Your packaging line goes down at 2:15 PM. The technician arrives at 2:22 PM, diagnoses a failed proximity sensor by 2:40 PM, replaces it by 2:55 PM, and the line restarts at 3:00 PM. That breakdown took 45 minutes from failure to production. Multiply that by 8 breakdowns per week, and you are looking at 6 hours of lost production every week from repair time alone.
MTTR tells you how long, on average, it takes to get a machine from "broken" to "running again." It is one of the most practical numbers in maintenance because it directly measures how fast your team responds to problems. A lower MTTR means less downtime, more production, and higher OEE.
What is MTTR?
MTTR stands for Mean Time to Repair. It is the average time it takes to restore a piece of equipment to normal operation after a failure. The clock starts when the failure is detected and stops when the machine is back in production.
MTTR includes everything that happens during a repair event:
- Notification and response time (getting the right technician to the machine)
- Diagnosis time (figuring out what is wrong)
- Waiting time (getting parts, tools, or additional help)
- Actual repair time (hands-on fix)
- Verification time (testing to confirm the fix works)
MTTR is sometimes defined more narrowly as just the hands-on repair time. In this article, we use the broader definition that includes everything from detection to restart, because that is what matters to your production schedule.
The MTTR Formula
The formula is straightforward:
MTTR = Total Repair Time / Number of Repairs
Total repair time is the sum of all downtime caused by corrective maintenance events in a given period. Number of repairs is the count of those events.
Worked Example
A bottling line had 6 unplanned breakdowns in January:
| Breakdown | Failure | Repair Time |
|---|---|---|
| #1 | Conveyor belt tracking off | 25 min |
| #2 | Proximity sensor failure | 45 min |
| #3 | Pneumatic cylinder leak | 90 min |
| #4 | HMI communication error | 15 min |
| #5 | Filler nozzle clog | 30 min |
| #6 | Drive belt worn, slipping | 55 min |
Total repair time = 25 + 45 + 90 + 15 + 30 + 55 = 260 minutes
Number of repairs = 6
MTTR = 260 / 6 = 43.3 minutes
This means that on average, each breakdown on this line takes about 43 minutes from failure to restart. Notice that the pneumatic cylinder repair (90 minutes) pulls the average up significantly. That single event is worth investigating with a root cause analysis to understand why it took so long.
MTTR vs MTBF vs MTTF
MTTR is one of three related reliability metrics. Each measures a different aspect of equipment performance.
| Metric | Full Name | What It Measures | Lower or Higher is Better? |
|---|---|---|---|
| MTTR | Mean Time to Repair | Average time to fix a failure | Lower is better |
| MTBF | Mean Time Between Failures | Average time a machine runs before failing (repairable equipment) | Higher is better |
| MTTF | Mean Time to Failure | Average lifespan before failure (non-repairable items like light bulbs, seals) | Higher is better |
The relationship between them:
- MTBF tells you how often a machine breaks down. A pump with an MTBF of 720 hours fails about once every 30 days of continuous operation.
- MTTR tells you how long each breakdown lasts. If that same pump has an MTTR of 2 hours, you lose about 2 hours every 30 days.
- Availability connects them: Availability = MTBF / (MTBF + MTTR). For our pump: 720 / (720 + 2) = 99.7%.
To improve equipment availability, you can either increase MTBF (make failures less frequent through better maintenance) or decrease MTTR (make repairs faster when failures do occur). In practice, you should work on both.
MTTR Benchmarks by Industry
MTTR varies widely depending on equipment complexity, spare parts availability, team skill level, and how well repair knowledge is documented. Here are typical ranges based on industry data and our work with manufacturing plants.
| Industry | Typical MTTR | Best-in-Class | Common Bottleneck |
|---|---|---|---|
| Automotive Assembly | 30-60 min | <20 min | Parts availability |
| Food & Beverage | 45-90 min | <30 min | Sanitation requirements |
| Pharmaceuticals | 60-120 min | <45 min | Validation and documentation |
| Packaging | 20-45 min | <15 min | Diagnosis time |
| Chemical / Process | 2-8 hours | <2 hours | Safety procedures, permits |
| Metals & Steel | 1-4 hours | <1 hour | Equipment size and access |
| Discrete Manufacturing | 30-75 min | <25 min | Technician skill gaps |
If your MTTR is above the typical range for your industry, focus on the common bottleneck column first. That is usually where the biggest gains are hiding.
What Drives High MTTR
When MTTR is too high, it is almost always caused by one or more of these five factors:
1. Slow detection and notification
If operators do not report failures immediately, or if the notification system relies on someone walking to the maintenance office, you lose 10-20 minutes before the technician even knows there is a problem. Plants using automated alerts (machine sends a text or notification when it faults) cut this phase to under 2 minutes.
2. Long diagnosis time
This is the #1 driver of high MTTR in most plants. A technician arrives at a machine with no information and starts from scratch. If the machine has 200 possible failure modes and the technician has never worked on this exact problem before, diagnosis can take an hour or more. Access to repair history, troubleshooting guides, and experienced colleagues dramatically shortens this phase.
