Free tool
MTBF and MTTR Calculator: Check What Your Numbers Count
Your report says the case packer runs 364 hours between failures, and it still goes down every couple of weeks. The division is simple. What moves the number is which hours go on top and which stops count as failures. This calculator shows MTBF and MTTR as your report counts them and on a stated basis, and how far to trust a number built on a few failures.
Works on one asset or a group of identical assets. Free, and nothing you enter leaves the page.
Why use it
A good MTBF can come from how it’s counted
MTBF is hours divided by failures. Both halves depend on choices: which hours, which stops count, and where the repair clock starts and stops.
“MTBF went up in the quarter we ran short weeks.”
Calendar hours keep counting when the line doesn’t run.
Caught byTime base check
“MTTR looks fine, but the line sat half a shift waiting on a part.”
The repair clock leaves out waiting.
Caught byRepair clock check
“MTBF jumped after we changed how breakdowns get coded.”
Breakdowns logged as PM or adjustments drop out of the count.
Caught byFailure definition check
“One bad month swings the number all over the place.”
A few failures can’t pin MTBF down.
Caught bySample size check
If any of these sound familiar, the calculator shows how many hours each one is worth.
Who it’s for
Check the number before it sets a PM interval
Maintenance managers
Building the monthly KPI deck.
Reliability engineers
Before resetting a PM interval.
Plant managers
When the report shows an improvement nobody saw on the floor.
CI leads
Baselining a TPM pilot.
CMMS admins and planners
Setting up failure codes and malfunction start and end fields.
How it works
Three steps
Enter one period’s hours and events
Calendar, scheduled and planned stop hours, the breakdowns however they were logged, and repair time. Take them from one month or quarter of work orders.
Answer a few questions on how they’re counted
Which hours your report divides by, which stops it counts, where its repair clock starts and stops, and whether the asset is repaired or replaced. Not sure is always an option.
Take both numbers to your next maintenance review
You get two MTBF numbers, as reported and on a stated basis, the 90% range around the stated one, MTTR split into repair, waiting and restart, and availability both ways. On case packer 2: 364 h reported, 122 h stated, a range of 68 to 245 h, and 3.0 h of active repair inside 5.3 h of restoration. Copy it, email it or print it.
The method
Five checks on how your MTBF is counted
Each check tests one choice behind the number. The examples come from case packer 2, the sample asset in the tool below.
Time base
Which hours does your report divide by?
Passes whenMTBF divides by operating hours.
Why it mattersCalendar hours keep counting when the asset isn’t scheduled. IEC 60050-192 defines MTBF on operating time.
Passes974 operating hours.
FailsThe report divides 2,184 calendar hours.
Failure definition
Which stops count as failures?
Passes whenEvery stop where the asset couldn’t do its job counts, whatever the work order type.
Why it mattersCoding a breakdown as PM or an adjustment raises MTBF without changing the machine.
Passes6 breakdowns plus those 2 make 8.
FailsOne breakdown logged as PM and one repeat logged as an adjustment are left out.
Repair clock
Where does the repair clock start and stop?
Passes whenThe report says which clock it uses.
Why it mattersActive repair and time back in production can differ by more than double. ISO 14224 separates active repair time from down time.
Passes3.0 h active repair and 5.3 h restoration, labeled.
FailsA report MTTR from tech start to repair done, called downtime.
Sample size
How many failures sit behind the number?
Passes whenThere are enough failures to trust a single number.
Why it mattersWith 8 failures, the 90% range runs from 68 to 245 hours (NIST chi-square method).
PassesThe range is shown next to the number.
FailsAn MTBF built on 3 failures.
Repaired or replaced
Is the asset repaired or replaced when it fails?
Passes whenMTBF is used for repaired assets and MTTF for replaced parts.
Why it mattersIEC applies MTBF to repairable items only.
PassesCase packer 2 is repaired.
FailsA bearing’s life reported as MTBF.
Try it
Check an MTBF number
Open the example to see finished results, or enter one month or quarter from your own maintenance report.
Get started
Pick a starting point
Have your maintenance report and work order list for one period in front of you. Your numbers stay in this page and clear when you close it.
Common questions
What counts, and where the methods come from
Does MTBF include repair time?
No. IEC 60050-192 defines MTBF as the mean operating time between failures, so the hours an asset spends down for repair stay out. On case packer 2, the 42 hours of failure downtime come off the 1,040 scheduled hours before dividing by 8 failures.
Should I use calendar hours if I don’t track run hours?
Use scheduled hours minus planned stops and failure downtime: 974 hours on case packer 2. If calendar hours are all you have, label the result calendar MTBF so nobody compares it with an operating-time number. Calendar time gives the sample 364 h; operating time gives 122 h.
Does an operator reset count as a failure?
Most plants set a minimum duration below which a stop doesn’t count; this calculator starts at 10 minutes, and you can change it. Short stops belong in OEE as minor stops, the performance loss in Nakajima’s six big losses. Counting case packer 2’s 14 operator-cleared stops would drop its reported MTBF from 364 h to 109 h.
Do failures found at PM count?
A stop where the asset couldn’t do its job counts, whatever the work order says, in line with the IEC 60050-192 definition of failure as the loss of ability to perform as required. A PM that finds wear before the asset stops does not count. Case packer 2’s one breakdown logged as PM is worth 23 hours of reported MTBF.
Can MTTR include waiting for parts?
Restoration time does; active repair time doesn’t. ISO 14224 separates active repair time from down time, which also takes in waiting and restart. On case packer 2, active repair is 3.0 h per failure and restoration is 5.3 h, so say which one you report.
How many failures do I need?
Show the range with the number. The NIST/SEMATECH e-Handbook gives a chi-square method for it: with 8 failures, case packer 2’s 90% range runs from 68 to 245 hours. With fewer than 5 failures, treat any single MTBF as rough.
Can I combine identical machines?
Yes, when they’re alike and counted the same way. The NIST/SEMATECH e-Handbook pools operating time and failures across identical units: add the hours, add the failures, then divide. Leave out a machine with a different duty or different coding habits.
Does an MTBF of 500 hours mean it runs 500 hours?
No. Under a constant failure rate, about 63% of runs fail before reaching the MTBF, because the chance of lasting a full MTBF is about 37%. For case packer 2’s stated 122 h, expect most failures to come sooner than that.
What is a good MTBF?
Published targets disagree because they skip the definition: the same case packer reads 364 h or 122 h depending on which hours and stops are counted. Compare your own stated-basis trend quarter over quarter, with the range beside it.
MTBF or MTTF for bearings and belts?
MTTF. IEC 60050-192 uses MTBF for repairable items and mean time to failure for items that are replaced, so a bearing or belt you throw away gets MTTF. The machine it sits in, like case packer 2, keeps MTBF.
Inherent or operational availability?
Inherent availability uses active repair: MTBF ÷ (MTBF + active repair), 97.6% on case packer 2. The downtime version uses restoration, which adds waiting and restart: 95.9%. Label which one you report.
Site assessment
Want a second set of eyes on your maintenance numbers?
Our fixed-fee site assessment reviews your floor processes, records and reports and ends in a written findings report.
About the site assessment