How to Calculate MTTR (Mean Time to Repair), With Worked Examples

The repair that took six hours — and the three that took one

The hydraulic press goes down on a Tuesday morning. Your technician spends two hours hunting for the fault, thirty minutes sourcing a fitting from the storeroom, another hour waiting for the line pressure to bleed off safely, and then forty-five minutes on the actual repair. Call it four and a half hours of downtime — for a job that, once he knew the root cause, was under an hour of hands-on work.

That gap between what a repair takes and what it should take is what Mean Time to Repair measures — and closing it is one of the clearest levers you have on both downtime cost and equipment availability.

By the end of this guide you will know exactly how to calculate MTTR, how to interpret the number against your fleet, and how MTTR connects to the broader reliability picture — including availability and downtime cost per hour.

---

What MTTR actually measures

MTTR (mean time to repair) is the average elapsed time from the moment a failure is detected to the moment the asset is returned to full operating condition — including diagnosis, parts retrieval, the physical repair, and any testing before sign-off.

It is one of the three core reliability metrics every maintenance team should be tracking alongside MTBF and OEE. Where MTBF tells you how long equipment runs between failures, MTTR tells you how quickly you recover when it doesn't.

MTTR = Total Repair Time ÷ Number of Repairs

"Total repair time" is the sum of every minute the asset was out of service across all repair events in your measurement period. "Number of repairs" is the count of those events — not the count of parts replaced or work orders written, but distinct failure-and-restore cycles.

MTTR does not include planned preventive maintenance downtime. If you scheduled a 4-hour PM window on Saturday, that window does not go into the MTTR calculation. MTTR is a metric for unplanned corrective events only.

---

The MTTR formula in practice: two worked examples

Example 1 — single asset over one quarter

A packaging line conveyor experienced five unplanned failures in Q1. Your technician logged the following repair durations:

Repair event	Repair time (hours)
Jan 8 — belt slippage	1.5
Jan 29 — encoder fault	3.0
Feb 14 — bearing seizure	6.0
Mar 3 — motor contactor	2.0
Mar 21 — belt slippage (repeat)	1.0
Total	13.5

MTTR = 13.5 hours ÷ 5 repairs = 2.7 hours per repair

That 2.7-hour average is the baseline you manage against going forward. The bearing seizure on February 14 — a 6-hour event — stands out immediately. If you can understand why that repair took four times longer than the belt jobs (Was the bearing hard to access? Was the replacement part not stocked? Did diagnosis take two hours?), you have a specific improvement target.

Example 2 — fleet rollup across three assets

Suppose you are tracking MTTR across three CNC machines over a 6-month period:

Asset	Total repair time (hrs)	Number of repairs	MTTR (hrs)
CNC-01	8.0	4	2.0
CNC-02	18.0	3	6.0
CNC-03	5.0	5	1.0
Fleet total	31.0	12	2.6

Fleet MTTR = 31.0 hours ÷ 12 repairs = 2.6 hours per repair

CNC-02 deserves your attention: three failures, but an average of 6 hours each. That pattern suggests a systemic diagnosis or parts-availability issue — possibly a component that is not stocked, or a failure mode that is not yet well-understood by the team. CNC-03, by contrast, fails more often (5 events) but recovers in an hour on average, which points to well-understood, easily addressed faults. For CNC-03, the priority is reducing failure frequency (an MTBF problem); for CNC-02, the priority is cutting recovery time (an MTTR problem). The distinction matters for where you invest.

---

Why MTTR drives downtime cost directly

Every hour of MTTR has a price. The exact cost depends on your operation — line rate, product value, labor standing by — but the arithmetic is straightforward:

Downtime cost per event = MTTR (hrs) × Downtime cost per hour ($)

If a production line costs your operation $8,000 per hour in lost throughput and standing labor, and your MTTR on that line averages 3 hours, each unplanned failure costs roughly $24,000 before parts or overtime.

Research benchmarks give context to the scale of these costs. Aberdeen Research (via ReliaMag, 2026) put average unplanned downtime cost at $260,000 per hour across sectors; discrete manufacturers typically fall in the $10,000–$50,000 per hour range (ReliaMag citing Aberdeen and Siemens, 2024). Your operation may be well below those figures — but even at $5,000 per hour, a 3-hour MTTR on a key asset represents a $15,000 event, every time it happens.

That is why reducing MTTR by even 30–45 minutes per event compounds quickly across a fleet. For a step-by-step method for calculating your specific downtime cost per hour, see the downtime cost calculation guide.

