Académique Documents
Professionnel Documents
Culture Documents
P(X<3000) = P(Z<
30003675
430
P(X<3000) = P(Z<
30004500
)
980
= P(Z>-1.570)
= P(Z>-1.531)
= 1-P(Z<1.570)
= 1-P(Z<1.531)
= 1- 0.9418
= 1- 0.9371
= 0.0582
= 0.0629
Therefore the probability that a bulb made from process I is regarded as bad in is 0.0582
and the probability that a bulb made from process II is regarded as bad is 0.0629. Out of a
certain number, n of bulbs produced by process I, 5.82 % will have a lifetime less than 3000
hours. Out of a certain number, n of bulbs produced by process II, 6.29 % will have a lifetime
less than 3000 hours. Process I is more suitable to use as it has a higher yield as there is a
larger percentage of bulb with a lifetime greater than 3000 hours being produced.
From Process I,
P(X<3000) = P(Z<
35003675
430
P(X<3000) = P(Z<
35004500
)
980
= P(Z>-0.4070)
= P(Z>-1.020)
= 1-P(Z<0.4070)
= 1-P(Z<1.020)
= 1- -1 (0.4070)
= 1- -1 (1.020)
= 1- 0.6579
= 1- 0.8461
= 0.3421
=0.1539
When specification for bad lamps is less than 3500 hours, the percentage of bad bulbs
produced by Process I and II are 34.21% and 15.39% respectively. There is then a much
higher percentage of bad bulbs produced by process I. Therefore, process II will be more
suitable.
Distribution Plot
Normal
30003500
Mean StDev
3675 430
4500 980
0.0009
0.0008
Density
0.0007
0.0006
0.0005
0.0004
0.0003
0.0002
0.0001
0.0000
1000
2000
3000
4000
5000
6000
7000
8000
2.
Let Y be the random variable diameter of one metal plug in centimetres.
Sample number, n=100
Sample mean, yy =25.10
Since the sample number, n is greater than 30 (n=100), the central limit theorem can applied
and the sample is approximately normal. Therefore, the unbiased estimate of the population
n
n1
=
variance is as follows:
2/ n] =
~N (, 2/ n)
For a 99.9% confidence interval,
P(a<z<b)=0.999
Upper tail probability= 0.0005, therefore
P(Z<z)= 0.9995
z= -1 (0.9995)
= 3.291
Therefore, P(-3.291<z<3.291)= 0.999
) s2 =
100
2
99 (0.12 )
100
(0.0144)/100] =
99
0.0144
99
(25.10 +3.291
0.0144
99
, 25.10 -3.291
0.0144
99
Dial setting
1
2
3
4
5
6
7
8
9
10
Standard deviation
0.13383655
0.144840909
0.092646281
0.143406958
0.144360659
0.138904444
0.170293864
0.186455237
0.180419265
0.087819259
b)
Mean Cut length(mm) against Dial settings
90
85
80
75
70
Mean Cut Length
(mm)
mean
length65
(mm)
60
55
50
45
1
Dial Settings
c)
10
Versus Fits
99
0.2
Residual
Percent
90
50
-0.2
-0.4
10
1
0.0
-0.6
-0.50
-0.25
0.00
0.25
0.50
50
60
70
80
Residual
Fitted Value
Histogram
Versus Order
90
Residual
Frequency
0.2
3
2
1
0
0.0
-0.2
-0.4
-0.6
-0.6
-0.4
-0.2
0.0
Residual
0.2
0.4
10
Observation Order
From the histogram, we can see that the standard deviation of the mean cut lengths is
positively skewed. In the Versus fits, the points are scattered about the dotted line showing
that there is a linear relationship between the 2 variables. In the normal plot probability, the
plots are evenly scattered along the line of best fit. This shows that the cut lengths values
follow a normal distribution.
d) Yes, the length of cut rod vary linearly with the dial settings. When the mean value of the
lengths cut for each dial setting is plotted in Minitab and a line of best fit is drawn, the
correlation coefficient between the 2 variables is equal to +0.9997 (positive correlation). This
shows that there is a strong linear relationship between the dial setting and the cut rod
length.
Fitted Line Plot
Mean Cut Lengths (mm) = 46.08 + 4.006 Dial setting
90
S
R-Sq
R-Sq(adj)
0.299147
99.9%
99.9%
80
70
60
50
0
10
Dial setting
e)
Length of metal plug to be cut = 68.0
0.1 mm