Managing Risk for Better Life

Class 4 – Weibull Distribution

Functions & Its Application

Ahmad Taufik, M.Eng, Ph.D

Risk and Reliability Specialist

Reliability Engineering Training Course

 Part I
Weibull Functions & Reliability Parameters

Reliability Engineering Training Course

 Weibull Distribution Functions
 The purpose of this class is to introduce the Weibull Function to
estimate the remaining life of component(s) or equipment.
 It is used by the manufacture to setup replacement time schedule and it
is also used by the user to setup maintenance time schedule
(preventive maintenance plan).
 Weibull Function (Reliability Function and Probability of Failure Function
– CDF) is a powerful method and has a vast application in industry for
spare-part management of electronics, instrument, electrical,
mechanical /rotating equipment components.

Reliability Engineering Training Course

 Comparison & Application of
Various Distribution Functions

Exponential Normal Lognormal Weibull

Distribution Distribution Distribution Distribution
Function Function Function Function
o Life Distribution o Life Distribution o Size Distribution o Corrosion
of Complex non of High Stress of Breaks & Resistance
repairable Component Leaks (Pipes) o Life Distribution
system o Stress – o Life Distribution of Many Basic
o Life Distribution Strength of Some Components
of Burn-in of Analysis Electronics (Capacitors,
Some o Tolerance Parts Relays Ball
Components Analysis Bearings,
Turbine Blades,
and Certain
Motors)
Reliability Engineering Training Course
 Weibull Distribution Functions

The family of Weibull distributions was introduced by the Swedish

physicist Waloddi Weibull in 1939; his 1951 article “A Statistical
Distribution Function of Wide Applicability” (J. of Applied Mechanics, vol.
18: 293–297) discusses a number of applications. .

Definition

A random variable X is said to have a Weibull distribution with

parameters  and  ( > 0,  > 0) if the pdf of X is

Reliability Engineering Training Course

 Weibull Distribution Functions

 Waloddi Weibull, a Swedish inventor and engineer invented the

Weibull distribution in 1937. The U.S. Air Force recognized the
merit of Weibull’s methods and funded his research to 1975.
 Leonard Johnson at General Motors improved Weibull’s methods.
He suggested the use of median rank values for plotting.
 The engineers at Pratt & Whitney found that the Weibull method
worked well with extremely small samples, even for 2 or 3 failures
 Weibull Reliability Functions has wide range of many engineering
application. The CDF characterizes electronic, electrical and
mechanical equipment at the component levels

Reliability Engineering Training Course

Probability Functions Representing Reliability

1. Reliability Function, R(t)

2. Cumulative Distribution Function (CDF) – F(t)
3. Probability Density Function (PDF) – f(t)
4. Hazard Function – h(t)
5. Relationships Among R(t), F(t), f(t), and h(t)

7
 Example of Plots of R(t), F(t), f(t), h(t) for Normal Distribution

R(t) f(t)

8 F(t) h(t)
 Relationships Among R(t), F(t), f(t), and h(t)

f(t) R(t) F(t) h(t)

 t f (t )
 

f(t) - f (u ) du f (u )du
t 0  f (u)du
t

dR (t ) dR (t )
 - 1  R (t ) 
R(t) R (t )dt
dt
dF (t ) dF (t )
F(t) 1  F (t ) - 1  F (t )  dt
dt

t t

e 0 
t

h(t) h(t ) e 0
 h ( u ) du h ( u ) du
1 e 0
h ( u ) du
-

Notes:
The matrix shows that any of the three other probability functions (given by the
columns) can by found if one of the functions (given by the rows) is known.
9
 CFR, IFR, DFR
↓
Hazard Function
 1
 t
h(t )    , t  0,   0,   0
↓   
Reliability Function
  t  
R (t )  exp     
↓     
Cumulative Distribution Function

  t  
↓ F (t )  1  exp     
    
Probability Density Function
 1
t   t  
f (t )    exp     
       
Weibull Distribution Function

Reliability Engineering Training Course

Weibull Distribution Function

Reliability Engineering Training Course

Weibull Distribution

 Probability Functions
 Characteristics of Weibull Distribution
 MTTF, Variance, Median and Mode
 Conditional Reliability

