Scribd
Upload
92 views

Reliability Statistic - Normal Distribution

The document discusses probability concepts and distributions that are commonly used in reliability analysis. It defines key probability terms and notation. It then explains several common reliability distributions - normal, lognormal, exponential, Weibull - and provides examples of how to calculate probabilities and reliability given distribution parameters. Applications discussed include calculating reliability given failure time distributions and specifications.

Uploaded by

Iwan Akur
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
92 views

Reliability Statistic - Normal Distribution

The document discusses probability concepts and distributions that are commonly used in reliability analysis. It defines key probability terms and notation. It then explains several common reliability distributions - normal, lognormal, exponential, Weibull - and provides examples of how to calculate probabilities and reliability given distribution parameters. Applications discussed include calculating reliability given failure time distributions and specifications.

Uploaded by

Iwan Akur
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 12

Chapter 2 1

• Safety Moment
• Reliability Math
Logic 2

Rules of Probability
1 The probability of obtaining an outcome A is denoted by P(A), and so on for other outcomes.
2 The joint probability that A and B occur is denoted by P(AB).
3 The probability that A or B occurs is denoted by P(A + B).
4 The conditional probability of obtaining outcome A, given that B has occurred, is denoted by P (A
| B).
5 The probability of the complement, that is, of A not occurring, is P(A¯ ) = 1 – P(A)
6 If (and only if) events A and B are independent, then and
that is, P(A) is unrelated to whether or not B occurs, and vice versa.)
Reliability 101 - Definition 3

 Reliability Statistic
• discrete functions, continuous functions and point processes.
For example, a switch may either work or not work when selected or a pressure
vessel
may pass or fail a test—these situations are described by discrete functions
Continuous functions describe those situations which are governed by a
continuous variable, such as time or distance travelled. The electronic
equipment mentioned above would have a reliability function in this class.
Data Distribution 4

DATA

Continuous Discrete

Normal Exponential Weibull Binomial Poisson

Log Normal Gamma

𝑋𝑋 2
Normal Distribution 5

Examples;
- Baby born weight
- Measurement Result
- People Ages

Sign;
- Hazzard Rate increased
- Curve is simetric
Normal Distribution Examples 6

Find P(X > 13) ~> Find P(Z > 1,67)


µ=11 𝑥𝑥 − 𝜇𝜇 13−11 2
𝑧𝑧 = 𝑧𝑧 = = =1.67
𝜎𝜎 1.2 1.2
σ=1.2
Area under curve is 0,95254 , Therefor P (Z > 1,67) = 1 – 0,95254 =
Norm Distribution, Cont 7

Find P(X < 8) ~> Find P(X < 1,67)


µ=11 𝑥𝑥 − 𝜇𝜇 8 −11 −3
𝑧𝑧 = 𝑧𝑧 = = =− 2.5
𝜎𝜎 1.2 1.2
σ=1.2

Probability P < Area under for z -2.5 is 0,00621,


Therefor P (Z < - 2,5) =0,00621
Norm Distribution, Cont 8

Proportion of wires that will survive a pull of at least .5 g but not over 13.2 g
Change to mathematic form, P (8.5 < X < 13.2)
µ=11 𝑥𝑥 − 𝜇𝜇
𝑧𝑧 =
σ=1.2 𝜎𝜎
Solve P (X > 8.5 );
8.5 −11 − 2.5
𝑧𝑧 = = =− 2.083 >> 0.01876
1.2 1.2

Solve P (X < 13.2);


13.2 −11 2.2
𝑧𝑧 = = = 1.833 >> 0.96638
1.2 1.2

Probability P < Area under for z -2.5 is 0,00621,


Therefor P (-2.083 < Z < 1.833) = 0,094762
Lognormal Distribution 9

Examples;
- Vehicle mileage per year,
- Count switch operation,
- Repair time of a maintained system
- Fatigue Failure
- Economic Data, Maint. Cost

Sign;
- Hazzard Rate increased from zero then decrease
- Curve is right skewed
Lognormal Distribution Application 10

A manufacturer produces parts that have a median width of 2.0 cm. The population distribution is lognormal with
a shape parameter (σ) of 0.2. If the specification width for acceptable parts is 1.0 to 3.0 cm, what fraction of the parts
is rejectable?
1. Transform to statistical form, σ=0.2, P(1.0 < x < 3.0 )
2. Transform to lognormal formula,

ln 𝑥𝑥 −µ′
𝑧𝑧 =
Lognormal Distribution Application 11

A population of components, when tested at high laboratory stresses, fail according to a


lognormal distribution with 50=5000 and σ = 0.7. What percent of reliability expected on a 2000-
hour test ?

• Transform to statistical form, failure at F(2000 < x)


• Transform to lognormal formula, Known T50 µ = 5000, Then Ln T (u) = u’ = Ln 5000
• Transform to lognormal formula, Known x = 2000, Then Ln X = X’ = Ln 2000
• Known σ’ = 0,7
𝐿𝐿𝐿𝐿 2000 −𝐿𝐿𝐿𝐿 5000 𝐿𝐿𝐿𝐿 2000 −𝐿𝐿𝐿𝐿 5000 7.6 −8.52
• Z= = = = −1.31
0,7 0,7 0,7
• Find Z Value for -1.31 found 0.09510, > Failure probability
• Reliability = 1 – F(T) = 1-0.09510=0.9049
• Therefor expected for 2000 hour test about 90,49%

https://www.youtube.com/watch?v=rLv1k_PxUOE
Lognormal Distribution Application (examples) 12

It has been observed that gun tube occur according to the


distribution with log mean µ’ = 7 and log standard deviation σ’ = 2.
a. Find the reliability for a 200 round mission
b. What is the MTTF and standard deviation
(Source MIL-HDBK-338B)

• Transform to statistical form, P(200< x)


• Transform to lognormal formula, Known µ’ = 7 and σ’ = 2
𝑥𝑥𝑥−𝜇𝜇′ 𝐿𝐿𝐿𝐿 𝑥𝑥 −𝜇𝜇′ 𝐿𝐿𝐿𝐿 200 −7 5,298 −7 −1,7
• F 𝑧𝑧 = = = = = = −0,85
𝜎𝜎𝜎 𝜎𝜎𝜎 2 2 2
• Find Z Value for -0.85 found 0.19766,
• Reliability = 1 – F(T) = 1-0.19766=0.80234
• Therefor for a 200 round mission about 80,23%

https://www.youtube.com/watch?v=rLv1k_PxUOE

You might also like