MTBM: Formula, MTBF vs. MTBM & Interval Optimisation

What is Mean Time Between Maintenance (MTBM)?

MTBM (Mean Time Between Maintenance) is the average operating time between two consecutive maintenance events — both planned and unplanned. If a press line undergoes 6 maintenance actions in a month (4 planned preventive services + 2 unplanned breakdowns) and the total operating time is 360 hours, the MTBM is 60 hours. MTBM answers a different question than MTBF: MTBF measures reliability (how often the machine fails). MTBM measures maintenance burden (how often the machine requires any maintenance attention, whether it fails or not). A machine with a high MTBF but a low MTBM is reliable but over-maintained. A machine with a low MTBF but a high MTBM is under-maintained and breaking down because of it. The relationship between MTBF and MTBM tells you whether your preventive maintenance strategy is calibrated correctly — and an MES provides the data to answer that question.

How do you calculate MTBM?

MTBM = Total Operating Time / Total Number of Maintenance Events

Where:

• Total Operating Time = the cumulative time the machine was actually producing. Excludes all downtime — planned and unplanned.
• Total Number of Maintenance Events = the sum of all maintenance actions: preventive (PM), corrective (CM — unplanned failures) and condition-based (CBM). Every event that requires a technician to touch the machine counts.

The gap between MTBF (180 h) and MTBM (60 h) tells the story: the machine fails every 180 hours on average, but maintenance touches it every 60 hours. The 4 planned events between every 2 failures are preventive. Whether those 4 events are necessary — or whether 2 would suffice — is the optimisation question that MTBM helps answer.

What is the difference between MTBM, MTBF and MTTR?

The critical relationship: MTBM ≤ MTBF — always. MTBM counts every maintenance event; MTBF counts only failures. MTBM can never be higher than MTBF. If they are equal, you have zero preventive maintenance — every maintenance event is a reactive repair. If MTBM is much lower than MTBF, you have a heavy preventive maintenance schedule. The ratio MTBF / MTBM tells you the preventive-to-total maintenance ratio. In the worked example: MTBF / MTBM = 180 / 60 = 3.0 — meaning for every failure, you perform 3 maintenance events total (2 preventive + 1 corrective).

What is the over-maintenance trap — and how does MTBM expose it?

Most manufacturing plants over-maintain. That is not opinion — it is a structural consequence of how maintenance intervals are set. The typical process:

1. The machine manufacturer recommends a PM interval (e.g., "lubricate every 500 hours").
2. After the first breakdown, the maintenance manager shortens the interval "to be safe" (e.g., "lubricate every 300 hours").
3. After the second breakdown (caused by something completely different), the interval is shortened again (e.g., "lubricate every 200 hours").
4. Result: the machine is lubricated 2.5× more often than necessary, consuming technician hours, spare parts and production time — while the actual failure cause (a worn seal, not lubrication) was never addressed.

MTBM makes this visible. When MTBM trends downward while MTBF stays flat or improves, the plant is adding maintenance activity without improving reliability. The maintenance team is busier, but the machine is not failing less. That is the over-maintenance trap.

How does an MES provide the data for MTBM optimisation?

MTBM requires accurate tracking of every maintenance event — planned and unplanned. In practice, this means:

• Every unplanned stop with alarm code, start time and end time. The SYMESTIC alarms module captures these automatically from PLC signals. At Neoperl, SPS-based alarm capture identified the specific alarm codes causing failures — enabling maintenance to target PM at the components that actually fail, not at the components the OEM manual says to check.
• Every planned maintenance event with duration and type. The MES logs planned downtime separately from unplanned downtime. When the maintenance team marks a stop as "planned PM" in the MES, the system calculates both MTBF (excluding that event) and MTBM (including it). This separation is critical — mixing planned and unplanned stops in a single metric is the most common MTBM calculation error.
• Historical trend data per machine. A single MTBM value is a snapshot. The value of MTBM is in the trend: Is the maintenance burden per machine increasing, stable or decreasing over months? The SYMESTIC production metrics module tracks MTBM alongside MTBF and MTTR per machine — providing the maintenance manager with the complete reliability picture in a single dashboard.
• Correlation between PM events and subsequent failure frequency. The most powerful MTBM insight: after extending the lubrication interval from 200 hours to 400 hours, did the failure rate change? The MES provides the before/after data that answers this question objectively. At Meleghy Automotive (6 plants, bidirectional SAP integration), this kind of data-driven interval adjustment contributed to the 10 % downtime reduction achieved in the first 6 months.

How does MTBM fit into maintenance strategy?

FAQ

Is MTBM the same as PM interval?
No. The PM interval is a planned target ("maintain every 400 hours"). MTBM is the measured reality ("the machine actually received maintenance every 60 hours on average, because it also broke down twice between the planned services"). MTBM includes both planned and unplanned events. If your PM interval is 400 hours but your MTBM is 60 hours, the unplanned failures dominate your maintenance schedule — and the PM programme is not preventing enough failures to justify itself.

Should MTBM be as high as possible?
Not unconditionally. Maximising MTBM means minimising maintenance events — which is good if MTBF stays high (the machine runs reliably with less maintenance). But if you increase MTBM by simply deleting PM events and MTBF drops as a result (more failures), you have saved PM cost and created more expensive unplanned downtime. The goal is to maximise MTBM while keeping MTBF at or above the current level. The MES provides both metrics simultaneously, so the trade-off is visible in real time.

How does MTBM relate to OEE?
MTBM affects OEE Availability indirectly. Every maintenance event — planned or unplanned — takes the machine out of production. Planned PM events are typically classified as "planned downtime" and excluded from OEE Availability. Unplanned failures are included. But both consume available production hours. A plant with MTBM of 40 hours on a machine running 480 hours/month performs 12 maintenance events per month. If each event takes 2 hours (including setup and restart), that is 24 hours/month — 5 % of available time consumed by maintenance activity, regardless of how OEE classifies it. MTBM therefore drives total equipment utilisation, even when OEE hides the planned portion.

How do I start tracking MTBM if I only have MTBF today?
If your MES already tracks unplanned downtime events (the basis for MTBF), adding MTBM requires one additional input: logging planned maintenance events with a distinct state code. In the SYMESTIC system, this means configuring a "planned maintenance" downtime reason that operators or maintenance staff select when a PM event starts. Once both planned and unplanned events are logged, MTBM is calculated automatically from the same machine-state data. At Brita, digital machine signals provided the stop data; adding PM reason codes turned MTBF tracking into full MTBM tracking with zero additional hardware.

Related: MTBF (Mean Time Between Failures) · MTTR (Mean Time To Repair) · TPM · Predictive Maintenance · OEE Explained · SYMESTIC Alarms Module · SYMESTIC Production Metrics · MES: Definition & Functions

About the author

Martin Brandel

MES Consultant at SYMESTIC. Dipl.-Ing. Nachrichtentechnik. Over 30 years in industrial automation — from commissioning process control systems in the beverage industry through to cloud MES machine connectivity. Has seen enough lubrication schedules set by fear rather than data to know what over-maintenance looks like. · LinkedIn