VIETNAM NATIONAL UNIVERSITY

HO CHI MINH CITY UNIVERSITY OF TECHNOLOGY

FACULTY OF APPLIED SCIENCE

Probability and Engineering Applications (MT2015)

Project 8: Materials Engineering Statistical

Characterization

Instructor: Dr. Cris Laquibla Armada

Class: TT03
Name: Lê Trung Nam
Student ID: 2451114
Probability and Engineering App. Student ID: 2451114

1 Theoretical Background: Normal Distribution

1.1 General Normal Distribution

A random variable X is normally distributed with mean µ and variance σ 2 if its Probability Density
Function (PDF) is given by:

1 −(x−µ)2
f (x) = √ e 2σ2 , −∞ < x < ∞ (1)
2πσ

Notation: X ∼ N (µ, σ 2 ), where E(X) = µ and V (X) = σ 2 .

1.2 Standard Normal Distribution

A standard normal random variable, denoted as Z, is a normal variable with µ = 0 and σ 2 = 1. Its
Cumulative Distribution Function (CDF) is denoted as:

Φ(z) = P (Z ≤ z) (2)

Values for Φ(z) are typically found in standard statistical tables (Appendix Table III).

1.3 Standardization Formula

To calculate probabilities for any general normal variable X, we transform it into Z using the standard-
ization formula:
X −µ
Z= (3)
σ
The probability P (X ≤ x) is obtained by finding the corresponding z-value:
 
X −µ x−µ
P (X ≤ x) = P ≤ = P (Z ≤ z) (4)
σ σ

x−µ
where z = σ .

2 Illustrative Example: Shaft Diameter

Problem Statement: The diameter of a shaft in an optical storage drive is normally distributed with
mean µ = 0.2508 inch and standard deviation σ = 0.0005 inch. The specifications are 0.2500 ± 0.0015
inch.

Part (a): Calculation for Current Process

Range Calculation
First, we determine the specific upper and lower limits of the specification range:

• Lower Limit: 0.2500 − 0.0015 = 0.2485 inch

Ho Chi Minh City University of Technology Page 1 of 7

Probability and Engineering App. Student ID: 2451114

• Upper Limit: 0.2500 + 0.0015 = 0.2515 inch

Thus, we need to find the probability P (0.2485 < X < 0.2515).

Standardization
X−µ
We transform the variable X to the standard normal variable Z using Z = σ :

• For x1 = 0.2485:
0.2485 − 0.2508 −0.0023
z1 = = = −4.6
0.0005 0.0005

• For x2 = 0.2515:
0.2515 − 0.2508 0.0007
z2 = = = 1.4
0.0005 0.0005

Table Lookup and Calculation

The probability is the area under the curve between these two Z-values:

P (0.2485 < X < 0.2515) = P (−4.6 < Z < 1.4)

= Φ(1.4) − Φ(−4.6)

Using the Standard Normal Table:

• Φ(1.4) = 0.91924

• Φ(−4.6) ≈ 0.0000

P = 0.91924 − 0.0000 = 0.91924

Conclusion
Approximately 91.92% of the manufactured shafts conform to the required specifications.

Part (b): Optimization (Re-centering)

Analysis
If the process mean is adjusted to the target µnew = 0.2500:

• Lower Z: z1 = 0.2485−0.2500
0.0005 = −0.0015
0.0005 = −3

• Upper Z: z2 = 0.2515−0.2500
0.0005 = 0.0015
0.0005 =3

Calculation

P (−3 < Z < 3) = Φ(3) − Φ(−3)

From the Z-table:

Φ(3) = 0.99865, Φ(−3) = 0.00135

P = 0.99865 − 0.00135 = 0.9973

Ho Chi Minh City University of Technology Page 2 of 7

Probability and Engineering App. Student ID: 2451114

Conclusion
By recentering the process mean, the yield increases significantly to 99.73%.

3 Problem Set Solutions

3.1 Problem 1: Uniform Distribution (Coating Thickness Scan)

Given: The waiting time W is uniformly distributed over 15 minutes.

1
W ∼ Uniform(0, 15) =⇒ f (w) = for 0 ≤ w ≤ 15
15

Part (a): Find P (W < 4)

Formula
Z x
P (W < x) = f (w) dw
0

Substitution
Z 4
1
P (W < 4) = dw
0 15

Calculation

h w i4 4
P (W < 4) = = − 0 ≈ 0.2667
15 0 15

Conclusion
There is a 26.67% probability that the waiting time is less than 4 minutes.

Part (b): Find P (W > 10)

Formula
Z 15
P (W > x) = f (w) dw
x

Substitution
Z 15
1
P (W > 10) = dw
10 15

Calculation

h w i15 15 − 10 5 1
P (W > 10) = = = = ≈ 0.3333
15 10 15 15 3

Conclusion

Ho Chi Minh City University of Technology Page 3 of 7

Probability and Engineering App. Student ID: 2451114

There is a 33.33% probability that the waiting time is greater than 10 minutes.

3.2 Problem 2: Normal Distribution (Coating Thickness)

Given: The thickness X ∼ N (µ, σ 2 ) with µ = 50, σ = 5.

Part (a): Find P (X < 45)

Standardization

X −µ 45 − 50 −5
z= = = = −1
σ 5 5

Table Lookup

P (X < 45) = P (Z < −1) = Φ(−1)

From the Standard Normal Table: Φ(−1) = 0.1587.

Conclusion
The probability that a randomly selected plate has a thickness less than 45 µm is 15.87%.

Part (b): Find P (48 ≤ X ≤ 56)

Standardization
We calculate Z-scores for both boundaries:

• Lower: z1 = 48−50
5 = −2
5 = −0.4

• Upper: z2 = 56−50
5 = 6
5 = 1.2

Table Lookup and Calculation

P (48 ≤ X ≤ 56) = P (−0.4 ≤ Z ≤ 1.2) = Φ(1.2) − Φ(−0.4)

Using the table values:

Φ(1.2) = 0.8849, Φ(−0.4) = 0.3446

P = 0.8849 − 0.3446 = 0.5403

Conclusion
The probability that the thickness lies between 48 µm and 56 µm is 54.03%.

3.3 Problem 3: Exponential Distribution (Time to Corrosion)

Given: Mean µ = 12 years. Rate parameter λ = 1

µ = 1
12 . CDF: P (X ≤ x) = 1 − e−λx .

Ho Chi Minh City University of Technology Page 4 of 7

Probability and Engineering App. Student ID: 2451114

Part (a): Probability perforation occurs within 8 years

Substitution

1 2
P (X ≤ 8) = 1 − e− 12 ·8 = 1 − e− 3

Calculation

P (X ≤ 8) ≈ 1 − e−0.6667 ≈ 1 − 0.5134 = 0.4866

Conclusion
The probability that corrosion perforation occurs within 8 years is 48.66%.

Part (b): Probability pipe lasts more than 15 years

Formula

P (X > 15) = 1 − P (X ≤ 15) = e−λ·15

Substitution

15
P (X > 15) = e− 12 = e−1.25

Calculation

P (X > 15) ≈ 0.2865

Conclusion
The probability that the pipe lasts more than 15 years without perforation is 28.65%.

3.4 Problem 4: Central Limit Theorem (Average Hardness)

Given: Population mean µ = 210, σ = 15. Sample size n = 25.

Part (a): Approximate P (X̄ < 205)

Standard Error Calculation

σ 15 15
σX̄ = √ = √ = =3
n 25 5

Standardization

X̄ − µ 205 − 210 −5
Z= = = ≈ −1.67
σX̄ 3 3

Ho Chi Minh City University of Technology Page 5 of 7

Probability and Engineering App. Student ID: 2451114

Table Lookup

P (X̄ < 205) = P (Z < −1.67) = Φ(−1.67)

From the Z-table: Φ(−1.67) = 0.0475.

Conclusion
The probability that the sample mean hardness is less than 205 HV is approximately 4.75%.

Part (b): Approximate P (205 ≤ X̄ ≤ 215)

Standardization

• Lower Z: z1 = 205−210
3 ≈ −1.67

• Upper Z: z2 = 215−210
3 ≈ 1.67

Calculation

P (205 ≤ X̄ ≤ 215) = Φ(1.67) − Φ(−1.67)

P = 0.9525 − 0.0475 = 0.9050

Conclusion
There is a 90.50% probability that the sample mean falls between 205 and 215 HV.

3.5 Problem 5: Confidence Intervals (Mean Hardness of Alloy)

Given: n = 20, x̄ = 225, s = 14. Use t-distribution (df = 19).

Part (a): Construct a 95% Confidence Interval

Critical Value Lookup

For 95% confidence, α = 0.05 =⇒ α/2 = 0.025.

t0.025,19 = 2.093

Margin of Error Calculation

s 14
E = tα/2 · √ = 2.093 · √
n 20
E = 2.093 · 3.1305 ≈ 6.55

Interval Calculation
We calculate the lower and upper bounds:

Ho Chi Minh City University of Technology Page 6 of 7

Probability and Engineering App. Student ID: 2451114

• Lower Bound: x̄ − E = 225 − 6.55 = 218.45

• Upper Bound: x̄ + E = 225 + 6.55 = 231.55

Conclusion
The 95% confidence interval for the true mean hardness is (218.45, 231.55) HV.

Part (b): Construct a 90% Confidence Interval

Critical Value Lookup

For 90% confidence, α = 0.10 =⇒ α/2 = 0.05.

t0.05,19 = 1.729

Margin of Error Calculation

14
E = 1.729 · √ = 1.729 · 3.1305 ≈ 5.41
20

Interval Calculation

• Lower Bound: 225 − 5.41 = 219.59

• Upper Bound: 225 + 5.41 = 230.41

Conclusion
The 90% confidence interval for the true mean hardness is (219.59, 230.41) HV.

Ho Chi Minh City University of Technology Page 7 of 7