Calculating Failure Rate: 0.8 Reliability in 2000 Hours

  • Thread starter Thread starter scotchpie
  • Start date Start date
  • Tags Tags
    Failure Rate
Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
3 replies · 12K views
scotchpie
Messages
2
Reaction score
0
I am currently studying for a business qualification I am stubling over a question.

It concerns calculating failure rate. The equation I know for this needs the number of failures over the total time.

However the question is: "A device has a reliability of 0.8 over 2000 hours, calculate the failure rate."

How can one firstly work of the number of failures from this? It doesn't give any other information, ie total sample size etc. just the reliability over 2000 hours.

Thanks
Andrew
 
Engineering news on Phys.org
I'm pretty sure 0.8 is the number of failures. To me it seems like the question gives the answer, but hen I guess I'm not sure what they are looking for. Wik has some info: http://en.wikipedia.org/wiki/Failure_rate
In words appearing in an experiment, the failure rate can be defined as

The total number of failures within an item population, divided by the total time expended by that population, during a particular measurement interval under stated conditions. (MacDiarmid, et al.)
So the population size would be 1, time is 2000 hours, and number of failures is .8.
 
Last edited:
I believe reliability means 'rate of success', i.e. it works reliably 0.8 or 80% of the time. So 1 is all the time, and 1-0.8=0.2 would be unreliability, which is related to the failure rate.

Look on the page for MIL-HDBK-217F
Reliability Prediction of Electronic Equipment- Notice F
http://www.weibull.com/knowledge/milhdbk.htm#200

One can download the pdf file.

Weibull.com is a good resource for statistics and statistical analysis.
 
Last edited:
"A device has a reliability of 0.8 over 2000 hours, calculate the failure rate."

Just to make this simple for you place it in your math formula like it says.

0.8/2000 0.8 over 2000 hours. The answer of the efficiency is very low, but it's the truth according to the math. .0004 per hour is the answer according to the information furnished. Gee, slave labor does better than an efficiency rating like that, and it sounds like someone needs to pay someone better wages to get the job done. The machine is malfunctioning - tell whoever asked the question to fix the machine or up their wages to fair.

For those of you who do not recognize significant numbers, in its stated form this is not a percentage. If someone meant it that way, their distinction is incorrect.

What is the reliabillaty per hour is what I think the equation boils down to.

Then stated as a failure rate in 20,000 attempts you only get 8 successes... that being the case expressed as a percentage, failure rate is 20,000/8 which is 2500% failure rate.
 
Last edited: