Control Chart Concepts - Exercises

1. In the manufacturing of corn flakes, moisture content is a critical variable. Suppose that when the
manufacturing process is in control, the moisture content of flakes just after milling is normally distributed
with mean .11 and variance .0016. Find 3-sigma control limits and the centerline for an x̄-chart which
uses samples of 36 flakes.1
2. A sample of flakes yields an average moisture content of .132. Should the process be judged out of control?2
3. A sample of flakes yields an average moisture content of .112. Should the process be judged out of control?3
4. A sequence of samples taken hourly yields the following moisture content averages:
Time 1 2 3 4 5 6 7
x̄ .115 .105 .125 .126 .131 .129 .128
Plot the x̄-chart. Identify any out of control points in the above sequence.4
5. Suppose the process mean shifts to .10 at time 11. What is the probability that the control chart will
detect this shift?5
6. Suppose the process mean shifts to .125 at time 15. What is the probability that the control chart will
detect this shift?6
7. Suppose the process mean shifts to .095 at time 25. What is the probability that the control chart will
detect this shift?7
8. Samples of size 4 are taken from a process daily. The relevant measurements should be normally distributed
with a mean of 15 and a variance of 1.44 when the process is in control. Find the centerline and control
limits for an appropriate 3-sigma x̄-chart.8
9. The following measurements were taken from the process on a particular day: 18, 22, 13, 7. Find the
average, and use the control chart constructed in the previous question to decide if the process is in
control.9

10. The following measurements were taken from the process on another day: 14, 10, 9, 7. Find the average,
and use the control chart constructed in the previous question to decide if the process is in control.10

11. If the mean for the above process shifts to 17 on a particular day, find the probability that this shift will
be detected by the control chart on that day.11

12. The clockspeed of a certain component is required to be within 29.5 MHz and 36.5 MHz. An x̄-chart has
been set up to control the production of these components. Subgroups of size 16 were used to set up the
3σ control limits: 31.0 and 34.0 MHz. Compute the ARL until a process mean shift to 30.5 MHz will be
detected.12
13. The moisture content of bran flakes has been monitored with the use of an x̄-chart which is based on
hourly random samples of 4 units (A unit is 10 g flakes). The control limits are 7 .20 and 5.20. How many
samples must be taken on average, until a mean shift of size 0.5 is detected by this control chart?13
1k
√
= 3, σ = .0016 = .04, µ = .11. Therefore, the centerline is .11, UCL = .11 + 3(.04)/6 = .13, LCL = .09.
2 Yes, since x̄ >UCL.
3 No, since x̄ lies between the control limits.
4 The process is judged out of control at time 5.
5 Since σ/√n = .0067, the probability of an out of control point occurring is P (X̄ < .09) + P (X̄ > .13) = P (Z < (.09 −

.1)/.0067) + P (Z > (.13 − .1)/.0067) = P (Z < −1.49) + P (Z > 4.47) = .068.

6 P (X̄ < .09) + P (X̄ > .13) = P (Z < (.09 − .125)/.0067) + P (Z > (.13 − .125)/.0067) = P (Z < −5.22) + P (Z > .746) = .2236.
7 P (X̄ < .09) + P (X̄ > .13) = P (Z < (.09 − .095)/.0067) + P (Z > (.13 − .095)/.0067) = P (Z < −.746) + P (Z > 5.22) = .2236.
8 centerline=15, UCL= 15 + 3(1.2)/2 = 16.8, LCL = 13.2.
9 x̄ = 15. In control.
10 x̄ = 10 < 13.2. Out of control.
11 σ/√n = .6. The probability of detecting the shift is then P (X̄ > 16.8) + P (X̄ < 13.2) = P (Z > −.2/.6) + P (Z < −3.8/.6) =

.6293.
12 σ̂ = 2. Therefore, p = P (Z < 1) + P (Z > 7) = .8413; ARL= 1.188.
d
13 σ̂ = 2/3. Therefore, p = P (Z < −4.5) + P (Z > 1.5) = .067, ARL= 14.92.
d

