MTBF and MTTR: measuring unplanned downtime

To effectively reduce unplanned downtime, you first need to measure it correctly. Beyond the cumulative stop volume that appears in OEE, two finer indicators illuminate the exact nature of the problem: MTBF (Mean Time Between Failures), which measures equipment reliability, and MTTR (Mean Time To Repair), which measures restoration speed. Understanding the distinction between these two indicators — and knowing which one to target — makes the difference between effective initiatives and those that scatter their efforts. This article describes in detail the calculation, interpretation and use of MTBF MTTR unplanned downtime in manufacturing.

The target audience: maintenance managers, reliability engineers, production directors and project managers structuring an equipment reliability improvement initiative on their industrial site.

Definitions and calculation formulas

First, precisely fix what each of the two indicators covers.

MTBF (Mean Time Between Failures) — average operating time between two failures. Measures intrinsic equipment reliability.

Formula: MTBF = Total operating time / Number of failures

Example: over a 720-hour planned operating period, a line experiences 12 failures. Its MTBF is 720 / 12 = 60 hours. This means that on average, the line operates 60 hours before the next failure.

MTTR (Mean Time To Repair) — average repair and restart time after a failure. Measures responsiveness and corrective maintenance effectiveness.

Formula: MTTR = Total downtime for repair / Number of failures

Example: across the same 12 failures, cumulative repair time is 18 hours. MTTR is 18 / 12 = 1.5 hours. This means that on average, a repair operation takes 1h30.

Availability — derived from the two previous indicators. Formula: Availability = MTBF / (MTBF + MTTR). In the example: 60 / (60 + 1.5) = 97.6%. This availability is the Availability factor of OEE for the breakdown component.

Why distinguish MTBF from MTTR

Two machines can have the same contribution to unplanned downtime while presenting very different MTBF/MTTR profiles. This distinction is essential because it points to completely different improvement levers.

Case 1 — Low MTBF, low MTTR. The machine breaks down often but restarts quickly. The dominant problem is reliability (root causes of failures). Priority lever: reinforced preventive maintenance, root cause analysis of failure modes, possibly predictive maintenance.

Case 2 — High MTBF, high MTTR. The machine breaks down rarely but stays down for a long time when it does. The dominant problem is corrective maintenance organization. Priority lever: improving intervention speed (procedures, training, spare parts availability).

Case 3 — Low MTBF, high MTTR. The machine breaks down often AND stays down for a long time. Critical situation calling for a global program combining reliability and maintenance organization. This is often the case of old equipment at end of life without renewal policy.

Case 4 — High MTBF, low MTTR. Optimal situation, reliable and well-maintained equipment. Lever: maintain this performance and capitalize for other equipment in the fleet.

This 4-case typology directs corrective actions much more precisely than simple monitoring of global stop volume.

Reference orders of magnitude by industry

MTBF and MTTR values vary significantly by industrial sector and equipment type. Some reference orders of magnitude to position your own site:

Automotive Tier-1 (presses, assembly lines). Typical world-class MTBF: 150-300 hours. Typical MTTR: 30-90 minutes. Underperforming sites: MTBF 40-100 hours, MTTR 2-6 hours.

Pharmaceutical (packaging lines, lyophilization). Typical world-class MTBF: 80-200 hours. Typical MTTR: 45-120 minutes (quality procedures extend restart). Underperforming sites: MTBF 30-80 hours, MTTR 3-8 hours.

Food processing (process, packaging lines). Typical world-class MTBF: 60-150 hours. Typical MTTR: 20-60 minutes. Underperforming sites: MTBF 20-60 hours, MTTR 1-4 hours.

Plastics (injection presses). Typical world-class MTBF: 200-500 hours. Typical MTTR: 30-90 minutes. Underperforming sites: MTBF 50-200 hours, MTTR 2-5 hours.

These orders of magnitude are indicative. The most useful comparison remains that of your site vs itself over time, rather than vs external benchmarks whose exact scope may differ.

