Feeder System MTBF and MTTR Guide: Measuring and Improving Reliability

Reliability is measurable, and that means it is improvable

When a vibratory feeder jams twice per shift, the production team knows it is unreliable. When it jams once per week, they might consider it fine. But "once per week" is not a reliability metric — it is an anecdote. To improve feeder reliability, you need to measure it consistently, compare it against benchmarks, and track the effect of changes over time. MTBF (Mean Time Between Failures) and MTTR (Mean Time To Repair) are the two metrics that make this possible.

MTBF tells you how long the feeder typically runs before a failure stops production. MTTR tells you how long it takes to get the feeder back online after a failure. Together, they determine the availability of the feeder: the proportion of scheduled time that the feeder is actually producing parts. Availability is the first component of OEE (Overall Equipment Effectiveness), and for feeder systems, it is often the component with the most room for improvement.

This guide covers the definitions and calculations for MTBF and MTTR, benchmark values for different feeder types, data collection methods, root cause analysis for low MTBF, strategies to improve both metrics, and how MTBF and MTTR connect to OEE and total cost of ownership. It builds on concepts from our automated feeding system TCO guide and our feeder retrofit and upgrade guide.

MTBF and MTTR: definitions and calculations

MTBF is the average operating time between consecutive failures. It is calculated as total operating time divided by the number of failures in that period. For example, if a feeder runs 720 hours in a month and experiences 3 failures, MTBF = 720 / 3 = 240 hours.

MTTR is the average time from failure to restored operation. It is calculated as total downtime due to failures divided by the number of failures. If those 3 failures caused 2.5 hours, 1.0 hour, and 0.5 hours of downtime respectively, MTTR = (2.5 + 1.0 + 0.5) / 3 = 1.33 hours.

Availability is derived from both metrics: Availability = MTBF / (MTBF + MTTR). In the example, Availability = 240 / (240 + 1.33) = 99.4%. This sounds high, but in a plant running 24/7, 0.6% unavailability equals about 52 hours of downtime per year from this feeder alone.

There are important nuances in how these metrics are defined. "Failure" must be consistently defined — does it include only unplanned stops, or also scheduled maintenance? The most useful approach for feeder systems is to count any event that stops the feeder from delivering parts to the downstream process, regardless of cause. Scheduled preventive maintenance windows are excluded from the operating time calculation but tracked separately.

• MTBF = Total operating time / Number of failures: measures how long the feeder runs between problems.
• MTTR = Total repair time / Number of failures: measures how quickly the feeder is restored after a problem.
• Availability = MTBF / (MTBF + MTTR): the proportion of time the feeder is producing parts.
• Define "failure" consistently: any event that stops part delivery, regardless of root cause.

Benchmark values for different feeder types

MTBF varies significantly by feeder type, application complexity, and operating environment. A simple vibratory bowl feeder running one part type in a clean environment will have a much higher MTBF than a multi-part flexible feeder with vision sorting in a harsh environment. The following benchmarks are based on industry data and manufacturer experience; actual values depend on the specific installation.

These ranges are wide because the operating context matters enormously. A vibratory bowl running dry, clean screws at moderate speed in a temperature-controlled plant will sit at the top of its range. The same bowl running oily stamped parts with burrs in a foundry environment will sit at the bottom. The benchmarks are useful as a reality check: if your tooled vibratory bowl shows an MTBF of 50 hours, something is wrong and needs investigation.

Data collection: CMMS, manual logs, and PLC tracking

The quality of MTBF and MTTR data depends entirely on how consistently failures and repair times are recorded. Three methods are common, each with different accuracy and effort levels.

Manual paper logs are the simplest to start with but the least reliable. Operators write down when a jam occurred, what they did to clear it, and how long it took. The problem is consistency: some operators log every event, others only log the major ones. Repair time estimates are often rounded or approximate. Manual logs are acceptable for getting started, but they should be replaced with a more systematic method as soon as possible.

CMMS (Computerized Maintenance Management System) entries are more structured. When a feeder fails, the operator or technician creates a work order, records the failure mode, and logs the time to repair. CMMS data is more consistent than paper logs and enables trend analysis across equipment. The limitation is that small events (a 30-second jam cleared by the operator without calling maintenance) are often not recorded, which inflates the apparent MTBF.

PLC-based tracking is the most accurate method. The feeder controller already knows when the feeder is running and when it is stopped. By logging run/stop transitions with timestamps, the PLC captures every failure event automatically, including brief jams that operators would not bother to log. MTTR is measured from the stop timestamp to the restart timestamp. The data can be exported to a CMMS or MES for analysis.

