SPC MODELING
Create a Control Limit chart for the following: You are a manager at RadioTag Inc., a maker of RFID tags for US military and private use (e.g., Wal Mart). Your process assembles radio id tags to certain specifications – 95.5 confidence intervals, however as they come off the assembly line a test is run on each tag to determine whether the frequency works or does not (hint the data are dichotomous distributed binary). Below are the data you collected over a 2-week period: GRAPH GOES UNDER THIS
Day sample size non-working tags
1 100 15
2 100 13
3 100 12
4 100 11
5 100 16
6 100 19
7 100 5
8 100 8
9 100 19
10 100 15
11 100 2
12 100 6
13 100 9
14 100 10
P bar =
SD of P bar =
UCLP Bar =
LCLP bar =
B) Graph and Plot the data. Round answers to three places.
Clearly identify the P bar, s.d. of Pbar, UCL, LCL (make sure the graph elements are labeled) and identified in A above. Place a box around these final answers. Failure to place a box will result in failure to earn points.
C) Compare p value to a. What is the result?
Day | n | np | p = np / n |
1 | 100 | 15 | 0.1500 |
2 | 100 | 13 | 0.1300 |
3 | 100 | 12 | 0.1200 |
4 | 100 | 11 | 0.1100 |
5 | 100 | 16 | 0.1600 |
6 | 100 | 19 | 0.1900 |
7 | 100 | 5 | 0.0500 |
8 | 100 | 8 | 0.0800 |
9 | 100 | 19 | 0.1900 |
10 | 100 | 15 | 0.1500 |
11 | 100 | 2 | 0.0200 |
12 | 100 | 6 | 0.0600 |
13 | 100 | 9 | 0.0900 |
14 | 100 | 10 | 0.1000 |
Average | 0.1143 | ||
p_bar |
p_bar = 0.1143
SD of p_bar, Sp = [p_bar * (1 -
p_bar) / n]1/2 = sqrt(0.1143*(1 -
0.1143) / 100) = 0.0318
Confidence level = 95.5%, so, α = 0.045 and α/2 = 0.0225
So, zα/2 = Normsinv(1 - 0.0225) = 2.00
UCLp = p_bar + zα/2 *
Sp = 0.1143 + 2*0.0318 = 0.1779
LCLp = Max(0, p_bar - zα/2 *
Sp)= Max(0, 0.1143 - 2*0.0318) =
0.0507
SPC MODELING Create a Control Limit chart for the following: You are a manager at RadioTag Inc.,...
Create a Control Limit chart for the following: You are a manager at RadioTag Inc., a maker of RFID tags for US military and private use (e.g., Wal Mart). Your process assembles radio id tags to certain specifications – 95.5 confidence intervals, however as they come off the assembly line a test is run on each tag to determine whether the frequency works or does not (hint the data are dichotomous distributed binary). Below are the data you collected over...