Reliability Engineering

(Concept)
 Introduction

 Reliability is the integral of the distribution of probabilities of

failure-free operation from the instant of switch-on to the first
failure
 The reliability of a component (or a system) is the
probability that the component (or the system) will not fail for
a time t
 Reliability is the probability that a device will operate without
failure for a given period of time under given operating
conditions
 Reliability is the probability of an item performing its
intended function over a given period of time under the
operating conditions encountered
 Introduction

4 elemen yang memegang peranan penting dalam mengkarakterisasikan

reliability (kehandalan) suatu item adalah :

1. Probability (Peluang)
Nilai reliability (kehandalan) suatu komponen / item / sistem
dinyatakan dalam bentuk peluang, yang nilainya bervariasi satu
dengan lainnya.

2. Intended function
Untuk menentukan kehandalan suatu komponen diperlukan suatu
kriteria yang jelas, apa fungsi dari komponen tersebut? pada
kondisi seperti apa komponen tersebut dinyatakan ‘fail’ ?
 Introduction

3. Time (waktu)
Reliability (kehandalan) merupakan fungsi waktu, oleh karena itu
definisi satuan waktu harus dipertimbangkan (cycle, jam, menit,
dll)

4. Operating condition
Kondisi operasi sangat mempengaruhi kinerja suatu item,
reliability tidak akan ada artinya jika tidak memasukkan unsur
kondisi operasi. Kondisi operasi tersebut dapat berbentuk
temperatur, kelembaban, dan lain-lain.
 Failure

 A deviation in the properties of the item from the prescribed

conditions is considered as a fault

 A state of fault is denoted by the term failure

 An item is considered to have failed under one of three

conditions :
1. When it becomes completely inoperable
2. When it still operable but is no longer able to perform its
intended function satisfactorily, or
3. When serious deterioration has made ir unreliable or unsafe
for continued use, thus necessitating its immediate removal
from service for repair or replacement
 Failure
The pattern of failure (pola kegagalan)

The bathtub curve

Failure
The six pattern of failure (6 pola kegagalan)
 Failure
The history of pattern of failure
Failure
 Reliability Measurements
Reliability Function :
Consider a test in which a large number of trials are made.
Each trial has an equal chance of resulting in event A. If A
does not result, however, then event B does.
In a large finite number of trials with X results of A and Y
results of B, the probabilities :
X Y
P( A)  P( B) 
X Y X Y
 Reliability Measurements
Reliability Function (cont.):
Consider a fixed number of components N0 under test.
After time t, with Ns(t) surviving and Nf (t) failing, the
probability of survival is

N s (t ) N s (t )
Psurvival    R (t )
N s (t )  N f (t ) N0
The probability of failure is
N f (t ) N f (t )
Pfailure    F (t )
N s (t )  N f (t ) N0
 Reliability Measurements
Reliability Function (cont.):

N s (t ) N s (t ) N 0  N f (t )
R(t )   
N s (t )  N f (t ) N0 N0
N f (t )
 1
N0
 1  F (t )

Reliability
Unreliability
function
function
 Reliability Measurements
Reliability Function (cont.) :

The reliability of a component (or a system) is the probability

that the component (or the system) will not fail for a time t

R(t )   f (t ) dt
t
 t
R(t )   f (t ) dt  1   f (t ) dt  1  F (t )
t 0

R(t) = reliability function

F(t) = unreliability function (cdf = cumulative distribution
function)
f(t) = probability density function (pdf)
 Reliability Measurements
Failure Rate :

• The rate at which failures occur in the interval 0 to t

• Total number of failures within an item population, divided

by the total number of life units (total operating time)
expended by that population during a particular
measurement interval under stated conditions
(Mil-Std-721)
 Reliability Measurements
Failure Rate (cont.) :

N s (t ) N s (t ) N 0  N f (t ) N f (t )
R(t )     1
N s (t )  N f (t ) N0 N0 N0
dR(t ) 1 dN f (t ) dN f (t ) dR(t )
   N0
dt N 0 dt dt dt
1 dN f (t ) N 0 dR(t ) N s (t )
 R(t ) 
N s (t ) dt N s (t ) dt N0

1 dR(t ) f (t ) # failures
λ(t)  (t )    
R(t ) dt R(t ) Total time
 Reliability Measurements
Failure Rate (cont.) :
Discrete cases :

f f = Number of failure in test time


T T = Total test time
 Reliability Measurements
Mean Time Between Failure (MTBF) / MTTF :

MTBF  m  E (t ) dF (t ) d [1  R(t )] dR(t )

f (t )   
 dt dt dt
  t f (t ) dt
0

  
dR(t )
MTBF   t dt   t R(t ) dt   R(t ) dt
0
dt 0 0

Discrete cases :

T 1
m 
 f
 Reliability Measurements
Example :
A light bulb manufacture wants to know the performance
of their products. 500 light bulb were tested. After 300
hours, 20 of them were failed.
Calculate:
Reliability, Failure Rate, MTBF,
 Distribution in Reliability Engineering

Exponential Distribution
Probability density function (pdf) :

f (t )   e t t 0
Cumulative distribution function (cdf) / Unreliability function :
F (t )  Q(t )  1  e  t t 0
Reliability function :
R(t )  e  t t 0
Failure rate : MTBF :
1
 (t )   E (t ) 

 Distribution in Reliability Engineering

Exponential Distribution (cont.)

 Distribution in Reliability Engineering

Weibull Distribution
Probability density function (pdf) :
 1
   t    t  
f (t )      exp     ;t 0
         
Cumulative distribution function (cdf) / Unreliability function :
  t  
F (t )  1  exp     ; t 0
    

Reliability function :
  t   (n)   e  x x n1 dx
R(t )  exp     0
    
Failure rate : MTBF :
   t 
 1
1 
 (t )      MTBF     1
      
 Distribution in Reliability Engineering

Weibull Distribution (cont.)

β=3
η = 50

β = 0.5

β=3
η = 100
β=3

β=3
β=1 η = 200
 Distribution in Reliability Engineering

Weibull Distribution (cont.)

β=3
β=3

β = 0.5 β = 0.5

β=1
β=1
 Distribution in Reliability Engineering

Normal Distribution
Probability density function (pdf) :
1 
 1t   
2

f (t )  exp     ;  t 
 2 
 2    

Reliability function :
 
1  1  t   2 
R(t )   f (t ) dt   exp     dt
t t  2  2    
 Distribution in Reliability Engineering

Normal Distribution (cont.)

Failure rate :
1  1  t   2 
exp    
f (t )  2  
2   
 (t )  
R(t ) 1  1  t   2 
t  2 exp  2     dt
 

MTBF :
E (t )  
 Distribution in Reliability Engineering

Normal Distribution (cont.)