• Manual logs: easy to start, low accuracy — useful for initial assessment but not for trend tracking.
• CMMS entries: structured and searchable, but misses brief operator-cleared events — good for maintenance planning.
• PLC tracking: captures every stop/start automatically — the gold standard for MTBF and MTTR data.

Root cause analysis for low MTBF

When MTBF is below benchmark, the first step is to classify failures by mode. The most common failure modes for vibratory feeder systems are: jamming at the tooling or discharge, misorientation causing downstream rejects, controller faults (amplitude drift, coil failure), sensor failures (part detection, bowl level), and mechanical wear (spring fatigue, surface coating degradation).

A Pareto analysis of failure modes typically reveals that one or two modes account for the majority of failures. For example, if 60% of failures are jamming at a specific tooling feature, the improvement effort should focus on that feature — not on the occasional sensor failure that accounts for 5% of events.

For each dominant failure mode, apply the 5-Why method to identify the root cause. A jam at the discharge might be caused by: (1) part geometry variation within the incoming lot, (2) tooling wear that narrows the track clearance, (3) insufficient air blow-off pressure, or (4) controller amplitude drift. Each root cause leads to a different corrective action: tighter incoming part inspection, tooling replacement, air pressure regulation, or controller recalibration.

The key discipline is to record the failure mode and root cause for every event, not just the repair action. Without this data, MTBF remains a number without a direction for improvement.

• Classify failures by mode: jam, misorientation, controller fault, sensor failure, mechanical wear.
• Apply Pareto analysis: focus on the one or two modes that cause the most failures.
• Use 5-Why on dominant modes: dig past the symptom to the root cause before implementing a fix.
• Record mode and cause for every event: the data is what turns MTBF from a score into an improvement tool.

Strategies to improve MTBF

Improving MTBF means preventing failures or delaying their onset. The strategies fall into three categories: preventive maintenance, spare parts strategy, and design upgrades.

Preventive maintenance addresses the known wear-out mechanisms before they cause failures. For vibratory feeders, the primary wear items are the bowl surface coating (polyurethane or epoxy), the leaf springs, the drive coil insulation, and the tooling edges. A preventive maintenance schedule based on operating hours — not calendar time — ensures that wear items are replaced before they fail. Typical intervals: bowl coating inspection every 2,000 hours, spring replacement every 5,000-8,000 hours, coil inspection every 10,000 hours.

Spare parts strategy ensures that when a wear item does fail, the replacement is available immediately. The recommended approach is to stock a critical spare kit for each feeder type: one set of leaf springs, one replacement coil, one set of common sensors, and one set of wear tooling. The cost of this kit is small compared to the downtime cost of waiting for parts to ship.

Design upgrades address recurring failure modes that preventive maintenance cannot solve. Common upgrades include: replacing standard leaf springs with high-fatigue-life springs, upgrading the bowl coating from standard polyurethane to a higher-wear formulation, adding an air knife or brush at the jam-prone tooling section to prevent part stacking, and installing a bowl level sensor to prevent overfilling, which is a frequent cause of jamming.

Strategies to reduce MTTR

Reducing MTTR means getting the feeder back online faster after a failure. The strategies here are diagnostic aids, modular design, and training.

Diagnostic aids help the technician identify the failure mode quickly. The most effective aid is a fault display on the HMI that shows the specific alarm condition (for example, "jam at discharge sensor 2" instead of "feeder fault"). When the controller can distinguish between a jam, a sensor failure, and a coil fault, the technician goes straight to the right component instead of troubleshooting by trial and error. A well-organized alarm list with recommended corrective actions for each alarm code can cut diagnostic time by 50% or more.

Modular design allows failed components to be swapped quickly. Quick-change tooling modules that slide out on rails, spring packs that bolt on from the outside of the base, and plug-in sensor cables with standard connectors all reduce the hands-on repair time. The design principle is: if a component is likely to fail, it should be replaceable without disassembling the feeder.

Training ensures that operators and maintenance technicians know the common failure modes and the standard corrective actions for each. A one-page troubleshooting guide posted at the feeder station — showing the top 5 alarm codes and the corrective action for each — is more effective than a 50-page manual that nobody reads during a line-down event. Training should also cover the difference between clearing a jam (operator task) and diagnosing a recurring failure (maintenance task).

• Specific alarm codes on the HMI: "jam at discharge sensor 2" instead of "feeder fault" — cuts diagnostic time by 50%+.
• Quick-change modules: tooling on rails, external spring packs, plug-in connectors — reduces hands-on repair time.
• One-page troubleshooting guide at the station: top 5 alarms and corrective actions — faster than searching a manual during a line-down event.
• Operator vs maintenance task boundaries: clear rules on when to clear a jam and when to call for diagnosis.

