Professional Documents
Culture Documents
b)
No. Meas.
3
0
48.7 48.9 49.1 49.3 49.5 49.7 49.9 50.1 50.3 50.5 50.7
Length range - cm
Number of
Measurements
1
2
2
0
1
1
2
0
1
b)
No. Meas.
8
7
6
5
4
3
2
1
0
48.1
Bin
(in)
48.1-50.0
50.1-52.0
52.1-54.0
54.1-56.0
56.1-58.0
58.1-60.0
60.1-62.0
62.1-64.0
64.1-66.0
68.1-70.0
Number of
Measurements
0
0
0
0
0
8
2
0
0
0
50.1
52.1
54.1
56.1
58.1
60.1
62.1
Length range - in
6.2 a)
Data arranged in bins with width 2in
6.1
64.1
66.1
68.1
6.2
6.3
3
Total no. meas. = 12
No.
Meas.
0
80
85
90
95
100
105
110
115
120
125
Bin
80-84.9
85-89.9
90-94.9
95-99.9
100-104.9
105-109.9
110-114.9
115-119.9
120-124.9
No.
0
1
2
0
2
2
3
1
0
No.
Meas.
4
0
8
8.5
9.5
10
6.4
Bin
8-8.49
8.5-8.99
9-9.49
9.5-9.99
10-10.49
10.5-10.99
No.
0
3
3
4
2
0
6.3
10.5
11
6.5
x 49.6cm
Median:
Arranging data in ascending order:
48.9, 49.2, 49.2, 49.3, 49.3, 49.8, 49.9, 50.1, 50.2, 50.5
Median
49.3 49.8
49.6cm
2
Standard Deviation:
S
( x i x )2
i 1 ( n 1)
n
(10 1)
(10 1)
(10 1)
S 0.53 cm
6.4
1
2
6.6
Median:
Arranging data in ascending order:
58.8, 59.1, 59.2, 59.3, 59.3, 59.8, 59.8, 60.0, 60.3, 60.4
Median
59.3 59.8
59.6in
2
Standard Deviation:
S
i 1
( xi x ) 2
( n 1)
(10 1)
(10 1)
(10 1)
S 0.54in
Modes: 59.3in,59.8in
6.5
1
2
6.7
x
x 105 psi
Median:
Arranging data in ascending order
89, 92, 94, 100, 104, 106, 108, 110, 110, 114, 115, 120
Median
106 108
2
Median 107
Standard Deviation:
S
( x i x )2
i 1 ( n 1)
n
(12 1)
(12 1)
(12 1)
Mode:
110psi
6.6
1
2
6.8
x
9.5 9.3 9.4 8.9 8.8 8.7 9.0 9.8 10.2 10 9.5 9.9
12
x 9.4bar
Median:
Arranging data in ascending order
8.7, 8.8, 8.9, 9.0, 9.3, 9.4, 9.5, 9.5, 9.8, 9.9, 10.0, 10.2
Median
9.4 9.5
2
Median 9.45
Standard Deviation:
S
( x i x )2
i 1 ( n 1)
n
(8.7 9.4) 2
(8.8 9.4) 2
(8.9 9.4) 2
S
(
12
1
)
(
12
1
)
(
12
1
)
1
2
S 0.50bar 1bar
Mode:
9.5bar
1 1 2
0.056
6 6 36
5.6%
1 1
2
0.056 5.6%
6 6 36
12%
6.7
6.12
6.13
6.14
n 5 5!
1
r 5 5!0
n 5 5!
1
r 0 0!5
n 5 5!
1
r 5 5!0
n 5 5!
1
r 0 0!5
6.9
n 6 6!
1
r 6 6!0
6.10
n 6 6!
1
r 6 6!0
2 2!
2
1 1!
2 2!
1
2 2!0
6.11
6.20
f ( x)
2 x3
P ( x )
3x
x
dx
35
35
2
27
8
1
35
35
xf ( x )dx
3x 2
dx
35
3x
3 x4
dx
35
4 35
2
3
81 16
140
1393
.
(x )
f ( x )dx
3x 2
(x )
dx
35
2
2
( x 2 2x 2 )
2
3
3x 2
dx
35
3x 4
3x 3
3x 2
3
dx 2
dx 2
dx
x5
35
35
35
5 35
2
2
3
3 5 25 194
. 2.774
5 35
1666
.
