Scribd
Upload
9 views

BA 14- Case Study Scenario[1]

The case study analyzes the relationship between machine speed and defect counts in a textile manufacturing factory. It includes a statistical analysis of defect probabilities and a hypothesis test to determine if the average defect count meets buyer requirements. The findings indicate that there is an 18% chance of defects and the average defect rate is statistically acceptable based on the hypothesis test results.

Uploaded by

Md.Tipu Sultan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views

BA 14- Case Study Scenario[1]

The case study analyzes the relationship between machine speed and defect counts in a textile manufacturing factory. It includes a statistical analysis of defect probabilities and a hypothesis test to determine if the average defect count meets buyer requirements. The findings indicate that there is an 18% chance of defects and the average defect rate is statistically acceptable based on the hypothesis test results.

Uploaded by

Md.Tipu Sultan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

CASE STUDY SCENARIO

A textile manufacturing factory produces 30 rolls(fabric) per day. Each


roll is processed using automated machines, and machine speed (in
meters per minute) is suspected to impact the defect count in each roll.
Management wants to analyze whether the current defect rates are within
acceptable limits as per buyer’s requirement. And wants to know whether
there’s actually a correlation between machine speed and defect count.
You have been hired as a Data Analyst to present an analysis on this.

Question 1: In which method you will select 10 sample? Justify your claim?

Roll_ID Machine_Speed Defect_Count

1 295.2 1

2 310.4 12

3 298.7 0

4 287.6 0

5 305.1 2

6 299.3 1

7 315.6 1

8 290.9 0

9 301.5 1

10 307.2 2

11 296.8 0

12 312.1 3

13 293.4 0

14 304.6 0

15 299.8 1
16 285.3 0

17 309.5 2

18 291.7 0

19 300 1

20 311 1

21 297.5 0

22 308.2 2

23 294 0

24 289.5 0

25 303.7 2

26 299 0

27 313.3 3

28 288.1 0

29 306.8 2

30 292.2 0

Question 2: Imagine you’re a quality engineer at that textile mill where, each fabric
roll can suffer two kinds of flaws: a color defect with probability P(C)=0.12, and a
texture defect with probability P(T)=0.09, and occasionally both occur together with
probability P(C and T)= 0.03. Assuming the defect events on any one roll are
independent of those on another, what is the probability that a randomly selected roll
will have at least one defect (i.e., either a color or a texture flaw)?
Question 3: Let’s say you select 10 sample from the roll. Your buyer’s requirement
is that on average not more than 1 defect count can be present in the rolls. Do a
hypothesis test for this? Let’s assume you selected the following rolls. (Consider,
level of significance= 5%, For Z=0.612, p-value is 0.542)
Roll No Defect Count
04 0
05 2
06 1
09 1
10 2
12 3
16 0
19 1
24 0
25 2

Answer 2:
“Either defect” on one roll

Use the Addition Rule for non-mutually exclusive events:

P(C or T) = P(C) + P(T) – P (C and T)


= 0.12+0.09−0.03
= 0. 18

So there’s an 18 % chance a randomly chosen roll has at least one of


these defects.

Answer 3: Textile Roll Defect Rate


Hypothesis Test
Formulate: Hypotheses:

• Null Hypothesis (H₀): μ = 1 (The average defect rate is


acceptable)
• Alternative Hypothesis (H₁): μ > 1 (The average defect rate is
greater than acceptable range)

Sample Data:
Sample Size, n = 10
Sample: [0, 2, 1, 1, 2, 3, 0, 1, 0, 2]
Population Mean (μ) = 1 (Assumed acceptable defect rate)

Step 1: Sample Mean (x̄)


x̄ = (0 + 2 + 1 + 1 + 2 + 3 + 0 + 1 + 0 + 2) / 10
= 12 / 10
= 1.2
Step 2: Sample Standard Deviation (s)

√Σ(xi − x̄)²
s= n−1
√9.6
=
9
=1.033

xᵢ xᵢ − x̄ (xᵢ − x̄)²
0 -1.20 1.44

2 0.80 0.64
1 -0.20 0.04

1 -0.20 0.04

2 0.80 0.64
3 1.80 3.24

0 -1.20 1.44

1 -0.20 0.04

0 -1.20 1.44

2 0.80 0.64
Sum of Squares = 9.6
Step 3: Z-Score Calculation
(x̄ − μ)
Z = (s / √n)
(1.2− 1)
=(1.033 / √10)

= 0.612
Step 4: p-value Calculation
Using Z = 0.612 and standard normal table (one-tailed):
p-value ≈ 0.542

Decision Rule:

Since the p-value (0.542) > 0.05, we fail to reject the null hypothesis.

The average defect rate is statistically not greater from the


acceptable rate at 5 percent level of significance. So our defect rate
is acceptable as per buyer’s requirement.

You might also like