---

What drives MTTR up — and how to address each factor

MTTR is not a single variable. It is the sum of several elapsed-time components, each with its own cause and fix:

1. Detection and notification time. How long between the failure occurring and someone knowing about it? Manual rounds create gaps. Dedicated condition checks or operator-reported fault logs tighten detection.

2. Diagnosis time. The largest variable in most unplanned events. A technician who has seen the failure before resolves it in minutes; a novel fault can take hours. Good failure history logs — recording what failed, what caused it, and what fixed it — reduce diagnosis time on repeat events. This is one of the concrete reasons a persistent PM history log (rather than a spreadsheet that gets overwritten) pays for itself.

3. Parts availability. Industry data suggests that 20%–30% of downtime duration is tied to parts availability (Oxmaint, 2026). A bearing sitting in the storeroom turns a 4-hour repair into a 90-minute one. This makes MTTR a direct input to stocking decisions.

4. Repair execution time. The actual hands-on work. Affected by technician skill, access to the asset, quality of documentation, and tools on hand. Standard repair procedures (SOPs) reduce variation here.

5. Testing and sign-off. Often underestimated. Some assets require a run-up sequence, pressure test, or quality check before returning to production. Build this into your logged repair time, or your MTTR will appear artificially low.

---

MTTR and availability: the direct connection

MTTR pairs with MTBF (mean time between failures) to calculate equipment availability — the percentage of scheduled time an asset is actually able to run:

Availability = MTBF ÷ (MTBF + MTTR)

For a worked example: if an asset has an MTBF of 480 hours and an MTTR of 6 hours:

Availability = 480 ÷ (480 + 6) = 480 ÷ 486 = 98.8%

Now suppose you cut MTTR from 6 hours to 2 hours through better parts stocking and documented SOPs:

Availability = 480 ÷ (480 + 2) = 480 ÷ 482 = 99.6%

The MTBF did not change — the asset failed just as often — but availability improved by 0.8 percentage points purely from faster recovery. At high production volumes, that 0.8 points represents real throughput. For a fuller treatment of how availability feeds into OEE, see the availability calculation guide.

---

Setting an MTTR target for your fleet

MTTR targets are not universal — they depend on asset criticality, the complexity of your failure modes, and how your maintenance team is resourced. A general framework:

• Establish your baseline first. You cannot set a meaningful target without knowing where you are. Pull repair logs for the past 6–12 months, calculate MTTR by asset and by asset class, and identify the outliers.
• Segment by criticality. A bottleneck asset that halts the entire line warrants a tighter MTTR target (and a more aggressive parts-stocking posture) than a non-critical utility asset with available backup.
• Target the biggest drivers. The fleet-level MTTR is often dominated by a small number of long events — typically the ones with parts-availability or difficult-diagnosis characteristics. Identify those and address them specifically.
• Track trend, not just level. A MTTR rising quarter over quarter on the same asset is a signal even if the absolute number looks acceptable.

MTTR should always be read alongside MTBF and the broader reliability picture described in our MTBF, MTTR, and OEE overview. A very low MTTR on an asset that fails every week is not a success story — it is a frequency problem in need of a preventive maintenance interval adjustment.

---

Tracking MTTR without a spreadsheet that breaks

For a small fleet — say, five to ten assets — a simple repair log in a spreadsheet is workable. Record the failure timestamp, the restore timestamp, and the calculated duration for each event. Sum the durations, divide by the count, and you have MTTR.

The method breaks down in two ways as you scale. First, the manual calculation across 30, 50, or 100+ assets becomes error-prone and time-consuming. Second, the repair log lives separately from the PM schedule and cost projections, so you cannot see whether an asset with a rising MTTR is also underperforming on its PM compliance — the two signals that, together, tell you whether you have a frequency problem or a recovery problem.

A persistent, multi-asset calculation engine — one that holds the repair history alongside the PM schedule and fleet cost rollup — keeps those signals in one place. That is a different tool from a free one-time calculator widget, which produces a single estimate with no saved registry, and a different tool from a full CMMS, which is built for work-order execution rather than interval planning and cost forecasting.

---

Do the math now — and save it

If you want to calculate MTTR, MTBF, and OEE for your fleet in a structured, pre-built format, the MTBF / MTTR / OEE Calculator Workbook is built for exactly this. Enter your repair events and operating hours by asset; the workbook calculates MTTR, MTBF, and OEE automatically, and shows how changes in repair time affect availability — so you can model the impact of a parts-stocking improvement or a new SOP before you implement it.

Download it as a standalone tool, or use it alongside the Maintenance Cost and Interval Planner to connect your reliability metrics to your PM schedule and annual cost forecast.