Reliability Engineering Training Course

 Characteristics of Weibull Distribution

 Weibull distribution can attain many shapes for various values

of shape parameter β.
 It can model a great variety of data and life characteristics,
including constant, increasing, and decreasing failure rates
(CFR, IFR, and DFR).
 Therefore, it is one of the most widely used lifetime
distributions in reliability engineering.

Two-Parameter Weibull Distribution:

 β = shape parameter, or slope parameter
 θ = scale parameter, or characteristic life parameter

Reliability Engineering Training Course

 Characteristics of Weibull Distribution

 Characteristic Effects of Shape (Slope) Parameter, β

 Effects of β on Weibull Hazard Function
(CFR, IFR, DFR depending on the value of β )
 Effects of β on Weibull PDF
 Effects of β on Weibull Reliability Function

 Characteristic Effects of Scale Parameter, θ

Reliability Engineering Training Course

Versatility of Weibull Model
 1
t
h(t )    , t  0,   0,   0
  

Reliability Engineering Training Course

Effects of β on Weibull Hazard Function

 1
t
h(t )  , t  0,   0,   0
   

Reliability Engineering Training Course

Effects of β on Weibull PDF

 1
t   t  
f (t )    exp     
       

Reliability Engineering Training Course

Effects of β on Weibull Reliability Function

  t  
R (t )  exp     
    

Reliability Engineering Training Course

Bathtub Curve
Typical information for components
Infant of auseful
Typical PC Likelihood of Failure
Component Mortality life period
Before Wearout
Rate (years)
Power Supply Low 3-6 Moderate

Motherboard Moderate 4-7 Low

Processor Low 7+ Very Low
System Moderate to
7+ Very Low
Memory High
Low to
Video Card 5-7 Low
Moderate
Low to
Monitor 5-7+ Moderate to High
Moderate
Moderate to
Hard Disk Drive 3-5 Moderate to High
High
Floppy Disk
Low 7+ Low
Drive
CD-ROM Drive Moderate 3-5 Moderate
Modem Low 5-7+ Low
Keyboard Very Low 3-5 Moderate
Mouse Very Low 1-4 Moderate to High
20
 Effects of β on Weibull Reliability Function


 t 
R (t )  exp     
    
 0 < β < 1, R(t) decreases sharply and monotonically, and
is convex.

 β = 1, R(t) decreases monotonically but less sharply than

for 0 < β < 1, and is convex.

 β > 1, R(t) decreases as t increases. As wear-out sets in,

the curve goes through an inflection point and decreases
sharply.
Reliability Engineering Training Course
 Scale Parameter, θ

 The scale parameter influences both the mean and the spread of
the Weibull distribution.

 All Weibull reliability functions pass through the point (θ, 0.368).
    
R ( )  exp       e 1  0.368
    
 θ is also called characteristic life.
 θ has the same units as T, such as hours, miles, cycles,
actuations, etc.

CDF

Reliability Engineering Training Course

Example: Q-Q of Weibull Distribution and Weibull Fit
(works well)
 T~Weibull (1, 4000) Generate 50 data points

1-e-1
0.632

h
 Effects of θ on Weibull PDF


 1
t   t  
 f (t )  exp     
        

 Increasing the value of θ has

the effect of stretching out the
 pdf
 influences both mean and
spread
  Since the area under a pdf
curve is a constant value of
one, the “peak'' of the pdf
curve will also decrease with
the increase of θ

Reliability Engineering Training Course

 Part II
Example of Weibull Problems

Reliability Engineering Training Course

Example 1
A mining industry has a large industrial dump truck experiences
failures that seem to be well approximated by a two parameter
Weibull distribution with
(1) Characterized the failure process based on the values of θ
and β.
  18, 000 km and   2.7
(2) What is the 10,000 km reliability?
10000 2.7
( )
R (10000 )  e 18000
 0.815

(3) What is the 24,000 km reliability?