Connecting MTBF and MTTR to OEE and TCO

MTBF and MTTR are not standalone metrics. They feed directly into OEE and TCO calculations, which are the business-level measures that justify investment in reliability improvement.

OEE = Availability × Performance × Quality. Availability is determined by MTBF and MTTR. For a feeder with MTBF of 1,000 hours and MTTR of 1 hour, Availability = 1,000 / 1,001 = 99.9%. If MTBF drops to 200 hours and MTTR increases to 2 hours, Availability = 200 / 202 = 99.0%. That 0.9% difference sounds small, but over 8,760 operating hours per year, it equals 79 hours of additional downtime — roughly 3.3 days of lost production.

The TCO connection is more direct. Each hour of feeder downtime has a cost: the lost production value, the labor cost of the repair, and any scrap or rework caused by the failure. If a feeder failure costs $500 per hour in lost production and $100 per hour in maintenance labor, and the feeder experiences 40 hours of downtime per year, the annual downtime cost is $24,000. Over a 10-year equipment life, that is $240,000 — often more than the initial purchase price of the feeder.

This is why investing in reliability improvement — better coatings, preventive maintenance, spare parts kits — pays for itself. A $2,000 investment that increases MTBF by 50% and reduces annual downtime by 20 hours saves $12,000 per year in downtime costs. The payback period is under two months.

• OEE Availability = MTBF / (MTBF + MTTR): the first component of OEE, directly driven by reliability metrics.
• Downtime cost = failure hours × (lost production + maintenance labor): often exceeds the feeder purchase price over the equipment life.
• Reliability investments pay back quickly: a $2,000 improvement that saves 20 hours of downtime per year pays back in under two months at typical line rates.

Frequently Asked Questions

What is a good MTBF for a vibratory bowl feeder?

A well-maintained tooled vibratory bowl feeder in a general industrial environment typically achieves an MTBF of 500 to 2,000 hours. Simple single-part bowls in clean environments can exceed 5,000 hours. If your feeder MTBF is below 200 hours, there is likely a specific root cause that can be addressed through maintenance or design changes.

Should scheduled maintenance be counted as failures in MTBF?

No. MTBF should count only unplanned failures that stop the feeder from delivering parts. Scheduled preventive maintenance is excluded from the operating time and failure count. However, if a scheduled maintenance task discovers a condition that would have caused a failure if left unaddressed, that near-miss should be recorded separately for trend analysis.

How is MTTR different from MTBF?

MTBF measures the average time the feeder runs between failures — it reflects reliability. MTTR measures the average time to restore the feeder after a failure — it reflects maintainability. Improving MTBF prevents failures from happening. Reducing MTTR makes failures less costly when they do happen. Both contribute to higher availability.

Can PLC data automatically calculate MTBF and MTTR?

Yes. By logging the timestamps of feeder run/stop transitions, the PLC can automatically calculate MTBF (average time between consecutive stops) and MTTR (average time from stop to restart). This method captures every event, including brief jams that operators might not log manually. The data can be exported to a CMMS or MES for reporting and trend analysis.

What is the relationship between MTBF and OEE?

MTBF and MTTR determine the Availability component of OEE. Availability = MTBF / (MTBF + MTTR). For a feeder with MTBF of 1,000 hours and MTTR of 1 hour, Availability is 99.9%. If MTBF drops to 100 hours, Availability drops to 99.0%. Since OEE = Availability × Performance × Quality, a drop in Availability directly reduces OEE.

How do I justify the cost of reliability improvements to management?

Calculate the annual downtime cost using your current MTBF and MTTR: multiply annual failure hours by the cost per hour of downtime (lost production plus maintenance labor). Then estimate the reduction in downtime from the proposed improvement. The difference is the annual savings. Most feeder reliability improvements — spare parts kits, preventive maintenance schedules, coating upgrades — have payback periods of less than six months.

Conclusion

MTBF and MTTR turn feeder reliability from a subjective impression into a measurable, improvable metric. Start by defining what counts as a failure, collecting data consistently, and calculating baseline values. Compare against benchmarks to see where you stand. Use Pareto analysis and root cause investigation to identify the highest-impact failure modes. Then apply the appropriate combination of preventive maintenance, spare parts strategy, design upgrades, diagnostic aids, modular design, and training to improve both metrics. The business case is straightforward: every hour of avoided downtime pays for the improvement many times over. If you need help assessing your feeder reliability or planning an improvement program, reach out to our engineering team with your current operating data and we can recommend a targeted action plan.