7 10 13 16 19 22 25

Figure 11.5 X and R chart example.

To help determine the subgroup sample size, the following formulas can be used.
For a one-sided specification:
2
n = z1− α + z1−β 2 l
δ

For a two-sided specification:

2
n = z1− α + z1−β 2 l
2 δ
124 Chapter Eleven

where
Z = Normal distribution value (See Appendix B, “Normal Distribution
Probability Points—Area Above Z”)
Ơ &RQILGHQFH
ơ 5HOLDELOLW\
ƣ 'LIIHUHQFHRUVKLIWZHZDQWWRGHWHFW

Example: :HZDQWWRGHWHFWDƱVKLIWLQDSURFHVVWKDWKDVDRQHVLGHGVSHFLILFDWLRQ
+RZPDQ\FRQVHFXWLYHVDPSOHVVKRXOGEHVHOHFWHGZLWKDQƠRIDQGDơRIIRU
each subgroup?
2 2
n = z1− α + z1−β l l
2
n = (2.326 + 2.576) 2
= 3.80 ≈ 4
δ l5

Example: :HZDQWWRGHWHFWDƱVKLIWLQDSURFHVVWKDWKDVDWZRVLGHGVSHFLILFDWLRQ
+RZPDQ\FRQVHFXWLYHVDPSOHVVKRXOGEHVHOHFWHGZLWKDQƠRIDQGDơRIIRU
each subgroup?
2 2
n = z(1− α)/2 + z1−β l l
2
= n = (1.96 + 1.65) 2
= 5.77 ≈ 6
δ l5

X-BAR AND s CONTROL CHARTS

The X-bar and s control charts assume a normal distribution and are usually used with a
subgroup size of greater than 10. A minimum of 25 subgroups is necessary to construct
the chart.

X chart s chart

ΣX Σ( X − X )2
X= (Subgroup) s= (Subgroup)
n n −1
ΣX Σs
X= (Centerline) s= (Centerline)
k k

X UCL = X + A3 * s s UCL  B4 * s

X LCL = X − A3 * s s LCL  B3 * s

s
s (Estimate of sigma)
c4
 Control Charts 125

where
n = Subgroup size
k = Number of subgroups
(See Appendix O, “Control Chart Constants.”)
Example: A process has an X-bar of 5.498 and an s-bar of 0.046. Twenty-five subgroups
of 11 samples are taken (Table 11.3). Calculate the centerlines and the upper and lower
control limits, and construct an X-bar and s chart (Figure 11.6).

Σ X 137.488
X= = = 5.498
k 25
Σ s 1.155
s= = = 0.046
k 25
X UCL = X + A3 * s = 5.498 + 0.927 * 0.046 = 5.541

X LCL = X − A3 * s = 5.498 − 0.927 * 0.046 = 5.455

s UCL = B4 * s = 1.679 * 0.046 = 0.077

s LCL = B3 * s = 0.321 * 0.046 = 0.015
126 Chapter Eleven

Table 11.3 Data for X and s chart.

Subgroup X s
1 5.449 0.032
2 5.487 0.039
3 5.456 0.037
4 5.612 0.121
5 5.479 0.033
6 5.535 0.106
7 5.516 0.016
8 5.465 0.045
9 5.551 0.036
10 5.520 0.031
11 5.489 0.030
12 5.474 0.026
13 5.481 0.035
14 5.480 0.031
15 5.537 0.030
16 5.439 0.047
17 5.505 0.098
18 5.485 0.080
19 5.475 0.071
20 5.590 0.057
21 5.487 0.040
22 5.489 0.039
23 5.492 0.021
24 5.475 0.025
25 5.479 0.029
Sum 137.448 1.155
Average 5.498 0.046
 Control Charts 127

X chart
5.650

5.600

5.550
UCL

5.500 X

5.450 LCL

5.400
1 4 7 10 13 16 19 22 25

s chart
0.140

0.120

0.100

0.080 UCL

0.060
s
0.040

0.020 LCL

0.000
1 4 7 10 13 16 19 22 25

Figure 11.6 X and s chart example.