6.12
3
2
2 2 2
6.21
x3
1 x 3
20
P( x )
1 x
20 dx 4 20
1
(81 1) 1
80
xf ( x)dx x
1x
x3
dx
20
5 3
20 dx 5 20
1
243 1
100
2.42
(x )
f ( x ) dx
(x
)2
x3
dx
20
(x 2 2
x 2)
3
x5
x4
dx 2
dx
20
1 20
6
1 x
[
20 6
2 x
x3
dx
20
x3
dx
20
x4
4
]
1
1 1
2 2.42
( 2.42) 2
[ (729 1)
( 243 1)
(81 1)]
20 6
5
4
1
0.458
6.13
6.22
(a) for x 0
P ( x 0)
f ( x )dx
f ( x )dx
3x 2
x3
1
dx
0.0286 or 2.86%
35
35 0 35
f ( x )dx
0dx 0
x 2
3x 2
x3
8
dx
2 35
35 35
2 x 3
3x 2
35 dx 1
2
f ( x )dx
3x 2
x3
8
dx
0.2286 or 22.86%
35
35 2 35
(b) for 0 x 1 P (0 x 1)
F( x )
6.23
3 x
6.14
6.24
P (r )
In this case:
n
n!
r !( n r )!
r
p=0.95
n=4
P( 2)
4
4!
6
( 3 !)(1!)
2
6 0.9025 0.0025
0.0135
4
3
1
( 0.95) (1 0.95)
3
P( 3)
4
4!
4
(3 !)(1!)
3
4 0.8574 0.05
0.1715
6.25
or
99.95%
We want the probability that at most 2 computers will fail. The is the
probability that 0, 1 or 2 failures. Define success as a computer failure. Then
p=0.1. Then we want the probability that 0, 1 or 2 will fail.
20
0
20
0.1 (1 0.1) 0.121577
0
P (0)
20
1
19
0.1 (1 0.1) 0.27017
1
P (1)
20
2
18
0.1 (1 0.1) 0.28518
2
P (2)
6.15
6.26
20
20 !
190
2! 18 !
2
P ( 2)
20
3
17
(0.1) (1 0.1) .1901;
2
P ( 3)
20
20 !
1140
3 ! 17 !
3
20
4
16
(0.1) (1 0.1) 0.098;
4
P ( 4)
20
20 !
4845
4 ! 16 !
4
20
5
158
0.032;
(0.1) (1 0.1)
5
P ( 5)
20
20 !
15504
5 ! 15!
5
P( 6)
0.7351
6
6!
1
6! 0!
6
73.51%
(b) For at least two parts to be satisfactory, we should find the sum of
probabilities for 2,3,4,5 and 6 parts to be satisfactory:
P (number of successes 2) 1 P (0) P (1)
6
0
6
( 0.95) (1 0.95)
0
P (0)
6
6!
1
0! 6!
0
156
. 10 8
6
1
5
(0.95) (1 0.95)
1
P (1)
6
6!
6
1! 5 !
1
6.16
6.28
n
n!
r !( n r )!
r
P (r )
P(3)
85.7%
b- Exactly one month with power failure in four months: n=4, r=3
4
3
1
( 0.95) (1 0.95)
3
P( 3)
4
4!
4
3 !(4 3)!
3
17.15%
P ( 0)
5
1
4
5
( 0.95 ) ( 0.05 ) 2.969 10
1
5
5!
5
1! 4 !
1
5
2
3
.
10 3
( 0.95 ) (0.05) 1128
2
5
5!
10
2! 3 !
2
5
3
2
(0.95 ) ( 0.05) 0.0214
3
5
5!
10
3 ! 2!
3
5
4
1
(0.95 ) ( 0.05) 0.2036
4
5
5!
5
4 ! 1!
4
P (1)
P ( 2)
P ( 3)
P( 4)
4
P(i ) 0.226
i 0
6.29
P (r )
n
n!
r !( n r )!
r
16
0
16
(0.01) (0.99)
0
P (r 0)
P (r 0) 0.85
16
16 !
1
0 ! 16!
0
6.17
6.30
Binomial distribution
n 10 10!
4950
r 2 !98
n 10 0!
7528 0
r 5 !95
n 10 10! 13
1.73x0
r 10 !90
Note: One cannot normally compute 100! with a hand calculator it is a very large number. However
100!/(2!98!) can be rewritten 100x99x98!/(2!98!) =100x99/2!, which can be readily computed.
n 10 10!
1
r 0 0!1
n 10 10!
10
r 1 1!9
This must be solved by trial and error for p. Using a spreadsheet, the answer is
p=0.0155
6.31
Binomial distribution
n 10 10 !
10
r 1 !9
6.19
n 10 10!
3921 5
r 4 !96
n 10 10! 17
2.5310
r 15 !85
Note: One cannot normally compute 100! with a hand calculator it is a very large number. However
100!/(1!99!) can be rewritten 100x99x98!/(1!99!) =100/1!, which can be readily computed.
n 10 10!
