3/5/2024

Reliability Engineering
The Failure Distribution (Chapter 2)

The Failure Distribution

• The Reliability Function

• Mean Time To Failure
• Hazard Rate Function
• Bathtub Curve
• Conditional Reliability

2
 3/5/2024

2.1 The Reliability Function

• Reliability is defined as the probability that a system (component) will

function over some time period t
Let T = a random variable, the time to failure of a component
R(t) is the probability that the time to failure is greater than or equal to t
R(t)= Pr{T  t}
where R(t )  0 , R(0)  1, and
lim t  R (t )  0

Often called the SURVIVAL FUNCTION

3

2.1 The Reliability Function

F(t) is the probability that a failure occurs before time t:

F(t)= 1- R(t)= Pr {T < t}

where F (0)  0 and limt  F (t )  1

R(t) = the reliability function

F(t) = the cumulative distribution function (CDF) of the
failure distribution
4

4
 3/5/2024

The Reliability Function

• f(t) = Probability Density Function (PDF):

d F (t) d R (t)
f(t) = = -
dt dt

f (t )  0 and 
0
f (t )dt  1

PDF of the Failure Distribution f(t):

6
 3/5/2024

CDF of the Failure Distribution F(t):

Reliability Function R(t)

8
 3/5/2024

The Reliability Function

• CDF of the failure distribution F(t):

F(t) =  0 f(t ) dt '

t

• Reliability Function R(t):


R(t) =  t f(t ) dt '
where f(t) is Probability Density Function (PDF):

Remember: F(t) = 1- R(t), 0 ≤ R(t) ≤ 1, and 0 ≤ F(t) ≤ 1

9

Finding Failure Probabilities

Pr{a  T  b} = F(b) - F(a)

= R(a) - R(b)

a b

10

10
 3/5/2024

Example

The passive components of a distribution system for natural gas

has the following reliability function:

t2
R (t )  1  ; 0  t  10 yrs
100
Find: a. R(3 yrs)
b. The CDF, F(t)
c. Pr{1<T<3}
d. The probability density function, f(t)
11

11

Example

32 t2
a . R ( 3)  1   .9 1 R (t )  1  ; 0  t  10 yrs
100 100

b. F(t) = 1 - [1 - t2 / 100] = t2 / 100

c. Pr{1<T<3} = F(3) - F(1) =. 09-.01 =. 08

f(t)
dF (t ) 2t t
d. f (t )    ; 0  t  10
dt 100 50
10 t

12

12
 3/5/2024

Example 2.1
T: the time (in operating hours) to failure of a compressor,

a. What is the reliability for a 100 hr operating life?

b. Find the design life if a reliability of 0.95 is desired?
A design life is defined to be the time to failure tR that
corresponds to a specified reliability R, that is R(tR) = R.
c. What is the probability that the failure occurs within 10 to 100 hrs?
13

13

2.2 Time to Failure

Mean Time To Failure (MTTF)

• The mean of the failure distribution is only one of several measures of
central tendency of the failure distribution

MTTF = E(T)

14

14
 3/5/2024

2.2 Time to Failure

• Median time to failure:

= 0.5
• The median divides the distribution into two halves, with 50 percent
of the failures occurring before the median time to failure and 50
percent occurring after the median

15

15

2.2 Time to Failure

• Third frequently used average is the mode, or most likely observed

failure time, defined by:

f( t mode ) = max f(t)

0 t<

• For a small fixed interval of time centered around the mode, the
probability of failure will generally be greater than for an interval of
the same size located elsewhere within the distribution

16

16
 3/5/2024

2.2 Time to Failure

17

17

2.2 Time to Failure

• One measure that is often used to further describe a failure

distribution is its variance σ2,
• Variance represents an average squared distance a failure time will be
from the MTTF
• defined by:

= −

= −
18

18
 3/5/2024

Example

• For the distribution system, find the MTTF, Median, and Mode

19

19

Example 2.2

Plot function: http://www.fooplot.com

20

20
 3/5/2024

2.3 Hazard Rate Function

• In addition to the probability functions defined earlier the failure rate
or hazard rate function is often used in reliability
• hazard rate function provides an instantaneous (at time t) rate of
failure

• Hazard Rate Function or Failure Rate Function λ(t):

21

21

2.3 Hazard Rate Function

• Hazard Rate Function or Failure Rate Function λ(t):
( )
=
( )

= −

• Failure rates can be characterized as:

• Increasing Failure Rate (IFR) when λ(t) increasing
• Decreasing Failure Rate (DFR) when λ(t) decreasing
• Constant Failure Rate (CFR) when λ(t) increasing constant
22

22
 3/5/2024

Hazard Rate Function

• The cumulative failure rate over a period of time t is defined by:

= = − ln ( )

• Average failure rate between two times t1 and t2:

( ) − ( ) ln − ln ( )
1, 2 = =
− −

ln 0 − ln ( ) − ln ( ) ( )
= = =
−0
23

23

Hazard Rate Function

• Average Failure rates can be as:

• Increasing Failure Rate Average (IFRA) when AFR(t) increasing
• Decreasing Failure Rate (DFRA) when AFR (t) decreasing

24

24
 3/5/2024

Example

• For the distribution system, find failure rate function (hazard)?

Cumulative failure rate function after t period of time?
And average failure rate function after t period of time?

25

25

Example 2.5

• Given the linear hazard rate function

where t is measured in operating hours, what is the design life if a 0.98

reliability is desired?
What is the cumulative failure rate function?
What is the average failure rate function?

26

26
 3/5/2024

2.4 Bathtub Curve

27

27

28

28
 3/5/2024

Human Mortality Curve

age Male Female
0 0.007644 0.006275
5 0.000202 0.000152 Probability Male Death
10 0.00011 0.000113
16 0.00081 0.000375
0.024
17 0.000964 0.000423
20 0.00129 0.000456 0.019
25 0.001379 0.000499
30 0.001389 0.000628 0.014
35 0.00177 0.000953
40 0.002589 0.001514 0.009
45 0.003891 0.002264
50 0.005643 0.003227 0.004
55 0.008106 0.004884
60 0.012405 0.007732 -0.001
65 0.019102 0.012199 0 10 20 30 40 50 60 70
70 0.029824 0.019312 Age
75 0.046499 0.030582
80 0.073269 0.050396
85 0.120186 0.086443
90 0.192615 0.147616 29

29

2.5 Conditional Reliability

• Conditional reliability describes the reliability of a component or
system following a burn-in period To or after a warranty period To
• We define conditional reliability as the reliability of a system given
that it has operated for time To :

30

30
 3/5/2024

Residual MTTF

Residual MTTF (after To):

Where t’ = t + To

31

31

Example

• For the distribution system, calculate the reliability of 5 years given

that the burn-in period is 1 year? And calculate the MTTF after the
burn-in period?
t2
R (t )  1  ; 0  t  10 yrs
100

32

32
 3/5/2024

Example
t2
R(t)  1
100
R ( t  1) 1  ( t  1) 2 / 100
R ( t |1)  
R (1) 1  (1 / 100)
100  ( t  1) 2
  1.01  ( t  1) 2 / 99
100 (.99 )

Therefore: R(5|1) = 1.01-36/99=.646

where R(5) = 1 - .25 = .75
10
1 10 t2 1  t3 
1  T02 /100 T0
MTTF (T0 )  1  dt  t 
100 .99  300 T 1
0

1  10 3
1  5.67
or MTTF (1)  10  1    5.72 yr.
.99  300 300  .99
33

33

Example:

DFR:
.
0.5
=
1000 1000

Calculate the reliability of 11 year?

If 1 year consumed in the burn-in process, what is the reliability of 11
years?

34

34