Scribd
Upload
75 views3 pages

Weibull Analysis: Reliability, Failure Rate, Mean Time To Failure) That Are Derived From The

The Weibull distribution is commonly used in reliability engineering due to its versatility in modeling different failure patterns. Weibull analysis involves fitting life data to the Weibull probability density function to estimate the shape (β) and scale (η) parameters. These parameters and the location parameter (γ) are then used to derive useful reliability metrics like reliability, failure rate, and mean time to failure. The shape parameter β provides insights into the failure mechanism, with different values of β indicating infant mortality, random failures, early wearout, or rapid wearout.

Uploaded by

Hassani Mohamed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
75 views3 pages

Weibull Analysis: Reliability, Failure Rate, Mean Time To Failure) That Are Derived From The

The Weibull distribution is commonly used in reliability engineering due to its versatility in modeling different failure patterns. Weibull analysis involves fitting life data to the Weibull probability density function to estimate the shape (β) and scale (η) parameters. These parameters and the location parameter (γ) are then used to derive useful reliability metrics like reliability, failure rate, and mean time to failure. The shape parameter β provides insights into the failure mechanism, with different values of β indicating infant mortality, random failures, early wearout, or rapid wearout.

Uploaded by

Hassani Mohamed
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Weibull Analysis

The Weibull distribution is one of the most commonly used distributions in Reliability
Engineering because of the many shapes it attains for various values of . Weibull analysis
continues to gain in popularity for reliability work, particularly in the area of mechanical
reliability, due to its inherent versatility. The Weibull probability density function (Failure/Time
Distribution) is given by

t 1 (
f(t) = (
) e

t
)

t = Time to Failure

= Shape parameter
= Scale parameter
= Location parameter (locates the distribution along the abscissa. When the distribution starts at t=0, or at
the origin, = 0 )

The useful metrics (Reliability, Failure Rate, Mean Time To Failure) that are derived from the
above f(t) function are shown in Page 2

Data Requirements
Life data that is relevant to the failure mode is
critical

Individual Data
Total Units

Number of observed failures

Item Time to Failure

Grouped Data

Total Units

Number of groups

Failures in Group

Group End Time

Plotting Procedures

Order data from lowest to highest failure time

Estimate percent failing before each failure


time (median ranks)

Draw best line fit through data points plotted


on Weibull paper

Estimate Weibull parameters and from the


graph

Note: Parameters & can be easily derived


from a Weibull Software package, without
going through the above procedure
Example

A manufacturer estimates that its customers will operate a product, a portable A.C. Power Generator for 3000 hours per year, on
average. The company wants to sell the Generator with a 1-year warranty, but needs to estimate the percent of returns that will be
experienced in order to assess the warranty cost. Also wants to determine the Mean Time To Failure of the Generator. The
manufacturer authorizes a test program using 10 random samples of the product.
Times ( Hours) to failure for 10 samples are: 1:18200, 2:9750, 3:6000, 4:10075, 5:15000, 6:5005, 7:13025, 8:9500, 9:15050,
10:7000
Analysis Results are in Page 2; Using the Plot & Equations in Page 2 (1) Determine % of expected failures in the warranty period
using the Plot and the R(t) Equation. (2) Determine the Mean Time To Failure
Answers: (1) 3%; (2) 10929 Hours

Hilaire Ananda Perera ( http://www.linkedin.com/in/hilaireperera )


Long Term Quality Assurance

Use Of Weibull Analysis

= 2.58
= 12310 Hours

Q(t) = Probability of Failure as a percentage

t
( )
Reliability Function

Failure Rate Function

R(t) =

= Scale Parameter

t 1
(t) = ( )

Mean Time To Failure m = (

= Shape Parameter

t = Operating Time
( -1 +1) is the Gamma
Function evaluated at the
value of ( -1 +1)

+ 1)

NOTE: Mean Time To Failure is the inverse of Failure Rate only when = 1

Hilaire Ananda Perera ( http://www.linkedin.com/in/hilaireperera )


Long Term Quality Assurance

Interpreting The Weibull Plot


Value of

Product
Characteristics

<1

Implies infant
mortality. If
product survives
infant mortality, its
resistance to failure
improves with age

Inadequate stress screening or Burn-In


Quality problems in components
Quality problems in manufacturing
Rework/refurbishment problems

=1

Implies failures are


random. An old
part is just as good
(or bad) as a new
part

Maintenance/human errors
Failures are inherent, not induced
Mixture of failure modes

Implies early
wearout

Low cycle fatigue


Corrosion or erosion failure modes
Scheduled replacement may be cost effective

Implies old age


(rapid) wearout

Inherent material property limitations


Gross manufacturing process problems
Small variability in manufacturing or material

>1&<4

>4

If this occurs, suspect the following

Ref: RAC Reliability Toolkit

Hilaire Ananda Perera ( http://www.linkedin.com/in/hilaireperera )


Long Term Quality Assurance

You might also like