1
r 0 0!1
n 10 10!
10
r 1 1!9
6.20
This must be solved by trial and error for p. Using a spreadsheet, the answer is
p=0.0155
6.32 We are looking at the probability that more than 175 passengers will show up. This
can be solved as a binomial distribution problem. Consider success that a passenger
shows up, so p=0.95. We then want more than 175 successes out of 180 trials.
n 180 ! 180 79 !
160
178 !2 178!xxr 2
n180 ! 18079!
180
1 79 !1 79!xxr 1
n180 180!
1
r 180 180!x
P (176 ) 42.297 10 6 x 0.95176 (1 0.95) 4 0.03174
P (177 ) 955860 x 0.95177 (1 0.95) 3 0.01363
P (178 ) 16110 x 0.95 178 (1 0.95 ) 2 0.004363
6.22
6.23
6.33 We are looking for the probability that there will 5 or less defective components.
Consider success to be a defective component then p = 0.05, n=55 and r = 0,1,2,3,4,5
r=0:
r=1:
r=2:
r=3:
r=4:
5 5!5
1
0 0! 5!5
5 5!
5
1 1!54!
5 5!
1485
2 2!53!
5 5!
26 35
3 3!52
5 5!
34105
4 4!51
6.24
r=5:
5 5!
3478 61
5 5! 0
So P(x>5) = 1 ( 6.74 x10 3 0.03369 0.0842 0.1404 0.1755 0.1755 ) =0.384. The
probability that there will be more than 5 visits in an hour is 0.384
6.35 This can be solved as either a binomial distribution (p=0.001, n=4, r=1) or as a
Poisson distribution. For the poisson distibution, the expected occurrence () will be
4/1000 = 0.004. We are then looking for the probability of x = 1.
P (1) e 0.004 .004 1 / 1! 3.984 x10 3 .
6.36 This is a Poisson distribution. The expected number of crashes, , is (1/3)5=1.666.
Then, P (0) e 1.666 1.666 0 / 0! 0.189
6.37 This is a Poisson distribution problem although it can be done as a binomial
distribution. The expected frequency () of defects will be (1/50)x10 = 0.2 for the typical
10 m2 kitchen. The probability of 1 or more defects is 1-P(0).
P (0 ) e 0.2 0.2 0 / 0! 0.818 P(x1) = 1-0.818 = 0.182. If solved as a binomial
distribution , p= 0.02, n=10. For r = 0,
1 0 ! 0 10
1 (0)1xP .2(1.02) 0.81
0 !10
6.38 Poisson distribution. The fact that a bulb has failed does not affect the probability of
other failures. We need to know what the probability of two or more failures during the
day. The probability of 2 or more is:
P(x2) = 1 P(0) P(1). = 2.
P (0) e x / x! e 2 2 0 / 0! 0.13533
6.25
P (1) e x / x! e 2 21 / 1! 0.2707
6.26
6.43
The area from 100 to 100.5 is 0.2. From Table 6.3, z =
0.52
0.52
20%
20%
110 .5 100
, 0.96
100.75 100
0.78125
0.96
99.5
100
100.5
28.26%
100
for error greater than 100.75 or less than 99.25 (1000.75) we will have:
P ( z ) 2 (0.5 0.2826) 0.4348
43.48%
6.44
The area from 110 to 110.5 is 0.25. From Table 6.3, z =
0.67
0.67
110.5 110
, 0.75
111 110
1.33
0.75
6.27
100.75
6.45
6.827 6.832
0.5
0.01
a) z1
b) z1
c) z1
6.831 6.832
0.1
0.01
6.28
6.46
a) z1
7.76 7.75
1 .0
0.01
b) z1
c) z1
7.85 7.75
10
0.01
6.29
no. of
students
0.342
0.66
1.3785
2.247
2.8725
2.8725
2.247
1.3785
0.66
0.342
Of course, the number of students with each grade is an integer. If we round off, we will
only get 14 students total so some judgement is required for the additional student.
6.30
40%
10%
6.48
a)
x 10000
500
x = 9,360 hrs
b)
z1
7000 10000
6 area 0.5( fromTable 6.3)
500
z2
10000 10000
0 area 0
500
0.5 0 0.5
50% or1000 bulbs will fail
7 .85
7.9
7.95
8.0 5
8.1
8.15
8.2
8.25
6.50 (a) A 2 tolerance means that values greater than z = 2 or less than z =-2. For z =
2.0, the value from the normal distribution table is 0.4772 . The right hand tail area (for
z>2) is 0.5 0.4772 = 0.0228 The left hand tail has the same area. The probability of
rejection is then 2x0.0228 = 0.0456 or 4.56%.