1
 14. Refer to the previous question. What subgroup size is required to reduce the ARL to 10 for detecting a
mean shift to 5.70, while maintaining an in-control ARL of 370?14
15. An x̄-chart with UCL = 14.2 and LCL = 8.2 is used to monitor a process. If subgroups of size 25 are
used, compute ARL before a process mean shift to 12.2 is detected. 15
16. Suppose a process has a mean of 3 and a variance of 1.44 when in control, and we require an ARL of 50
for detecting shifts in the mean of 0.3 while the in-control ARL should be 370. Find the subgroup size.16
17. Suppose we require an ARL of 10 for a mean shift of 2.5 for a process whose variance is 25. What is n?17
18. A p-chart with UCL= .27 and LCL= .03 is used on subgroups of size 80. How many subgroups must be
inspected, on average, to detect a shift to p = .18?18
19. Refer to the previous question. What subgroup size is required to reduce the ARL to 10 for detecting a
shift to p = .18, while maintaining the in-control ARL at 370?19
20. A p-chart with centerline p = .1 and sample size 100 is used to control a process. If the process shifts to
p = .2, find the ARL.20
21. A p-chart with UCL = 0.12, LCL = 0.02 and sample size 235 is used to control a process. How many
subgroups will be inspected, on average, before a process shift to p = .08 will be detected?21
22. Refer to the previous question. What sampling frequency is necessary in order to ensure that a process
shift to p = .08 will be detected within 24 hours, on average?22
23. A p-chart with UCL = 0.15 and LCL = 0.01 is currently used to monitor the production of ski masks.
If an average run length of 20 is required to detect a shift to p = .1, what subgroup size should be used,
while maintaining an in-control ARL of 370?23
24. The following points have been plotted on an x̄-chart having centerline 11.2.
8.6 10.3 12.1 11.5 9.2 10.6 11.8 12.9 13.1 13.4 9.8 10.6 13.3 10.1
Is the process in control? (Apply the sensitizing rules.)24
25. The following points have been plotted on an x̄-chart having UCL = 13.5 and LCL = 1.5.
6.7 7.8 9.3 6.9 6.1 8.3 9.6 7.7 4.7 8.2 3.3 2.5 5.1 7.6
Is the process in control? (Apply the sensitizing rules.)25
26. The following points have been plotted on an x̄-chart having UCL = 14.2 and LCL = 8.2.
11.8 10.3 11.1 13.2 10.3 9.7 8.6 11.5 13.1 9.6 11.3 8.9 12.1 14.0
26
Is the process in control? (Apply the sensitizing rules.)
27. The following points have been plotted on an x̄-chart having UCL = 22.8 and LCL = 10.8.
17.2 15.3 18.4 19.6 15.8 13.7 11.1 14.7 13.8 17.4 21.1 17.6 19.2 22.4
27
Is the process in control? (Apply the sensitizing rules.)

14 p = .1. Therefore,n = 6.
d
15 σ̂ = 5. Therefore, p = P (Z > 2) + P (Z < −4) = .0228. n = 43.9
d
16 p = .02. Therefore, n= 15.
d
17 p = .1. Therefore, n= 12.
d
18 p = P (Z > 2.09) + P (Z < −3.49) = .0183; ARL = 54.6.
d
19 p = .1. Therefore, n = 374.
d
20 UCL = .19, LCL = .01. Therefore, p = P (Z > −0.25) + P (Z < −4.75) = .5987; ARL = 1.67.
d
21 p = P (Z > 2.26) + P (Z < −3.39) = .01225; ARL = 81.63.
d
22 1 sample every 17.64 minutes. Therefore, if 4 samples are taken per hour, the out-of-control condition will be detected in less
than 24 hours, on average.
23 p = .05. Therefore, n = 257.
d
24 No; 6 successive points increase.
25 No; 2 of 3 consecutive points outside 2σ warning limits.
26 Yes.
27 No; 4 of 5 consecutive points beyond 1σ limits.