24000 2.7
( )
R(24000)  e 18000
 0.1137
(4) What is the B10 life? (km at which 10% of the population will
fail, or 90% reliability is desired.)

R( β α )  1  α 100
t
( ) 2.7
R( β 10 )  0.9  e 18000
 β 10  7821.7

(5) If the Weibull slope can be changed to   1.7 by changing

the gear tooth design, how will this affect the above answers?

R(10000)  0.6920  0.815

R(24000)  0.1957  0.1137

Reliability Engineering Training Course

MTTF and Variance
 1
MTTF    1  
 
  2   1   
2

     1       1    
2 2

         

where ( x) is the gamma function:


( x)   y e dy x 1  y
0

To obtain the gamma function, use Standard Table in the textbook, and

( x)  ( x  1)( x  1), x  0
( x)  ( x  1)!, x  integer > 0
Reliability Engineering Training Course
Reliability Engineering Training Course
 Median and Mode

 Design Life and Bα Life

(tR / )
R(tR )  e  R  tR   ( ln R)1/ 

 ( B /  ) 
R( B )  e  1   / 100
B    ln(1   / 100) 
1/ 


 Median R( B50 )  0.5  e  ( B50 /  ) 


B50  tmed   (ln 2)1/ 

 Mode  (1  1/  )1/  for   1

tmode 
 0 for   1
Example 2
Given a Weibull distribution with a characteristic life of
127,000 hours and a slope of 3.74, find the mean and
standard deviation. Also find the probability of surviving
(reliability) the mean life.
1
Solution MTTF  θ Γ (1  )

1 1
A  Γ (1  )   (1  )   (1 . 27 )  0 . 9025
 3 . 74
2 2
B  Γ (1  )   (1  )   (1 . 53 )  0 . 88757
 3 . 74
 MTTF  127000 ( 0 . 9025 )  114617 . 5
 2
  2(B  A2)     34382.5

Reliability Engineering Training Course

Example 2
If for the dump truck Example 1, we have two designs, A and B.
Find the mean and standard deviation of the life.

Solution

A   18, 000 km and   2.7 MTTF 16007.57

  6388.294
B   18, 000 km and   1.7
MTTF 16063.74
  9752.76

Reliability Engineering Training Course

 Burn-In Concept in Manufacture
 Early failures (infant mortality) – common “reliability” problem, esp.
in electronic equipment  usually caused by manufacturing
“defects” quality problems. Problems are typically common in
“new” products and may disappear as technology matures.
 Ideal  build-in Q&R upfront and reduce such problems – but
hard to do with complex technology and pressure to reduce
product development cycle time.
 To achieve reliability goals and reduce field-failure, common
practice to “burn-in” components and systems to screen out units
that would fail early – esp. important in safety-critical applications
 can be viewed as a type of 100% inspection.
 Done at use condition or low-level stress environment to avoid
undue aging of components and systems.
 Burn-in is expensive  incorporate costs and benefits and decide
on optimal trade-off
Reliability Engineering Training Course
 Part III
Mechanical Application
of Weibull Functions Plot

Reliability Engineering Training Course

 Graphical Model Validation

 Weibull Plot
  t  
F (t )  1  R(t )  1  exp     
  h  
1
 ln ln   ln t   ln h is linear function of ln (time).
1  F (t )

• Estimate Fˆ (ti ) at ti using Bernard’s Formula (Median Rank Method)

For n observed failure time data (t1 , t2 ,..., ti ,...tn )

Median Rank Method ˆ i  0.3

F (ti ) 
n  0.4
Reliability Engineering Training Course
 Plotting The Weibull Graph Procedures