How to correctly measure MTBF and MTTR

The precision of MTBF/MTTR measurement depends on the quality of the underlying data collection. Several conditions must be met.

Precisely define what constitutes a “failure”. Not all unplanned downtime is failures. A raw material shortage, a quality stop, an operator wait are not failures in the MTBF/MTTR sense. Limit scope to stops linked to equipment failure (mechanical, electrical, electronic, hydraulic, pneumatic).

Measure actual operating time, not opening time. MTBF is calculated on the time when the equipment should have been operating, excluding planned breaks, planned changeovers, planned preventive maintenance. Without this precision, inter-period comparisons are biased.

Automatically capture starts and stops. Manual measurement introduces significant biases. A real-time measurement system with non-intrusive sensors (current, vibration, optical) captures events at the second and automatically calculates MTBF/MTTR, without entry errors.

Distinguish sub-periods to identify trends. An MTBF averaged over the year can mask recent degradation. Calculate MTBF month by month or week by week to identify inflection points.

Actions to improve MTBF (reduce failure frequency)

Improving MTBF means reducing failure frequency. Several levers contribute to this result.

Reinforced preventive maintenance on dominant failure modes. Analysis of failure histories to identify the 3-5 most frequent failure modes. Implementation of specific preventive routines (lubrication, clearance checks, preventive replacement of components with known service life).

Improved design and installation. Some recurring failures reflect an initial installation problem (misalignment, excessive vibration, insufficient protection against contamination). Investment in durable correction prevents failure repetition.

Predictive maintenance on critical equipment. Vibration, thermal, current or acoustic sensors to anticipate failures before they occur. See Predictive maintenance and unplanned downtime: conditions for success.

Operator training in first-level maintenance. Early detection of anomalies (unusual noise, vibration, nascent leak) by operators allows intervention before failure. Training investment quickly pays off.

Renewal of end-of-life equipment. Beyond a certain age or failure level, cumulative maintenance cost of old equipment exceeds amortization of new equipment. Capex decision to be informed by historical MTBF data.

Request a TeepTrak demo

Actions to improve MTTR (reduce restart time)

Improving MTTR means shortening time between failure detection and machine restart. Several levers contribute to this result.

Standardized diagnostic procedures. For recurring failures, documentation of the most efficient diagnostic sequence. This standardization avoids each technician redoing the same investigation from scratch.

Immediate availability of spare parts. Constitution of a minimum stock of critical spare parts on site. For very high-value parts, contract with supplier for delivery in under 24 hours.

Polyvalence of maintenance technicians. A technician capable of intervening on mechanical, electrical and automation is more efficient than a specialized technician who must call a colleague for each type of failure. Polyvalence training to organize over 12-24 months.

Modern maintenance tools. Tablet terminals with access to technical documentation, electrical schematics, intervention history, machine drawings. Reduces time lost searching for information.

Fluid operator-maintenance coordination. Early detection by operator, first-level intervention if possible, technician call if necessary, rapid briefing on technician arrival: each step where minutes can be gained with good practices.

Technician geolocation. On very large sites, knowing where the closest technician is allows optimization of internal travel. Mobile communication tools (smartphones, connected watches) facilitate this coordination.

Integrating MTBF/MTTR into management routines

MTBF/MTTR monitoring should integrate into regular site management routines, without becoming a statistical exercise disconnected from the field.

Weekly review. Production-maintenance team, 15-20 minutes. Review of the week’s failures, analysis of main causes, decision on actions to engage. No formal presentation — factual discussion.

Monthly assessment. Evolution of MTBF and MTTR on critical equipment, identification of drift, validation of maintenance investment arbitrations. Maintenance manager and production manager present.

Quarterly strategic review. Assessment of reliability programs, decisions on equipment end-of-life renewal, annual objective adjustment.

Ritual use — non-negotiable — of these indicators progressively builds a reliability culture that extends beyond simple numbers.

