Jose Alejandro Bastidas B.

SIMULATION

HOMEWORK 5 CHAPTER 3
ANALYSIS OF SIMULATION INPUT

1. The time required for 50 different employees to compute and record the number of hours worked
during the week was measured, with the following result in minutes:

Employee Time (min) Employee Time (min)

1 1.88 26 0.04
2 0.54 27 1.49
3 1.90 28 0.66
4 0.15 29 2.03
5 0.02 30 1.00
6 2.81 31 0.39
7 1.50 32 0.34
8 0.53 33 0.01
9 2.62 34 0.10
10 2.67 35 1.10
11 3.53 36 0.24
12 0.53 37 0.26
13 1.80 38 0.45
14 0.79 39 0.17
15 0.21 40 4.29
16 0.80 41 0.80
17 0.26 42 5.50
18 0.63 43 4.91
19 0.36 44 0.35
20 2.03 45 0.36
21 1.42 46 0.90
22 1.28 47 1.03
23 0.82 48 1.73
24 2.16 49 0.38
25 0.05 50 0.48

Use the chi-square test to test the hypothesis that these service times are exponentially distributed.
Let the number of class intervals be k = 6. Use the level of significance a = 0.05.

n = 50
With k = 6, the width of the interval is given by:
Class width = (max-min) / k = (5.5-0.01) /6=0.915

The mean (β) is equal to 1,206, therefore λ = 1 / β

Hypothesis testing
H0: The random variable has an exponential distribution
H1: The random variable does not have an exponential distribution
𝑛
(𝑂𝑖 − 𝐸𝑖 )2
𝜒02 = ∑
𝐸𝑖
𝑖=1

The expected value of an exponential distribution in an interval [a, b] is given by:

Jose Alejandro Bastidas B.
𝑏
1 −𝑥/𝛽
𝐸𝑖 = 𝑛 ∫ ℯ 𝑑𝑥
𝑎 𝛽

INTERVAL Li Ls CLASS MARK Oi Ei Oi* Ei* (Oi-Ei) (Oi-Ei)^2 (Oi-Ei)^2/Ei

1 0,01 0,925 0,4675 29 26,38392 29 26,38 2,62 6,843893 0,259396

2 0,925 1,84 1,3825 9 12,34566 9 12,34566 -3,35 11,19344 0,90667
3 1,84 2,755 2,2975 7 5,776827 12 10,33665 1,66 2,766744 0,267664
4 2,755 3,67 3,2125 2 2,703114
5 3,67 4,585 4,1275 1 1,264851
6 4,585 5,5 5,0425 2 0,591854
50 1,43373

The degrees of freedom are given by k-s-1 = 3-1-1 = 1

As a result, X02 = 1,433 and X0.05,1 = 3.8415> X02
With the results obtained, it is concluded that the null hypothesis is not rejected, and the data follow an
exponential distribution.

2. Studentwiser Beer Company is trying to find out the distribution of the breaking strength of their
glass bottles. Fifty bottles are selected at random and tested for breaking strength, with the following
results (in pounds per square inch):

218.95 232.75 212.80 231.10 215.95

237.55 235.45 228.25 218.65 212.80
230.35 228.55 216.10 229.75 229.00
199.75 225.10 208.15 213.85 205.45
219.40 208.15 198.40 238.60 219.55
243.10 198.85 224.95 212.20 222.90
218.80 203.35 223.45 213.40 206.05
229.30 239.20 201.25 216.85 207.25
204.85 219.85 226.15 230.35 211.45
227.95 229.30 225.25 201.25 216.10

Apply test for normality.

Apply chi-square test; apply it with at least two different choices for the number of intervals.
Do all the tests reach the same conclusion?

Using Minitab check the normality of the data.

Hypothesis testing
Ho: The random variable has a normal distribution
H1: The random variable does not have a normal distribution.
Jose Alejandro Bastidas B.

Since the p-value is greater than the level of significance, the null hypothesis is not rejected and it is
concluded that the data follow a normal distribution

For the other part, we graph ate the histogram.

Histograma de esfuerzos de rotura
Normal
Media 219,4
10
Desv.Est. 11,72
N 50

8
Frecuencia

0
200 210 220 230 240
esfuerzos de rotura

With n = 50 the number of classes that will be used will be k = 6

Class width = (max-min) / k = (243.1-198.4) /6=7.45
With mean μ = 219,356 and deviation σ = 11,71780714
Jose Alejandro Bastidas B.
Real Limit (x) Z= (x-μ)/σ P(x)

205,85 -1,152604736 P(x≤205.85) = 0,12453633

213,3 -0,516820249 P(x≤213.3) = 0,30264082
220,75 0,118964238 P(x≤220.75) = 0,54734816
228,2 0,754748725 P(x≤228.2) = 0,77480012
235,65 1,390533212 P(x≥235.65) = 0,082183511

̅x = 122.3
σ = 11.7
For interval 1 and 5 the probalities are:
Interval 1: P (x <207.34) = 0.153
Interval 2: P (x> 234.16) = 0.103

Interval Li Ls Oi Z- Z+ P(X) EI (Oi-Ei) (Oi-Ei)^2 (Oi-Ei)^2/Ei

1 198,4 207,34 10 -1,8 -1,0 0,153 7,665 2,335 5,450 0,711

2 207,34 216,28 11 -1,0 -0,3 0,244 12,218 -1,218 1,483 0,121
3 216,28 225,22 11 -0,3 0,5 0,295 14,752 -3,752 14,079 0,954
4 225,22 234,16 13 0,5 1,3 0,205 10,231 2,769 7,667 0,749
5 234,16 243,1 5 1,3 2,0 0,103 5,134 -0,134 0,018 0,003
1 50 2,536

X20 = 2.53
X20,05,2= 5.99
The degrees of freedoms are equal to 5-1-1 = 2
X20 < X 20.05,2 Is Not rejected
We conclude that the null hypothesis is not rejected, and it is concluded that the data follow a
normal distribution.

3. The Crosstowner was a bus that cut a diagonal path from northeast Atlanta to southwest Atlanta.
The time required to complete the route was recorded by the bus operator. The bus runs from
Monday through Friday. The times of the last fifth runs, in minutes, are as follows:

92.3 92.8 106.8 108.9 106.6

115.2 94.8 106.4 110.0 90.9
104.6 72.0 86.0 102.4 99.8
87.5 111.4 105.9 90.7 99.2
97.8 88.3 97.5 97.4 93.7
99.7 122.7 100.2 106.5 105.5
80.7 107.9 103.2 116.4 101.7
84.8 101.9 99.1 102.2 102.5
111.7 101.5 95.1 92.8 88.5
74.4 98.9 111.9 96.5 95.9
Jose Alejandro Bastidas B.
• Do a scatter plot

• Use Stat::Fit, or any other software, and do a histogram of the best distribution
Jose Alejandro Bastidas B.

• Use Stat: Fit, or any other software, and do a Goodness of fit test, and find the best distribution

Exponential
Distribution

Lognormal
Distribution
Jose Alejandro Bastidas B.

Normal
Distribution

Uniform
Distribution

We conclude that the best distribution is normal because of the value P we obtain.

• Use Stat: Fit, or any other software, and find the best distribution parameters

The parameters of the best distribution (normal) are μ = 99.222 and σ = 10.067