(b) This part is best solved using the binomial distribution. For 2 rejections, we define
success as a rejection. Then for 2 rejections, r = 2, n=20 and p = 0.056. The probability
6.31
of 2 rejected parts is then 0.1705. Similarly, for r = 4 and r = 10 we get 0.0099 and
4.5x10-9 respectively.
6.51
0.2 inches
x 7.25 ft 87 inches
d x x 0.3 acceptable
d
0.3
z
15
.
0.2
From Table 6.3, A 0.4332
0.3
0.14 inch
2.17
This means that the manufacturer has to cut the columns close to the standard size.
6.52
x 10.500 inches
0.005 inches
10.500
4
0.005
0.005
P ( x 10.520) 0.5 0.5 1 100% U sin gTable 6.3
a) z
10.485 10.500
3
0.005
10.515 10.500
z2
3
0.005
b) z1
6.32
10.520
6.53
x 25.00cm
0.02cm
1
0.02
0.02
P ( x 25.02) 0.5000 0.3413 0.8413 84.13%
a) z
24.95 25.00
2 .5
0.02
25.05 25.00
z2
2.5
0.02
b) z1
6.54
40%
10%
50,000
a) = 5000
From table 6.3 for area = 0.40
z 1.28
x 50,000
128
.
5000
x = 43,600 miles
6.33
b) z1
60,000 50,000
2
5000
area = 0.4772
50,000
z2
70,000 50,000
4
5000
60,000
70,000
area = 0.5
0.5-0.4772 = 0.0228
(100,000 tires)(0.0228) = 2280 tires fail between 60,000 to 70,000 miles
c)
20,000 50,000
6
5000
40%
10%
6.55
= 160 hrs
For A = 0.4, z 1.28
x 3600
128
.
160
x 3395.2
6.34
n
2
(196
. )( 2)
30
40
30 0.620 mph with confidence of 95%
x z
6.35
6.60
1- = 0.95 = 0.05
0.5-/2 = 0.475
z = 1.96 (From Table 6.3)
x z
2
50,000 196
. ( 5000 /
50,000 980 miles
100)
6.36
6.61
n 10 VCRs
x 1500 hours
S 150 hours
0.05
Find the 95% confidence interval on the mean of the life of the VCRs.
n 1 9
a)
. (FromTable 66. )
t 2262
0025
2
.
2
x t (S
2
n)
1500 2.262(150
1500 107 hours
10 )
t S 50 hrs
n
2
( 0.025)
2
Assuming S =150 hrs will remain the same (in reality should be recalculated and
it may change), using Table 6.6 we find n by trial and error.
n
15
20
25
36
2.145
2.093
2.064
2
t S
2
n
(hrs )
83
70
62
50
6.37
6.62
For this problem, we need to find a one-sided confidence interval of the form
x t
S
n
t-distribution
(a) x = 41.25x106, S = 0.30x106, n=10. For 99% confidence level, = 0.01 and = 10
1 = 9 from Table t = 2.821. The tolerance interval is then:
x t
S
n
41.25x10 6 2.821
0.3 x10 6
10
40.982 10 6
41.0x106 falls into this interval so the manufacturer cannot be confident that the average
strength exceeds this limit.
(b) If we change n to 20, then = 20 1 =19 and t = 2.539.
41.25x10 6 2.539
0.3 x10 6
20
41.080 10 6
Now the manufacturer can be confident the mean exceeds 41.0x10 6 psi.
6.38
6.63
n 10
x 910%
.
S 0.8%
0.05
n 1 9
0.025 t 2 2.262 ( From Table 6.6)
x t S
2
91 2.262 0.8
n
10
910
. 0.57
Cut confidence interval by one half > 0.57/2 = 0.285
Assuming that n > 30, and S to remain the same:
z
2
z 0.025
0.285
n
1.96
( S )
S 0.8
z 2
0.285
n 30
6.39
6.64
n8
x 88.5%
S 0.5%
0.05
n 1 7
0.025 t 2 2.365 ( FromTable 6.6)
x t S
2
88.5 0.42
Cut confidence interval by one half > 0.42/2 = 0.21
Assuming that n > 30, and S to remain the same:
z
2
z 0.025
0.21
n
1.960
( S )
S 0. 5
z 2
0.21
n 30
6.40
6.65
Find: N
0.5 2 0.495
z 2.575
1 in,
n
2.575
2.44
n
1 n 40
0.4 in,
n
2.575
0.34
0.2 n 18
n
n 15
x 25 ppm
S 3 ppm
0.05 n 1 14
2 0.025
2
14
, 0.025 26.119 ( FromTable 6.7 )
2
14
,1.025 5.6287
( n 1) S 2
2
14
,.025
14 32
( n 1) S 2
14 32
4.82;
22.17
2
26.119
14
5.6287
,1.025
6.41
6.67
The distribution governing the time interval between the arrival successive cars is
give to be
f(t,) = e-t
Where
1 1
(a)
P (t 6) 1 e ( 6 ) 1 e
6
8
0.5276
(b)
P(t 15) 1 P(t 15) 1 (1 e
(c)
15
8
) 0.1534
In order to have only two arrivals in 10 minutes, if the interval between the first
1
two arrivals is t, mins, the interval between the next arrival should be greater than (10-t )
