Scribd
Upload
83 views

Probability Distribution

Time dependent reliability functions • Times-to-failure obey a probability distribution • This probability value is a function of the time • Thus, the probability of failure or success of a component (i.e., reliability evolution) is a function of the time • In a system that is failed and is being repaired, times-to-repair are also distributed according to a probability distribution

Uploaded by

Jeff bales
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
83 views

Probability Distribution

Time dependent reliability functions • Times-to-failure obey a probability distribution • This probability value is a function of the time • Thus, the probability of failure or success of a component (i.e., reliability evolution) is a function of the time • In a system that is failed and is being repaired, times-to-repair are also distributed according to a probability distribution

Uploaded by

Jeff bales
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

LSU

Probability distributions in
reliability evaluation
Time dependent reliability functions
Times-to-failure obey a probability distribution

This probability value is a function of the time


Thus, the probability of failure or success of a component (i.e., reliability
evolution) is a function of the time
In a system that is failed and is being repaired, times-to-repair are also
distributed according to a probability distribution

Electrical & Computer Engineering Department

LSU

Terminology of distributions
In reliability evaluation, the random variable is frequently time.
If at = 0, the component or system is known to be operating then its
probability of failure at = 0 is zero.
As however, the probability of failure tends to unity as it is a
certainty that the component or system will fail given that the exposure
time to failure is long enough.

Electrical & Computer Engineering Department

LSU

Classroom activity
If at = 0, the component or system is known to be operating then its
probability of failure at = 0 is zero.
As however, the probability of failure tends to unity as it is a certainty
that the component or system will fail given that the exposure time to failure is
long enough.

The above characteristic is equivalent to what function in the


probability theory?

Electrical & Computer Engineering Department

LSU

Classroom activity
This characteristic is therefore equivalent to the cumulative distribution
function and is a measure of the probability of failure as a function of time
(or other random variable being considered).
the cumulative failure distribution function or more simply, the cumulative failure
distribution, designated

0 =0
: [0, ]
=1

Electrical & Computer Engineering Department

LSU

Terminology of distributions

The complement of () is known as the survivor function and designated as ()


() = 1 ()

The derivative of the cumulative distribution function of a continuous random variable gives the
probability density function.

In reliability evaluation, the failure density function, (), is defined as




=
=

Or

The probability of failure from = 0 to = 0

= 1
0

The total area under the failure density function must be unity and therefore () may also be
written as

Note: In the case of discrete random variables, the integrals are replaced by summations

Electrical & Computer Engineering Department

LSU

Terminology of distributions
failure density function.
(): probability of failure in time
(): probability of surviving beyond time

Electrical & Computer Engineering Department

10

LSU

Terminology of distributions
Hazard rate ()
The hazard rate is a measure of the rate at which failures occur in a given
period of time
The hazard rate is dependent on the number of failures in a given time period
and the number of components exposed to failure
In order to evaluate the hazard rate, the number of failures must be related per
unit to the number of components that are exposed to failure giving the
following definition of ()
o EX: the number of failures in a 100 component sample is less than that in a 1000
component sample if the same time period is considered and the components are
identical, yet the hazard rate should be the same

number of failures per unit time


=
number of components exposed to failure

Electrical & Computer Engineering Department

11

You might also like