1. Use Available Weibull Plot

Paper
2. Plot Data Points Failure
Probability (Y) as Linear
Function of X
3. Perform Linear Regression
X on Y
4. Find Shape Factor β and
Characteristics Life θ on
the Plot
5. Determine Reliability, R(t)
6. Determine MTBF Value

Reliability Engineering Training Course

Reliability Engineering Training Course
 Correlation of MTBF and θ for each particular β

Note :
The MTBF can be determined when θ and β have been calculated first

Reliability Engineering Training Course

Median Rank, 5% Rank and 95% Rank Confident Level

Reliability Engineering Training Course

Reliability Engineering Training Course
 ITEM Beta Values Eta Values
(Weibull Shape Factor) (Weibull Characteristic Life--hours)

Example Weibull
Low Typical High Low Typical High
Parameters for
My engineering judgment does not
Variety of Machinery support the range of a few values shown
Equipment Components in published databases

Machinery Equipment

taken from : Circuit breakers 0.5 1.5 3 67,000 100,000 1,400,000

Heinz P. Bloch in Compressors, centrifugal 0.5 1.9 3 20,000 60,000 120,000
Weibull Data Base Compressor blades 0.5 2.5 3 400,000 800,000 1,500,000
http://www.barringer1.com
Compressor vanes 0.5 3 4 500,000 1,000,000 2,000,000
Diaphragm couplings 0.5 2 4 125,000 300,000 600,000
See this home page for Gas turbine comp.
other item (electronics, blades/vanes 1.2 2.5 6.6 10,000 250,000 300,000
instruments, static, Gas turbine blades/vanes 0.9 1.6 2.7 10,000 125,000 160,000
equipment, lubrication, Motors, AC 0.5 1.2 3 1,000 100,000 200,000
Seal, etc)
Motors, DC 0.5 1.2 3 100 50,000 100,000
Pumps, centrifugal 0.5 1.2 3 1,000 35,000 125,000
Steam turbines 0.5 1.7 3 11,000 65,000 170,000
Steam turbine blades 0.5 2.5 3 400,000 800,000 1,500,000
Steam turbine vanes 0.5 3 3 500,000 900,000 1,800,000
Transformers 0.5 1.1 3 14,000 200,000 14,200,000
Reliability Engineering Training Course
 Concept
BallofBearing
10% Life (L10) – Bearing Life
Replacement

L10 Life = Time related to 10% Probability of Failure or 90% Reliability

Reliability Engineering Training Course

 Mechanical Application of Weibull Function

1. Failure Rate and Reliability of PSV

2. Heat Exchangers Inspection & Maintenance
3. Turbine Blade Maintenance
4. Ball Bearing Replacement (CASE STUDY)

Reliability Engineering Training Course

1. Failure Rate and Reliability of PSV

Reliability Engineering Training Course

1. Failure of PSV

Reliability Engineering Training Course

2. Failure of Heat Exchanger

Reliability Engineering Training Course

Inspection & Maintenance Plan

Reliability Engineering Training Course

3. Turbine Blades Maintenance

Reliability Engineering Training Course

Reliability Engineering Training Course
Reliability Engineering Training Course
Reliability Engineering Training Course
Reliability Engineering Training Course
Reliability Engineering Training Course
4. Ball Bearing Replacement – CASE STUDY

Reliability Engineering Training Course

 4. Ball Bearing Replacement Time Plan

No Failure (hr) The Failure Rate of 10 Ball Bearing is

given in the Table.
1 460
2 950 1. Construct CDF Vs Time Curve
3 1120 2. Determine the Shape Factor, β
4 1980
3. Determine the Scale Factor, θ
5 3250
4. Calculate the R(t) and F(t) at t =
6 5050
10,000 hr
7 8700
5. Calculate the MTBF
8 12000
6. Calculate B10 or L10
9 15500
7. Define Maintenance Plan
10 17590

Reliability Engineering Training Course