MTBF (Mean Time Between Failures) Calculation

[kctrinh]

Please help!

I used this equation Fail Rate= # of failure/total running time.Therefore, MTBF is 1/fail rate.

The total running time= unit that does not fail running time + running time of the fail unit.

Question: Let say that one of my unit failed, we repair it then it will continue to run again. How do one input as a running time for this failed unit? Are you suppose to include only upto the time it failed or even after it is fixed and ran again?

Best regards,

Kenny

[dna_leri]

Kenny,

In my opinion, you count the time before and after the repair.

If you only made one part and it first failed after 5 mins, you repaired it, and it fails again after 7 mins. At that point the MTBF (or average time between failures) is 6 mins.

See: http://www.vicr.com/documents/quality/Rel_MTBF.pdf
for a fuller explanation of the difference between MTBF and MTTF (Mean time to failure).

[Jim Wynne]

Originally Posted by kctrinh

Please help!

I used this equation Fail Rate= # of failure/total running time.Therefore, MTBF is 1/fail rate.

The total running time= unit that does not fail running time + running time of the fail unit.

Question: Let say that one of my unit failed, we repair it then it will continue to run again. How do one input as a running time for this failed unit? Are you suppose to include only upto the time it failed or even after it is fixed and ran again?

Best regards,

Kenny

Mean time between failures gives a good overview of MTBF and the calculations for it.

[Randy Stewart]

The other part of your OEE is the Mean Time To Repair. When you decide to start the counter on either of them is something that should be understood by Maintenance, Production and yourself.
Normally, MTBF is from "start-up" to fail, MTTR may start at Maintenance notification and run until they return the machine to production (after prove out of the repair). I have seen some engineers track MTTR until the "Okay to Run" was issued from QA. I have also seen where scheduled downtime was not computed in MTBF because the machine wasn't running.
Decide what data is going to help. When to start or stop isn't the main issue. Getting data and being able to use it to make you better is.

[pinpin]

Originally Posted by Randy Stewart

The other part of your OEE is the Mean Time To Repair. When you decide to start the counter on either of them is something that should be understood by Maintenance, Production and yourself.
Normally, MTBF is from "start-up" to fail, MTTR may start at Maintenance notification and run until they return the machine to production (after prove out of the repair). I have seen some engineers track MTTR until the "Okay to Run" was issued from QA. I have also seen where scheduled downtime was not computed in MTBF because the machine wasn't running.
Decide what data is going to help. When to start or stop isn't the main issue. Getting data and being able to use it to make you better is.

Dear Randy,

TS clause 7.3.3.2 requires manufacturing process design output shall incluide "data for quality, reliability, maintainaility and measurability" and clause 8.2.3.1 also talks about "process capability, reliability, maintainability and availability". Can you give example of these metrics in relation to manufacturing process ?