Thus,
1
1 10
1 810
e dt1 e 10 1.25e 1.25 0.3581
8
0
10
6.68 We are given that TBF has an exponential distribution with = 450 hours.
Therefore
1
1
450
(a)
P ( failure 300hours ) 1 e
( 300 )
1 e
300
450
0.4856
(b)
P ( failure 500hours ) e ( 500 ) 1 e
(c)
500
450
0.3292
[1 e
600
450
] [1 e
300
450
6.42
6.43
6.69 Let the pollution level, X, be in ppm. If we define y = log(X), then y is N(,),
where = 1.9031 and = 1.3010
Note: Here we have used y = log(X) rather than y = ln(X)
(a)
1.9542 1.9031
] P[ Z 0.393]
1.3010
(b)
(c)
If the mean pollutant level is reduced to 40 ppm, then the new value of = log(40
= 1.6021
P[ X 20] P[ y log(20)] P[Y 1.3010]
1.3010 1.6021
P[ Z
] P[ Z 0.2304]
1.3010
0.5 0.0915 0.4085
6.44
6.70 Let X be the time taken to fix the software bug. X has a mean of 200 mins, a
standard deviation of 30 mins. If y = log(X) is N(,) with = log(200) = 2.3010, =
log(30) = 1.4771
(a)
2.6990 2.3010
] P[ Z 0.2694]
1.4771
(b)
2 2.3010
2.3010 2.3010
Z
] P[0.2038 Z 0]
1.4771
1.4771
0.0808
P[
(c)
6.71
Chi-squared distribution.
S = 0,002 in., n = 10, = n 1 = 9. 95% confidence level: = 1-0.95 = 0.05.
/2 = 0.025, 1-/2 = 0.975. From Table 6.7, 2/2 = 19.023, 21-/2 = 2.704.
(n 1)S 2
(n 1)S 2
(10 1)0.002 2
(10 1)0.002 2
2
2
2 / 2
12 / 2
19.023
2.704
6
2
5
or 0.00137 0.00364
1.89 10 1.33 10
6.72
Chi squared distribution.
S = 5500, n = 8, = 7. = 1-0.99 = 0.01, /2 = 0.005, 1-/2 = 0.995.
2/2 = 20.278, 21-/2 = 0.9893.
(n 1)S 2
(n 1)S 2
(8 1)5500 2
(8 1)5500 2
2
2
2 / 2
12 / 2
20.278
0.9893
2
6
10442351 214.0 10 or 3231 14360
6.45
6.73
Chi squared distribution.
S = 10, n = 12, = 11. = 1-0.99 = 0.01, /2 = 0.005, 1-/2 = 0.995.
2/2 = 26.757, 21-/2 = 2.6032.
(n 1)S 2
(n 1)S 2
(12 1)10 2
(12 1)10 2
2
2
2 / 2
12 / 2
26.757
2.6032
2
41.11 422.5 or 6.41 20.55
We cannot be 99% confident that the standard deviation is less 15 mA.
6.74
There is an error in the problem. The required variance should be 0.0004 mm 2. (not m2).
(a) Chi squared distribution.
S = 0.01, n = 10, = 9. = 1-0.95 = 0.05, /2 = 0.025, 1-/2 = 0.975.
2/2 = 19.023, 21-/2 = 2.7004.
(n 1)S 2
(n 1)S 2
2
2 / 2
12 / 2
(10 1)0.012
(10 1)0.012
2
19.023
2.7004
(b) The maximum of the confidence interval is less than the desired variance so the part
is acceptable.
(c) S = 0.01, n = 5, = 4. = 1-0.95 = 0.05, /2 = 0.025, 1-/2 = 0.975.
2/2 = 11.143, 21-/2 = 0.4844.
(n 1)S 2
(n 1)S 2
(5 1)0.012
(5 1)0.012
2
2
2 / 2
12 / 2
11 .143
0.4844
5
2
4
3.590 10 8.258 10
The upper end of the confidence interval exceeds the allowable so the part is not
acceptable.