3. Missing spare parts
You diagnose the problem in 15 minutes, but the replacement part is not in the storeroom. Now you are waiting for a supplier delivery. Some plants keep 2-3 day emergency lead times on critical spares. Others lose entire shifts waiting for a $40 sensor because nobody tracked inventory levels. Review your spare parts strategy for your top 20 failure modes. Keep critical spares on-site.
4. Skill gaps
A senior technician with 15 years on the same equipment can diagnose most failures by sound and feel. A new hire with 6 months of experience cannot. When the senior tech retires, institutional knowledge walks out the door. Plants that document repair procedures and maintain searchable troubleshooting databases protect themselves from this risk.
5. No structured repair process
Without a consistent process, every technician approaches breakdowns differently. Some are methodical. Others jump straight to part swapping. The result is wildly inconsistent MTTR: the same failure takes 20 minutes with one technician and 90 minutes with another. A structured breakdown response process brings consistency and reduces the gap between your fastest and slowest repairs.
How to Reduce MTTR: Practical Steps
Reducing MTTR is not about working faster. It is about eliminating wasted time in the repair process. Here are the steps that deliver the biggest impact, in order of priority.
Improve notification speed
Get automated alerts working. When a machine faults, the assigned technician should know within 60 seconds. Use push notifications, text messages, or andon systems. The goal: zero minutes between failure and awareness.
Build a repair knowledge base
For your top 50 failure modes (the ones that account for 80% of your breakdowns), create a one-page troubleshooting guide: symptoms, most likely cause, repair steps, parts needed. Store these digitally where technicians can access them at the machine. This alone can cut diagnosis time by 30-50%.
Pre-position critical spare parts
Identify the parts that cause the most waiting time when they are out of stock. For each one, set a minimum stocking level and a reorder point. Most plants find that keeping $5,000-10,000 worth of critical spares on-site saves $50,000-100,000 in avoided downtime per year.
Cross-train your team
If only one technician can work on the PLC and that person is on vacation when it faults, your MTTR just doubled. Create a skills matrix showing which technicians are qualified on which equipment. Fill the gaps with targeted training. The goal: at least two qualified technicians for every critical system, on every shift.
Run post-repair reviews
After any repair that took longer than your target MTTR, spend 10 minutes asking: what slowed us down? Was it diagnosis, parts, skill, or something else? Track these answers over time. You will see patterns that tell you exactly where to invest.
Tracking MTTR Correctly
MTTR is only useful if you measure it consistently. Here are the rules that prevent the number from misleading you:
- Define start and end clearly. Start = when the failure is detected (machine stops producing). End = when the machine is back in production and making good parts. Not when the technician says "it is fixed," but when it is actually running.
- Include all repair events. Do not exclude outliers. That 4-hour repair on the gearbox is real downtime your production team felt. If you exclude it, your MTTR looks better on paper but your OEE tells the truth.
- Separate corrective from preventive. MTTR applies to unplanned corrective maintenance only. Planned PM shutdowns are not failures and should not be included in MTTR calculations. Mixing them will deflate your number and hide real problems.
- Track by machine and by failure mode. Plant-wide MTTR is useful for executive reporting, but for improvement work, you need MTTR by machine and by failure type. "Our MTTR is 45 minutes" is less useful than "Machine 12's bearing failures average 90-minute MTTR while all other failures average 30 minutes."
- Review monthly, trend quarterly. One month's data is too noisy to draw conclusions. Look at 3-month rolling averages to spot real trends. A single bad month does not mean your program is failing. Three bad months in a row means something changed.
Where Dovient Fits
Dovient directly targets the two biggest components of MTTR: diagnosis time and knowledge access.
- Faster diagnosis. When a technician logs a failure in Dovient, the system immediately shows past repairs on that machine with matching symptoms. Instead of starting from scratch, the technician sees what worked last time. Plants using Dovient report 40-60% reduction in diagnosis time within the first 3 months.
- Knowledge that stays. Every repair gets documented in a structured, searchable format. When your experienced technician retires, the knowledge does not leave with them. New hires can search for "Machine 12 proximity sensor fault" and find every previous instance, along with what caused it and what fixed it.
- MTTR tracking built in. Dovient automatically calculates MTTR by machine, by failure mode, and by technician. You can see trends without building spreadsheets. When MTTR creeps up on a specific machine, you know about it before it becomes a pattern.
- Spare parts intelligence. By tracking which parts are used in repairs, Dovient helps you identify which spares to keep on-site. If a specific sensor fails 3 times per year and it takes 6 hours to get a replacement delivered, that is a sensor you should stock locally.
If you want to see where your MTTR stands today, start by calculating it for your top 5 machines using the formula above. Then use our free OEE Calculator to see how your repair time connects to overall equipment performance. If you want help reducing your MTTR, schedule a conversation with our team.