Classic pitfalls in MTBF/MTTR usage

Several recurring errors degrade MTBF/MTTR usage.

Pitfall 1 — Confusion with global availability. MTBF/MTTR covers only equipment failures. Global availability (OEE Availability factor) also includes stops for changeover, quality, raw material, etc. Confusing the two leads to erroneous comparisons.

Pitfall 2 — Calculation on too few events. On a machine with few failures (e.g. 2 per month), monthly MTBF calculation is very unstable. Prefer rolling windows (3 or 6 rolling months) to stabilize measurement.

Pitfall 3 — Pursuit of MTBF without watching MTTR. Excessive focus on reliability can lead to neglecting maintenance organization. Both indicators should be monitored in parallel.

Pitfall 4 — Sterile competition between shifts. If MTBF per shift (morning, afternoon, night) becomes a competition indicator between operators, this can generate voluntary under-declarations. Present indicators as collective improvement tools, not as individual scoring tools.

Pitfall 5 — No memory of interventions. Without a CMMS (Computerized Maintenance Management System) that historicizes interventions, retrospective analysis is very difficult. This investment usefully complements the real-time measurement system.

Frequently asked questions

What minimum MTBF to target?
The acceptable minimum depends on sector and equipment type. The right reference is comparable sites in the same sector, or your own site’s progression over time. A generic target: MTBF above 80 hours for standard production equipment, above 200 hours for critical equipment with reinforced preventive maintenance.

Should we calculate MTBF/MTTR per machine or for the entire line?
Both. MTBF per machine identifies critical equipment to reliability-improve. MTBF of the entire line measures global production impact. A line with 10 independent machines has a mechanically lower global MTBF than each machine taken in isolation.

How to integrate very short stops (micro-stops) in calculation?
Micro-stops (under 5 minutes) are generally excluded from MTBF/MTTR because they don’t belong to the same register as structural failures. They are tracked separately with their own Pareto. But this boundary is conventional — it should be clearly defined and maintained over time.

Does MTBF always degrade with machine age?
Not always. A well-maintained machine can preserve its nominal MTBF for 15-20 years. A poorly maintained machine degrades from the first years. Age is not in itself a determining factor — maintenance policy and original material quality are more so.

How to react to sudden MTBF degradation?
Immediate investigation: raw material change, operator change on a shift, recent modification of machine parameters, ambient conditions (temperature, humidity), new reference produced. Identify the causal factor then correct.

What is the cost of a 50% MTBF improvement?
Very variable depending on starting point. For a machine where preventive maintenance is poorly structured, doubling MTBF can be achieved with reinforced preventive maintenance and operator training, i.e. a few thousand to tens of thousands of dollars per equipment. For an already well-monitored machine, gaining an additional 50% requires more significant investments (predictive maintenance, structural modifications).

What specific skills for MTBF/MTTR steering?
A reliability engineer (bachelor’s minimum, ideally master’s specialized) or an experienced maintenance manager. On medium-sized sites, this competence can integrate into the maintenance manager role. On very large sites, a dedicated reliability engineer position is justified.

Conclusion

Tracking MTBF MTTR unplanned downtime effectively structures an equipment reliability improvement initiative. MTBF measures intrinsic reliability and points to preventive and predictive maintenance. MTTR measures responsiveness and points to maintenance organization and technician training.

The systematic distinction between these two indicators avoids poorly targeted corrective actions. A structured approach combines MTBF improvement and MTTR improvement, based on objective data collected in real time.

Beyond the numbers, the stake is to build a reliability culture that perpetuates progress. Weekly production-maintenance rituals, sharing analyses with operators, valorizing achieved progress: organizational levers that complement technical levers.

For the global context of unplanned downtime reduction: Reducing unplanned downtime: method and tools. For the technological levers of predictive maintenance: Predictive maintenance and unplanned downtime.

More information about TeepTrak and our deployments in 450+ factories across 30+ countries at teeptrak.com.