6.46
6.75 There is an error in the problem. The required variance should be 0.0004 mm 2.
(not m2).
(a) Chi squared distribution.
S = 0.01, n = 10, = 9. = 1-0.99 = 0.01, /2 = 0.005, 1-/2 = 0.995.
2/2 = 23.589, 21-/2 = 1.7349.
(n 1)S 2
(n 1)S 2
(10 1)0.012
(10 1)0.012
2
2
2
2
/ 2
1 / 2
23.589
1.7349
5
2
4
3.81 10 5.188 10
(b) The maximum of the confidence interval is greater than the desired variance so the
part is not acceptable.
6.76 There is an error in the problem. The required variance should be 0.0004 mm 2.
(not m2).
(a) Chi squared distribution.
S = 0.01, = 1-0.99 = 0.01, /2 = 0.005, 1-/2 = 0.995.
(n 1)S 2
(n 1)S 2
2
2 / 2
12 / 2
We need to determine n so that the maximum of the confidence interval is 0.0004. We
are told that n = 10 was too small so n must be larger. The following calculations were
performed on a spreadsheet program.
n
11
12
13
14
nu
10
11
12
13
chisq
2.155845
2.603202
3.073785
3.565042
upper lim
0.000463855
0.000422557
0.000390398
0.000364652
6.47
6.77
xi
n 1
0.530
xi x
Low po int
High po int
6.48
6.79
x
xx
( x x) 2
y y
( y y) 2
20
30
40
50
1.02
1.53
2.05
2.55
-33.5714
-23.5714
-13.5714
-3.5714
1127.04
555.611
184.183
12.7549
-1.5271
-1.0171
-0.4971
0.0029
51.2669
23.9745
6.7463
-0.01036
60
75
100
3.07
3.56
4.05
6.4286
21.4286
46.4286
41.3269
459.185
2155.61
0.5229
1.0129
1.5029
2.3320
1.0345
0.25711
-6
8.4110
0.27342
1.0260
2.2587
7.1717
176.8214
4535.71
( x x)( y y)
3.3615
21.7050
69.7775
x 53.5714
y 2.5471
rxy
rxy
(x
(x
x )( y i y )
x )2 ( y i y )2
12
176.8214
435.71 7.1717
12
rxy 0.9804
6.49
6.80 (a)
x y
n x x
x y x x y
b
n x x
n xi y i
xi
yi
(xi)
20
30
40
50
60
75
100
1.02
1.53
2.05
2.55
3.07
3.56
4.05
400
900
1600
2500
3600
5625
10000
375
17.83
24,625
(yi)
xiyi
1.0404
2.3409
4.2025
6.5025
9.4249
12.6736
16.4025
20.4
45.9
82.0
127.5
184.2
267.0
405
52.5873
1132
7(1132) ( 375)(17.83)
0.0398
7( 24625) ( 375) 2
( 24625)(17.83) ( 375)(1132)
b
0.4587
7( 24625) ( 375 ) 2
y 3.898 10 2 x 0.4587
y V( mV)
x T(C)
(b)
4.5
4
3.5
3
2.5
Best Fit
Line
2
Output - mV
1.5
1
20
40
60
80
Temperature - C
6.50
100
6.51
6.81
(a)
xi(t)
yi(t)
( x i x)
( x i x) 2
( y i y)
( y i y) 2
0.0000
0.1000
0.2000
0.3000
0.4000
0.5000
4.98
1.84
0.68
0.25
0.09
0.03
-0.2500
-0.1500
-0.0500
0.0500
0.1500
0.2500
0.0625
0.0225
0.0025
0.0025
0.0225
0.0625
3.6683
0.5283
-0.6317
-1.0617
-1.2217
-1.2817
13.4567
0.2791
0.3990
1.1271
1.4925
1.6427
-0.9171
-0.0793
0.0316
-0.0531
-0.1833
-0.3204
=18.3971
=-1.5215
=0.1750
( y i y)( x i x )
x 0.25 sec.
y
rxy
rxy
13117
.
Volts
(x
(x
x )( y i y )
x )2 ( y i y )2
15215
.
0.1750 18.397
12
12
0.848
yi(lnV)
( x i x)
0.0000
0.1000
0.2000
0.3000
0.4000
0.5000
1.6054
0.6098
-0.3857
-1.3863
-2.4079
-3.5066
-0.2500
-0.1500
-0.0500
0.0500
0.1500
0.2500
( x i x) 2
0.0625
0.0225
0.0025
0.0025
0.0225
0.0625
( y i y)
2.5173
1.5216
0.5262
-0.4744
-1.4961
-2.5947
=0.1750
( y i y) 2
6.3368
2.3154
0.2769
0.2251
2.2382
6.7324
=18.124
( y i y)( x i x )
-0.6293
-0.2282
-0.0263
-0.0237
-0.2244
-0.6487
=-1.780
x 0.25
y 0.912
rxy
rxy
(x
(x
x )( y i y )
x )2 ( y i y )2
1780
.
