IE4230 Homework 5

17/02/2014
1. The fill volume of soft-drink beverage bottles is an important quality
characteristic. The volume is measured (approximately) by placing a
gauge over the crown and comparing the height of the liquid in the neck
of the bottle against a coded scale. On this scale, a reading of zero
corresponds to the correct fill height. Fifteen samples of size n = 10 have
been analyzed, and the fill heights are shown in the following table.
a) Set up 𝒙 and s control charts on this process. Does the process exhibit
statistical control? If necessary, construct revised control limits.
b) Set up an R chart, and compare with the s chart in part (a).
c) Set up an s2 chart and compare with the s chart in part (a).
• Control limits for the 𝑥 chart Control limits for the 𝑠 chart
UCL: 𝑥 + 𝐴3 𝑠 UCL: 𝐵4 𝑠
CL: 𝑥 CL: 𝑠
LCL: 𝑥 − 𝐴3 𝑠 LCL: 𝐵3 𝑠

Sample
𝒙𝟏 𝒙𝟐 𝒙𝟑 𝒙𝟒 𝒙𝟓 𝒙𝟔 𝒙𝟕 𝒙𝟖 𝒙𝟗 𝒙𝟏𝟎 𝒙 𝒔 𝑹 𝒔𝟐
Number
1 2.5 0.5 2 -1 1 -1 0.5 1.5 0.5 -1.5 0.5 1.3333 4 1.7778
2 0 0 0.5 1 1.5 1 -1 1 1.5 -1 0.45 0.9265 2.5 0.8583
3 1.5 1 1 -1 0 -1.5 -1 -1 1 -1 -0.1 1.1255 3 1.2667
4 0 0.5 -2 0 -1 1.5 -1.5 0 -2 -1.5 -0.6 1.1738 3.5 1.3778
5 0 0 0 -0.5 0.5 1 -0.5 -0.5 0 0 0 0.4714 1.5 0.2222
6 1 -0.5 0 0 0 0.5 -1 1 -2 1 0 0.9718 3 0.9444
7 1 -1 -1 -1 0 1.5 0 1 0 0 0.05 0.8960 2.5 0.8028
8 0 -1.5 -0.5 1.5 0 0 0 -1 0.5 -0.5 -0.15 0.8182 3 0.6694
9 -2 -1.5 1.5 1.5 0 0 0.5 1 0 1 0.2 1.1832 3.5 1.4000
10 -0.5 3.5 0 -1 -1.5 -1.5 -1 -1 1 0.5 -0.15 1.5284 5 2.3361
11 0 1.5 0 0 2 -1.5 0.5 -0.5 2 -1 0.3 1.2065 3.5 1.4556
12 0 -2 -0.5 0 -0.5 2 1.5 0 0.5 -1 0 1.1547 4 1.3333
13 -1 -0.5 -0.5 -1 0 0.5 0.5 -1.5 -1 -1 -0.55 0.6852 2 0.4694
14 0.5 1 -1 -0.5 -2 -1 -1.5 0 1.5 1.5 -0.15 1.2483 3.5 1.5583
15 1 0 1.5 1.5 1 -1 0 1 -2 -1.5 0.15 1.2704 3.5 1.6139
𝒙 𝒔 𝑹 𝒔𝟐
-0.0033 1.0662 3.2 1.2057
𝐴3 = 0.9754, 𝐵3 = 0.2837, 𝐵4 = 1.7163

Control limits for the 𝑥 chart Control limits for the 𝑠 chart
UCL: −0.0033 + 0.9754 × 1.0662 UCL: 1.7163 × 1.0662
= 1.0366 = 1.8299
CL: −0.0033 CL: 1.0622
LCL: −0.0033 − 0.9754 × 1.0662 LCL: 0.2837 × 1.0662
= −1.0433 = 0.3025

xbar chart s chart

1.5 2
1
1.5
0.5
0 1
-0.5
0.5
-1
-1.5 0
1 2 3 4 5 6 7 8 9 101112131415 1 2 3 4 5 6 7 8 9 101112131415
Sample Sample
b) Set up an R chart, and compare with the s chart in part (a).
• Control limits for the 𝑅 chart
UCL: 𝐷4 𝑅
CL: 𝑅
LCL: 𝐷3 𝑅

