+ Reply to Thread
Results 1 to 4 of 4

Thread: Converting instantaneous Reliability at time (t) to normalized failure rate (fpmh)

  1. #1
    Points: 465, Level: 9
    Level completed: 30%, Points required for next Level: 35

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Converting instantaneous Reliability at time (t) to normalized failure rate (fpmh)




    Given the reliability at 2100 hours =0.992, then lambda(2100 hours) = -Ln(0.992)=0.008. I need to convert the hazard rate (lambda) for 2100 hours to the normalized value in terms of failure rate per million hours (fpmh, 10^6 hours).
    I am foggy on the process from this point; should I use the pdf function to convert such as:
    F(t)=1-(1+0.008*10^6)^-1
    Then take the derivative f(t)=0.008 (1+0.008*10^6)^-2
    Then h(t): f(t)/1-F(t)

    Can someone point me in the right direction?

  2. #2
    Super Moderator
    Points: 10,218, Level: 67
    Level completed: 43%, Points required for next Level: 232
    Dragan's Avatar
    Location
    Illinois, US
    Posts
    1,838
    Thanks
    0
    Thanked 156 Times in 139 Posts

    Re: Converting instantaneous Reliability at time (t) to normalized failure rate (fpmh

    Quote Originally Posted by halfdome View Post
    Given the reliability at 2100 hours =0.992, then lambda(2100 hours) = -Ln(0.992)=0.008. I need to convert the hazard rate (lambda) for 2100 hours to the normalized value in terms of failure rate per million hours (fpmh, 10^6 hours).
    I am foggy on the process from this point; should I use the pdf function to convert such as:
    F(t)=1-(1+0.008*10^6)^-1
    Then take the derivative f(t)=0.008 (1+0.008*10^6)^-2
    Then h(t): f(t)/1-F(t)

    Can someone point me in the right direction?
    To begin, you're not going to find your answer by taking the derivative of a constant (i.e. the variable t does not appear on the right-hand side of F(t)), which is what you have presented above. I am assuming you have a constant failure rate and thus the density function is exponential, I suspect. In short, you need to get your notation correct so that you can get an appropriate response.

  3. #3
    Points: 465, Level: 9
    Level completed: 30%, Points required for next Level: 35

    Posts
    2
    Thanks
    0
    Thanked 0 Times in 0 Posts

    Re: Converting instantaneous Reliability at time (t) to normalized failure rate (fpmh

    Thank you for your comment. The equations for F(t) and f(t) were taken from "Applied Reliability", Tobias and Trindade, pg. 35, Example 2.3 Failure Rate Calculations as method to determine the cummulative hazard rate function h(t). The 0.992 value for reliability is taken from an exponential graph of reliability vs. time resulting from product life testing.

  4. #4
    TS Contributor
    Points: 18,828, Level: 86
    Level completed: 96%, Points required for next Level: 22

    Posts
    2,826
    Thanks
    10
    Thanked 517 Times in 492 Posts

    Re: Converting instantaneous Reliability at time (t) to normalized failure rate (fpmh


    For the notation issue, I think you can also refer to

    http://en.wikipedia.org/wiki/Survival_analysis

    which has some fundamental relationship between F, f, h

    Here you need to be more specific to say whether you are interested in the instantaneous hazard rate or the cumulative hazard. (supposed to be different)

    The second thing is that we are not sure about the parametric form of any one of F, f, h. If you can type the exact parametric formula for any one of them before doing any numerical substitution, that will be much better.

    The most important thing is that we are not sure, say if you are concerning changing from 1 unit/scale to another unit/scale, or you are given some information and want to know how to make use of the information and the relationship I quoted above to find the desired (cumulative) hazard.

+ Reply to Thread

           




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts






Advertise on Talk Stats