0.175 18.124
12
12
6.52
6.82
For the best fit line, the equ. is reading=1.0525weight-1.1661
For the forced zero, reading=0.9987weight.
40
reading
30
20
y = 1.0525x - 1.1661
y = 0.9987x
10
0
-10 0
10
20
weight - lb
6.53
30
40
6.83
DelP
1.96
4.2
4.9
5.48
5.91
7.3
7.73
9
9.9
Sy.x=
rsquared=
DelP - in H2O
Voltage
0.31
0.65
0.75
0.85
0.91
1.12
1.19
1.38
1.52
12
10
8
6
4
2
0
y = 6.5606x - 0.0629
0.5
1.5
Voltage
6.84 For the data given in Problem 6.79, we first compute the following quantities,
using x = T and y = V
0.0390
7 24625 (375) 2
n xi 2 ( xi )2
n xi 2 yi ( xi )( xi y i )
n xi ( xi )
2
14590
0.4590
31750
/2,n-2
(a)
( y
Yi ) 2
0.2394
n2
i
6.54
0.025,5
=t
b S yx t
2
( xi ) 2
1
2
n n ( xi 2 n x 2 )
, n 2
1
375 2
2
7 7 ( 24625 7 53.17 2 )
0.4595 0.5284 (0.0688,0.9879)
(b)
When T = 70 C, V = aT + b = 3.1929V
n 1
( x * x) 2
n
xi 2 n x 2 where
x* = 70 and y* = 3.1929V
o
7
24625 7 53.212
3.1929 0.6739 (2.5191,3.8668)
(c)
,n-2
At = 0.05, t
0.05,5
=t
= 2.015
7
24625 7 53.212
3.1929 0.5281 3.7211volts
6.85 Based on the given data, we compute the following quantities, where x = T oF and
y = in ksi.
6.55
(a)
the tensile strength, (ksi). The coefficients of the linear regression model are
a
(b)
exp
0.05,10
=t
= 1.812
o
exp S yx t
2
,n 2
n 1
( x * x) 2
n
xi 2 n x 2 where
x* = 670 and
S yx
( y
Yi ) 2
115 .6
3.34
n2
10
i
)
12 818 10 4 12 750 2
( 29.2056 6.3153) ( 22.89,35.52) ksi
(c)
Therefore, 95% confidence interval for change in strength for a 100 F change is =
(-4.34, -3.09) ksi
o
(d)
,n 2
n 1
( x * x) 2
n
xi 2 n x 2
where
x* = 550, = 0.05, n = 12
,n-2
0.05,10
=t
= 1.812
Then,
0.0372 550 54.099 3.34 1.812
12 1
(550 750)
12
818 10 4 12 750 2
= 33.6359 ksi
6.57
6.86
a) Best fit line for t and V
2
xi (t)
yi (V)
xi
0
0.01
0.04
0.09
0.16
0.25
0.0
0.1
0.2
0.3
0.4
0.5
4.98
1.84
0.68
0.25
0.09
0.03
yi
24.800
3.3856
0.4624
0.0625
0.0081
0.0009
xiyi
0
0.184
0.136
0.075
0.036
0.015
x y
n x x
n xi y i
6( 0.446) (15
. )(7.87)
8.694
6(0.55) (15
. )2
x y x x y
b
n x x
2
(0.55)(7.87) (15
. )(0.446 )
3.485
6( 0.55) (15
. )2
0.0
0.1
0.2
0.3
0.4
0.5
xi
0
0.01
0.04
0.09
0.16
0.25
.
15
0.55
yi
xiyi
1.6054
0.60977
-0.38566
-1.3863
-2.4079
3.5066
yi
2.5774
0.37181
0.14874
1.9218
5.7982
12.2959
0
0.060977
-0.077132
-0.41589
-0.96316
-1.7533
5.47129
.
231139
.
31485
6.58
x y
n x x
n xi y i
6( 3.1485) (15
. )( 5.47129)
10.175
6(0.55) (15
. )2
x y x x y
b
n x x
2
2
i
b y i a x i y i
n2
62
11369
.
12
S y .x
S y .x
.
)( 5.47129 ) ( 10.175)( 3.1485)
( 23.1139) (16319
62
12
Sy .x 0.0378
The standard error for (T vs ln(V)) is much less than for (t vs V) therefore there is
less data scatter around this best fit curve.