UCL: 1.7770.× 3.2 = 5.6863

CL: 3.2
LCL: 0.2230 × 3.2 = 0.7137
R chart s chart
6 2
1.5
4
1
2
0.5
0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 101112131415
Sample Sample
c) Set up an s2 chart and compare with the s chart in part (a).
• Control limits for the 𝑠 2 chart
𝑠2 1.2057
UCL: 𝜒𝛼2 2,𝑛−1 UCL: × 27.0933 = 3.6297
𝑛−1 9
CL: 𝑠2 CL: 1.2057
𝑠2 2 1.2057
LCL: 𝜒1−(𝛼 2) ,𝑛−1 LCL: × 1.2412 = 0.1663
𝑛−1 9

Here, 𝛼 = 0.0027.

s2 chart s chart
4 2
3 1.5
2 1
1 0.5
0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 101112131415
Sample Sample
2. A process is controlled with a fraction nonconforming control chart with
three-sigma limits, n = 100, UCL = 0.161, center line = 0.080, and LCL = 0.
a) Find the equivalent control chart for the number nonconforming.
• The fraction nonconforming control chart
𝑝 1−𝑝
UCL: 𝑝 + 3 UCL: 0.161
𝑛
CL: 𝑝 CL: 0.080
𝑝 1−𝑝
LCL: max 0, 𝑝 − 3 LCL: 0
𝑛

The nonconforming control chart

UCL: 𝑛𝑝 + 3 𝑛𝑝 1 − 𝑝 UCL: 16.1
CL: 𝑛𝑝 CL: 8
LCL: max 0, 𝑛𝑝 − 3 𝑛𝑝 1 − 𝑝 LCL: 0
b) Use the Poisson approximation to the binomial to find the probability
of a type I error.
• From (a), the control chart signals if there are more than 16 defects. The
type I error is
ℙ 𝐷 ≥ 17 𝑛 = 100, 𝑝 = 0.08
≅ ℙ 𝐷 ≥ 17 𝜆 = 𝑛𝑝 = 8
𝑑
16 8
=1− 𝑑=0 𝑑! 𝑒 −8 = 0.0037
Excel: =1-POISSON.DIST(16,8,TRUE)

c) Use the correct approximation to find the probability of a type II error

if the process fraction nonconforming shifts to 0.2.
• We will use the normal approximation to the binomial. The parameters
are
𝜇 = 𝑛𝑝 = 100 × 0.2 = 20
𝜎 2 = 𝑛𝑝 1 − 𝑝 = 100 × 0.2 × 0.8 = 16.
The type II error is
ℙ 𝐷 < 16.1 𝑛 = 100, 𝑝 = 0.2
≅ ℙ 𝐷 < 𝟏𝟔. 𝟓 𝜇 = 20, 𝜎 2 = 16
16.5−20
=Φ = Φ −0.875 = 0.1908
4

d) What is the probability of detecting the shift in part (c) by at most the
fourth sample after the shift?
• From (c), the probability of an out-of-control signal is 0.8092. The
requested probability can be computed by the cumulative density
function of a geometric distribution with parameter 𝑝 = 0.8092:
1 − 1 − 0.8092 4 = 1 − 0.19084 = 0.9987.
3. A payment process that reimburses members for out-of-network health
expenses is to be controlled through use of a fraction nonconforming
chart. Initially, one sample of size 200 is taken each day for 20 days, and
the results shown in the following Table are observed.
a) Establish a control chart to monitor future performance.
3+2+⋯+2 50 Day Nonconforming Units
• 𝑝= = = 0.0125. 1 3
20×200 4000
2 2
The fraction nonconforming control chart 3 4
4 2
𝑝 1−𝑝 5 5
UCL: 𝑝 + 3
𝑛 6 2
7 1
CL: 𝑝 8 2
9 0
𝑝 1−𝑝 10 5
LCL: max 0, 𝑝 − 3 11 2
𝑛
12 4
13 1
14 3
UCL: 0.0361 15 6
CL: 0.0125 16 0
17 1
LCL: 0 18 2
19 3
20 2
p chart
0.04
0.03
0.02
0.01
0
1 2 3 4 5 6 7 8 9 1011121314151617181920
Sample
b) What is the smallest sample size that could be used for this process
and still give a positive lower control limit on the chart?
• From (a), we have

𝑝 1−𝑝
𝑝−3 >0
𝑛
9𝑝 1 − 𝑝
𝑝2
>
𝑛
9 1−𝑝 9 × 0.9875
𝑛> = = 711.
𝑝 0.0125
Hence, the smallest sample size is 712.