Loading…
Loading…
Calculate MTBF from field data, or predict it from your bill of materials using MIL-HDBK-217F2 parts count — free, with no signup required to calculate. The parts-count calculator is free for up to 10 components; the full report and CSV export come with a free account.
Observed MTBF measures fielded equipment from accumulated operating hours and failures. Predicted MTBF estimates a design's reliability from its components before field data exists. This page provides a calculator for each.
MTBF = Total operating time / Number of failuresFailure rate = Number of failures / Total operating timeFailure rate = 1 / MTBF — under the constant-failure-rate assumptionFIT = Failure rate per hour × 1,000,000,000FIT = 1,000,000,000 / MTBF in hoursWorked example: a fleet accumulates 8,760 operating hours and records 2 failures. MTBF = 8,760 ÷ 2 = 4,380 hours (about half a year of continuous operation between failures); failure rate λ = 2 ÷ 8,760 ≈ 0.000228 per hour; FIT ≈ 228,311 failures per billion hours.
Before field data exists, MTBF is predicted from the design. The parts-count method of MIL-HDBK-217F Notice 2 assigns each component category a base failure rate, adjusted for quality level and operating environment, and sums them into the system failure rate — needing only the bill of materials, which makes it the standard early-design estimate.
Worked example: the bundled Controller-board example — 830 components in a Ground Fixed environment — predicts a system MTBF of 589,692 hours (λ ≈ 1.70 failures per million hours). Load it in the calculator above to see the per-line breakdown.
Parts count vs parts stress: parts count needs only the BOM; parts-stress analysis additionally uses each component's actual electrical and thermal operating conditions, giving a more accurate prediction once the detailed design exists. BQR's fiXtress performs stress-based prediction from the schematic.
| Observed MTBF (field data) | Predicted MTBF (MIL-HDBK-217F2) | |
|---|---|---|
| What it measures | Actual field performance to date | Expected reliability of a design |
| Inputs | Total operating hours and failure count | Bill of materials, quality levels, environment |
| When to use it | Equipment already operating in the field | During design, before field data exists |
| Method | MTBF = operating time ÷ failures | Per-component failure rates summed per the handbook |
| Tool on this page | Field-data calculator | BOM parts-count calculator |
MTBF applies to repairable systems; non-repairable items use MTTF instead. It is a population average under a constant-failure-rate assumption — not a guaranteed service life: a one-million-hour MTBF does not mean a unit lasts a million hours, but that one failure is expected per million accumulated unit-hours. Predictions compare designs and support planning; they are estimates, never guarantees of field performance.
MTBF (Mean Time Between Failures) is the average operating time between failures of a repairable system. It is a statistical measure of reliability: an MTBF of 100,000 hours means that across a large population, one failure is expected per 100,000 accumulated operating hours — not that any single unit will run that long.
FIT (Failures In Time) is the number of failures per billion device-hours. FIT = 1,000,000,000 ÷ MTBF in hours. Equivalently, FIT = failure rate per hour × 1,000,000,000. An MTBF of 1,000,000 hours equals 1,000 FIT.
Parts count fits early design, when the bill of materials is known but detailed electrical and thermal stresses are not. Once the design matures, parts-stress analysis — which uses each component's actual voltage, current and temperature conditions — gives a more accurate prediction. Parts count first, parts stress when the data exists.
The calculators are free and no account is needed to calculate: the field-data calculator is free without restriction, and the MIL-HDBK-217F2 parts-count calculator is free for up to 10 components. A free account is asked for only when downloading the full report or CSV export; paid plans remove the component limit.
Your projects and part-number libraries are saved to your own computer, not on BQR servers. BOM lines are processed to compute the prediction and are not retained; analytics record only anonymous usage events, never component data.
Go deeper: The Ultimate Guide to Calculating MTBF · Reliability Knowledge Hub · fiXtress — stress-based MTBF prediction