6.59
6.87
DelP
0.05
0.07
0.09
0.12
0.15
0.17
0.19
0.21
0.23
0.25
Sy.x=
Q
2
2.35
2.7
3.12
3.5
3.72
3.85
4.1
4.35
4.45
ln(DelP)
-2.99573
-2.65926
-2.40795
-2.12026
-1.89712
-1.77196
-1.66073
-1.56065
-1.46968
-1.38629
0.105469 Sy.x=
ln(Q)
0.693147
0.854415
0.993252
1.137833
1.252763
1.313724
1.348073
1.410987
1.470176
1.492904
pred(lnQ)
0.693541
0.862349
0.988434
1.132764
1.244715
1.307509
1.363311
1.413523
1.459164
1.500996
e
0.000394
0.007934
-0.00482
-0.00507
-0.00805
-0.00621
0.015238
0.002536
-0.01101
0.008092
e/Sy.x
0.04402
0.886565
-0.53839
-0.56645
-0.89932
-0.69442
1.702752
0.283388
-1.23055
0.904233
log10(DelP)
-1.30103
-1.15490196
-1.04575749
-0.92081875
-0.82390874
-0.76955108
-0.7212464
-0.67778071
-0.63827216
-0.60205999
log10(Q)
0.30103
0.371068
0.431364
0.494155
0.544068
0.570543
0.585461
0.612784
0.638489
0.64836
0.008949
y = 0.5017x + 2.1965
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
0
1.6
1.4
y = 12.179x + 1.5506
1.2
1
ln(Q)
Q-cfm
0.8
0.6
0.4
0.2
0
0.05
0.1
0.15
0.2
0.25
0.3
-4
-3
DelP- psi
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-0.5
2
1
e/Sx.y
log10(Q)
-1
-1
ln(DelP)
y = 0.5017x + 0.9539
-1.5
-2
0
-1
0.05
0.1
0.15
0.2
0.25
0.3
-2
log10(DelP)
DelP
The ln(Q) vs ln(DelP) is clearly the better fit than the straight day (Q vs. DelP). The fit
with log10 is different but is equivalent to the ln fit and is just as good. The standardized
residuals were computed for the ln regression analysis. The value of the standardized
residual at DelP = 0.19 is 1.702 and this point is suspect although it is not clearly an
outlier.
6.60
6.88
Data
T
90
90
90
90
90
120
120
120
120
120
rpm
2000
3000
4000
5000
6000
2000
3000
4000
5000
6000
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.99993859
R Square 0.99987718
Adjusted R Square
0.99977892
Standard Error
0.00530498
Observations
10
ANOVA
df
Regression
Residual
Total
SS
MS
4 1.145549 0.286387
5 0.000141 2.81E-05
9 1.14569
F
Significance F
10176.2 5.85E-10
Q = -0.369+0.0005267T-4.3214E-8rpmsq+5E-7rpmxT
Q - l/s
V - m/s
6.61
6.89
V
4
5
6
7
8
This data is non-linear and the 2nd order fit is superior. Note the points at the ends the
linear fit is above them and the 2nd order passes through them.
Q - l/s
linear fit
4
3
2
1
0
y = 0.695x - 1.796
4
V - m/s
6.62
6.90
Velocity
0.07
0.08
0.11
0.135
0.145
0.185
0.19
0.22
0.24
0.285
0.295
0.4
0.3
Velocity
Voltage
0.01
0.115
0.29
0.48
0.59
0.81
0.88
1.02
1.12
1.325
1.4
0.2
0.1
0
0
0.5
1.5
Voltage
T
200
300
400
500
600
700
800
900
1000
1100
1200
1300
sigma
41.5
40.3
39.2
37.8
35
32.5
28.5
23
16
9.5
6.5
5
sigma
6.91
(a)
50
40
30
20
10
0
500
1000
1500
Temp
(b) Due to the S shape of the curve, at least a 3 rd order curve will be required.
(c) The 4th order curve fit is shown.
6.63
6.92
Let L1 be the length of the shelves and L2 be the length between sides of the cabinet.
The clearance between the two is = L2 L1. We want to be 99% confident that is
greater than equal to zero.
S SL22 SL21 0.06 2 0.03 2 0.067
Assume large sample size for the standard deviations so we can use the normal
distribution. We want to be sure that 99% of the shelves will not be too long. Using Table
6.3, For 99% confidence level, A = 0.49 and z =2.33
.This value of z corresponds to
The longest the nominal length the shelves can be is 24 0.156 = 23.844 in long. Note:
We did not concern ourselves with how short the shelves can be.
6.64