128 Chapter Eleven

c-CHARTS
c-charts assume a Poisson distribution and are usually used with a constant sample
size, counting the number of defects per item. A minimum of 25 subgroups is necessary
to construct the chart.

c = (Subgroup count)
Σc
c= (Centerline)
k
UCL = c + 3* c

LCL = c − 3* c

(A calculated LCL of less than zero reverts to zero.)

c
(Estimate of sigma for subgroup)

where
k = Number of subgroups
Example: A process is evaluated using a constant sample size of 48 items. When the
items are inspected, a count of the number of defects is recorded (Table 11.4). Twenty-
five subgroups of 48 samples are taken. Calculate the centerline and the upper and lower
control limits, and construct a c-chart (Figure 11.7).

Σ c 136
c= = = 5.440
k 25
UCL = c + 3* c = 5.440 + 3* 5.440 = 12.437

LCL = c − 3* c = 5.440 − 3* 5.440 = − 1.557 reverts to 0

 Control Charts 129

Table 11.4 Data for c-chart.

Subgroup c
1 6
2 7
3 5
4 2
5 11
6 7
7 7
8 0
9 1
10 4
11 15
12 5
13 0
14 6
15 3
16 7
17 6
18 8
19 1
20 7
21 1
22 7
23 11
24 3
25 6
Sum 136
Average 5.440
130 Chapter Eleven

c-chart
16

14

12 UCL

10

6
c
4

0
1 4 7 10 13 16 19 22 25

Figure 11.7 c-chart example.

u-CHARTS
u-charts assume a Poisson distribution and are usually used with a variable sample size
(although a constant sample size may be used), counting the number of defects per item.
A minimum of 25 subgroups is necessary to construct the chart.

c
u (Subgroup)
n
Σc
u= (Centerline)
Σn

(Floating control limits) (Static control limits)

u u
UCL = u + 3* UCL = u + 3*
n Σn / k

u u
LCL = u − 3* LCL = u − 3*
n Σn / k
(A calculated LCL of less than zero reverts to zero.)
u u
s s≅
n Σn / k
(Estimate of sigma for subgroup) (Estimate of sigma for chart)
 Control Charts 131

where
n = Subgroup size
k = Number of subgroups
Example: A process is evaluated using a variable sample size. When the items are
inspected, a count of the number of defects is recorded. Twenty-five subgroups are
evaluated (Table 11.5). Calculate the centerline and the upper and lower control limits,
and construct a u-chart (Figure 11.8).

Table 11.5 Data for u-chart.

Subgroup n c u UCL LCL
1 113 77 0.681 1.232 –1.977
2 85 72 0.847 1.274 –1.118
3 99 125 1.263 1.251 –0.737
4 118 121 1.025 1.226 –0.511
5 111 80 0.721 1.235 –0.356
6 59 79 1.339 1.338 –0.241
7 123 117 0.951 1.221 –0.153
8 101 80 0.792 1.248 –0.081
9 105 128 1.219 1.243 –0.022
10 118 111 0.941 1.226 0.029
11 91 66 0.725 1.264 0.072
12 74 103 1.392 1.297 0.109
13 98 99 1.010 1.253 0.143
14 106 110 1.038 1.241 0.172
15 116 75 0.647 1.229 0.199
16 88 123 1.398 1.269 0.223
17 104 43 0.413 1.244 0.245
18 45 77 1.711 1.394 0.265
19 100 91 0.910 1.250 0.283
20 95 114 1.200 1.257 0.300
21 103 102 0.990 1.245 0.316
22 118 76 0.644 1.226 0.331
23 102 71 0.696 1.247 0.345
24 60 90 1.500 1.335 0.357
25 92 88 0.957 1.262 0.370
Sum 2424 2318
132 Chapter Eleven

u-chart
1.800

1.600

1.400

1.200 UCL

1.000
u
0.800
LCL
0.600

0.400

0.200
1 4 7 10 13 16 19 22 25

Figure 11.8 u-chart example.

