Reference no: EM132182109
Question 1. (Understanding Data) Attached to the exam is a CSV file titled:
{IE365_FE_Q1.csv}.
This data is from Valles Global Fireworks Division. In the data, we are testing 400 iterations of burn time (in seconds) of the new Arias 3000 bottlerocket.
Develop an appropriate histogram. What kind of distribution might this process follow? Hint - something is wrong in the data
- what might be happening in the samples? Additional Hint - I didn't randomize the data and placed the data in order of testing(aka a time series).
Question 2. (Statistics) The Leo and Saija Fireworks Company is developing a new Green line of fireworks in response to demand for reduction in greenhouse gases produced. Therefore, they are testing two new accelerants that they hope creates less CO2. Attached is a CSV file titled:
CSV_FE_Q2.csv
This file has results of testing of CO2 created (in mmol) from the original accelerant (a), the new Green accelerant(b), and the Super-Green accelerant(c).
Develop an appropriate inferential test and use the 5 paragraph format to discuss results. Hint: Contrast!!
What can Saija and Leo conclude based on their tests (given a 5% α)? Please use precise terms involving Ho and Ha.
Question 3. (Cumulative Sum Chart) The data in the following table are maximum temperature readings from accelerant testing (in 0F ) for a new firework (the Juan Diaz Cherry Bomb). The target value is µ0 = 950 Estimate the process standard deviation and set up a tabular CUSUM chart for this process using standardized values h = 5 and k = 1 . Interpret the chart. Note:
Read values down from left.
Question 4. (Process Capabilities): Pines Industries is making a Roman Candle firework. The firework consists of an accelerant fuse that is inserted into a casing (similar to a shaft and bearing). The internal diameter of the accelerant (shaft) is normally distributed and has sample mean of µˆ1 = 1.480cm and a sample standard deviation of σˆ1 = 0.015cm. The external diameter of the casing (bearing) is normally distributed and has sample mean of µˆ1 = 1.500cm and a sample standard deviation of σˆ1 = 0.01cm.
What percent of items will fail (where the external diameter of the accelerant is larger than the internal diameter of the casing)?
Question 5. (Machine Learning): In less than 300 words, describe how machine learning could be used for fault detection in sequence or time series data.
Extra Credit. (Multivariate Control Chart): The SohnCo firework scientists are developing a new Chrysanthemum shell that burns red then blue. The two quality characteristics are the powder amount (in mg) for the red burn and the powder amount (in mg) for the blue burn. Based on samples of size n=25 and assuming the mean values of the quality characteristics and the covariance matrix were computed from 50 preliminary samples:
x- = [55]
[30]
S = [200 130]
[130 120]
Sample Number |
x¯1(Red) |
x¯2(Blue) |
1
|
58 |
32
|
2
|
60 |
33
|
3
|
50 |
27
|
4
|
54 |
31
|
5
|
63 |
38
|
6
|
53 |
30
|
7
|
42 |
20
|
8
|
55 |
31
|
9
|
46 |
25
|
10
|
50 |
29
|
11
|
49 |
27
|
12
|
57 |
30
|
13
|
58 |
34
|
14
|
75 |
45
|
15
|
55 |
27
|
Construct a T2 control chart using these data. Use phase II limits.
Attachment:- Assignment.rar