Σ c 2318
u= = = 0.956
Σ n 2424
u
UCL = u + 3*
n

u
LCL = u − 3*
n

The control limits will vary for each subgroup because the sample size n varies from
group to group.
 Control Charts 133

np-CHARTS
np-charts assume a binomial distribution and are usually used with a constant sample
size, counting the defective items. A minimum of 25 subgroups is necessary to construct
the chart.
np = (Subgroup count)
Σ np
np = (Centerline)
k
UCL = np + 3* np(1 − ( Σ np / Σ n))

LCL = np − 3* np(1 − ( Σ np / Σ n))

(A calculated LCL of less than zero reverts to zero.)
s ≅ np(1 − ( Σ np / Σ n)) (Estimate of sigma)

where
n = Subgroup size
k = The number of subgroups
Example: A process is evaluated using a constant sample size of 200 items. When the
items are inspected, a count of the number of defective items is recorded. Twenty-five
subgroups of 200 samples are taken (Table 11.6). Calculate the centerline and the upper
and lower control limits, and construct an np-chart (Figure 11.9).

Σ np 100
np = = = 4.000
k 25
UCL = np + 3* np(1 − ( Σ np / Σ n)) = 4.000 + 3* 4.000(1 − (100 / 5000)) = 9.940

LCL = np − 3* np(1 − ( Σ np / Σ n)) = 4.000 − 3* 4.000(1 − (100 / 5000)) = − 1.940 ,

which reverts to zero.

134 Chapter Eleven

Table 11.6 Data for np-chart.

Subgroup n np
1 200 2
2 200 5
3 200 3
4 200 5
5 200 11
6 200 1
7 200 5
8 200 6
9 200 2
10 200 0
11 200 5
12 200 8
13 200 4
14 200 2
15 200 6
16 200 3
17 200 4
18 200 6
19 200 4
20 200 1
21 200 1
22 200 2
23 200 5
24 200 0
25 200 9
Sum 5000 100
 Control Charts 135

np-chart
12

10 UCL

4 np

0
1 4 7 10 13 16 19 22 25

Figure 11.9 np-chart example.

p-CHARTS
p-charts assume a binomial distribution and are usually used with a variable sample size
(although a constant sample size may be used), counting the defective items. A minimum
of 25 subgroups is necessary to construct the chart.

np
p (Subgroup)
n
Σ np
p= (Centerline)
Σn

(Floating control limits) (Static control limits)

UCL = p + 3* ( p(1 − p)) / n) UCL = p + 3* ( p(1 − p)) / ( Σ n / k )

LCL = p − 3* ( p(1 − p)) / n) LCL = p − 3* ( p(1 − p)) / ( Σ n / k )

(A calculated LCL of less than zero reverts to zero.)
s ≅ ( p(1 − p)) / n) s ≅ ( p(1 − p)) / ( Σ n / k )
(Estimate of sigma for subgroup) (Estimate of sigma for chart)
136 Chapter Eleven

where
n = Subgroup size
k = Number of subgroups
Example: A process is evaluated using a variable sample size. When the items are
inspected, a count of the number of defective items is recorded. Twenty-five subgroups
are evaluated (Table 11.7). Calculate the centerline and the upper and lower control
limits, and construct a p-chart (Figure 11.10).

Table 11.7 Data for p-chart.

Subgroup n np p UCL LCL
1 2108 254 0.120
2 1175 117 0.100 0.142 0.203
3 1658 248 0.150 0.138 0.175
4 2173 239 0.110 0.135 0.176
5 1720 238 0.138 0.138 0.176
6 1891 250 0.132 0.136 0.175
7 1664 145 0.087 0.138 0.194
8 1685 127 0.075 0.138 0.199
9 1967 327 0.166 0.136 0.167
10 976 95 0.097 0.145 0.212
11 2012 201 0.100 0.136 0.182
12 1187 124 0.104 0.142 0.200
13 1784 173 0.097 0.137 0.187
14 1390 132 0.095 0.140 0.198
15 2075 229 0.110 0.135 0.178
16 2077 236 0.114 0.135 0.177
17 1232 126 0.102 0.142 0.200
18 1914 228 0.119 0.136 0.178
19 2401 315 0.131 0.134 0.168
20 1975 255 0.129 0.136 0.174
21 1365 132 0.097 0.140 0.198
22 1505 117 0.078 0.139 0.203
23 1725 129 0.075 0.137 0.199
24 2105 315 0.150 0.135 0.168
25 1855 242 0.130 0.137 0.176
Sum 43619 4994
 Control Charts 137

p-chart
0.170

0.150

UCL
0.130

p
0.110

0.090 LCL

0.070

0.050
1 4 7 10 13 16 19 22 25

Figure 11.10 p-chart example.

Σ np 4994
p= = = 0.114
Σ n 43619
UCL = p + 3* ( p(1 − p)) / n)

LCL = p − 3* ( p(1 − p)) / n)

The control limits will vary for each subgroup because the sample size n varies from
group to group.

g-CHARTS
g-charts assume a binomial distribution and are usually used to chart the interval between
rare events. A minimum of 25 observations is necessary to construct the chart.

g = Interval between rare events

Σg
g= (Centerline)
k
UCL = g + 3* g * ( g + 1)

LCL = g − 3* g * ( g + 1)

s ≅ g * ( g + 1)
(Estimate of sigma for chart)
138 Chapter Eleven

where
k = The number of subgroups
Example: A company wishes to track the days between injuries to evaluate the presence
of a pattern or trend. Twenty-five subgroups are evaluated (Table 11.8). Calculate the
centerline and the upper and lower control limits, and construct a g-chart (Figure 11.11).

Table 11.8 Data for g-chart.

Subgroup Days
1 0
2 17
3 15
4 12
5 29
6 17
7 17
8 10
9 11
10 14
11 25
12 15
13 47
14 16
15 9
16 22
17 16
18 7
19 11
20 29
21 11
22 17
23 37
24 13
25 16
Sum 433
 Control Charts 139

g-chart
80

70 UCL

60

50

40

30

20
g
10

0
12

12

12

12

12

12

12

12

12

11 012

/1 2
1/ 12

2/ 13

3/ 13
13
12 01
20

20

20

20

20

20

20

20

20

0
20

20

20
/2

/2

/2
1/

1/

1/

1/

1/

1/

1/

1/

1/

1/

1/

1/
/1

/1
1/

2/

3/

4/

5/

6/

7/

8/

9/
10
Figure 11.11 g-chart example.

Σ g 433
g= = = 17.320
k 25
UCL = g + 3* g * ( g + 1) = 17.320 + 3* 17.320 * (17.320 + 1) = 70.759

LCL = g − 3* g * ( g + 1) = 17.320 − 3* 17.320 * (17.320 + 1) = − 36.119 ,

which reverts to zero.

140 Chapter Eleven

X AND mR (MOVING RANGE) CONTROL CHARTS

XDQGP5 PRYLQJUDQJH FRQWUROFKDUWVDVVXPHDQRUPDOGLVWULEXWLRQDQGDUHXVHGZLWK
individual values. A minimum of 25 observations is necessary to construct the chart.

X chart mR chart
X (Individual observation) mR = X 2 − X1 , X 3 − X 2 ...

ΣX Σ mR
X= (Centerline) mR = (Centerline)
k k −1

UCL = X + d 2 * mR UCL  D4 * mR

LCL = X − d 2 * mR LCL = 0

mR
s (Estimate of sigma)
d2

(See Appendix O, “Control Chart Constants.”)

where
k = Number of subgroups
Example: A manager wants to evaluate a low-volume production process. Due to the
cost of testing, one sample is measured per shift (Table 11.9). Calculate the centerline
and the upper and lower control limits, and construct an XDQGP5FKDUW )LJXUH 

Σ X 174.700
X= = = 6.988
k 25
Σ mR 7.200
mR = = = 0.300
k − 1 25 − 1
UCL = X + d 2 * mR = 6.988 + 1.128 * 0.300 = 7.326

LCL = X − d 2 * mR = 6.988 − 1.128 * 0.300 = 6.650

UCL = D4 * mR = 3.267 * 0.300 = 0.980

LCL = 0
 Control Charts 141

Table 11.9 Data for X and mR chart.

Subgroup X mR
1 7.2 0.000
2 7.5 0.300
3 7.1 0.400
4 6.9 0.200
5 6.8 0.100
6 6.6 0.200
7 6.3 0.300
8 6.9 0.600
9 7.2 0.300
10 7.1 0.100
11 6.9 0.200
12 7.2 0.300
13 7.4 0.200
14 7.5 0.100
15 7.4 0.100
16 7.3 0.100
17 7.7 0.400
18 6.7 1.000
19 7.2 0.500
20 6.8 0.400
21 6.6 0.200
22 6.7 0.100
23 6.2 0.500
24 6.7 0.500
25 6.8 0.100
Sum 174.700 7.200
142 Chapter Eleven

X chart
7.800

7.600

7.400
UCL
7.200

7.000 X
6.800

6.600 LCL

6.400

6.200

6.000
1 4 7 10 13 16 19 22 25

mR chart
1.200

1.000
UCL
0.800

0.600

0.400
mR
0.200

0.000
1 4 7 10 13 16 19 22 25

Figure 11.12 X and mR chart example.

 Control Charts 143

PRE-CONTROL CHARTS
Pre-control charts, sometimes referred to as stoplight charts, are distribution free and
are used to record individual values during production startup before there are enough
data to use statistical control charting methods. Pre-control charts divide the part/
process specification into three zones—red, green, and yellow—and plot a data point.
Depending on which zone the latest data point falls in, the process continues as is, is
further monitored, or stopped and adjusted. Figure 11.13 demonstrates the relationship
between the zones and part print tolerance.

X = (Centerline/nominal value)
UTL = X + USL
LTL = X – LSL

USL
UPCL = X +
2
LSL
LPCL = X −
2

LTL LPCL UPCL UTL

Print limit

Print limit

7% 86% 7%

1/2 print
tolerance
Red Yellow Green Yellow Red
zone zone zone zone zone
Print tolerance

Figure 11.13 Assumptions for underlying pre-control.

144 Chapter Eleven

where
X = Target value
UTL = Upper tolerance limit
LTL = Lower tolerance limit
USL = Allowed positive deviation from nominal
LSL = Allowed negative deviation from nominal
UPCL = Upper pre-control limit
LPCL = Lower pre-control limit
Interpretation rules:
• If five parts in a row fall between LPCL–UPCL (green zone), run the process.
• If one part falls between the LTL-LPCL or UPCL-UTL (yellow zone), restart
the count.
• If two parts in a row fall between the LTL-LPCL or UPCL-UTL (yellow zone),
adjust the process and restart the count.
• If one part exceeds the LTL or UTL (red zone), adjust the process and restart
the count.
• If 25 parts in a row fall between the LPCL–UPCL (green zone), run the
process but at a reduced sampling frequency.
Example: A machining process has a nominal dimension of 1.000 inch ± 0.010 inch.
Twenty-five subgroups are evaluated (Table 11.10). Construct a pre-control chart to
monitor the initial production run (Figure 11.14).

UTL = X + USL = 1.000 + 0.010 = 1.010

LTL = X – LSL = 1.000 – 0.010 = 0.990

USL 0.010
UPCL = X + = 1.000 + = 1.005
2 2
LSL 0.010
LPCL = X − = 1.000 − = 0.995
2 2
 Control Charts 145

Table 11.10 Data for pre-control chart.

Subgroup X
1 1.003
2 1.001
3 1.004
4 1.004
5 0.993
6 1.002
7 0.999
8 0.999
9 1.013
10 0.997
11 1.002
12 1.003
13 1.002
14 1.002
15 1.000
16 1.003
17 0.994
18 0.988
19 0.994
20 0.998
21 1.001
22 0.996
23 1.002
24 1.000
25 0.997
146 Chapter Eleven

Pre-control chart
1.015
Red zone
1.010 UTL
Yellow zone
1.005 UPCL
Green zone
1.000 Target
Green zone
0.995 LPCL
Yellow zone
0.990 LTL
Red zone
0.985
1 4 7 10 13 16 19 22 25

Figure 11.14 Pre-control chart example.