What's a good OEE benchmark for packaging lines?

World-class OEE for packaging lines is 85%, but most real-world lines run 55-70% in their first year and 65-75% in steady state. A line running above 75% OEE after 18 months is well-managed. Anything below 55% indicates either a mismatch between machine capability and product requirements, or operational issues with changeover, material supply, or maintenance discipline.

What MTBF should I expect from packaging equipment?

Modern servo-driven packaging machines typically deliver 600-1,200 hours MTBF in steady-state operation. Mechanical-heavy legacy lines run 300-600 hours. Anything advertised above 2,000 hours MTBF should be questioned — it usually reflects lab conditions, not production reality. Track MTBF internally starting at commissioning; supplier claims rarely match field experience.

How much can I reduce changeover time?

Most legacy packaging lines can cut changeover time by 40-70% through SMED methodology: converting internal setup to external setup, pre-staging format parts, and adding quick-change tooling. A 45-minute changeover routinely drops to 15-20 minutes after a focused SMED project. Quick-change tooling investment typically pays back in 12-18 months on multi-SKU lines.

How do advertised speeds compare to actual production speeds?

Advertised speed is the mechanical maximum running ideal product under lab conditions. Actual sustained production speed is typically 70-85% of advertised for well-matched applications and 50-70% for poorly matched ones. Ask suppliers for OEE data from existing customers running similar products, not the datasheet number.

What warranty terms should I insist on?

Standard packaging equipment warranties run 12 months from installation or 18 months from shipment, whichever comes first. Push for 24 months on major components and insist on a performance guarantee specifying minimum sustained throughput and maximum scrap rate during the warranty period. Without a performance guarantee, warranty covers defects but not fitness for purpose.

What upgrade paths matter for future-proofing?

Three matter most: servo-driven axes (vs. mechanical cams) for format flexibility, open industrial Ethernet protocols (EtherCAT, Profinet) for MES integration, and spare capacity in the electrical cabinet for additional modules. Avoid machines with proprietary controllers that lock you into one supplier for integration and spare parts.

Understanding Packaging Machine Specs: IPM, CPM, OEE, MTBF

The 5 Critical Specs on Every Datasheet

Packaging machine datasheets look comprehensive but are written to sell equipment, not to inform buying decisions. Five specifications determine whether a machine fits your application. Everything else is secondary.

Speed (expressed as packages per minute, cycles per minute, or units per hour). The headline number. Always ideal-product, ideal-condition speed. Use it as a ceiling, not a target.

Fill weight or volume range. The minimum and maximum product the machine can handle. Critical for multi-SKU lines. The useful range is narrower than advertised — typically the middle 60% of the range delivers acceptable accuracy.

Format or package size range. The minimum and maximum bag, pouch, carton, or container dimensions the machine can run. Format changes outside this range require tooling changes or are unsupported.

Power and utilities. Electrical service (voltage, phase, kVA), compressed air (CFM at PSI), cooling water if required. Often overlooked until installation, these drive facility upgrade costs.

Footprint and weight. Floor space required including maintenance access, plus operating weight. Heavy machines may require reinforced floors.

For the broader question of which machine fits which application, the Machine Selector pillar covers the matching logic.

IPM vs CPM: Why They Differ

Datasheets conflate several different speed metrics under "throughput." Understanding the distinction prevents the most common spec-reading mistake: buying a machine that never delivers its advertised speed in production.

Ideal Production Rate (IPM, Ideal Packages per Minute) is the theoretical maximum cycle rate assuming ideal product, ideal material, no stops, no changeovers, no operator intervention. This is the number in bold type on the datasheet. It is a mechanical ceiling, not a real-world target.

Cycle Rate per Minute (CPM) is the actual sustained cycle rate the machine achieves in production, averaged across a production run including minor stops and speed losses but excluding planned downtime. CPM is typically 75-90% of IPM for well-matched applications.

Net Production Rate is the finished, saleable packages per minute including all losses — rejects, jams, changeovers, and stops. This is the number that drives real economics. Net production rate is typically 50-75% of IPM in the first year and 65-80% in steady state.

The gap between IPM and net production is not a quality problem. It is physics and operations. A VFFS rated at 180 IPM running a free-flowing granular product in a standard pillow pouch may deliver 150 CPM and 125 net packages per minute. The same machine running a sticky product in a quad-seal pouch may deliver 110 CPM and 75 net. Both are running as designed.

Ask suppliers for CPM data from real customer installations running products similar to yours. If they cannot provide it, the IPM number is a marketing claim, not a performance commitment.

OEE: The Only Number That Matters

Overall Equipment Effectiveness (OEE) is the single metric that captures real packaging line performance. It is the product of three factors: availability, performance, and quality.

Availability = Run Time / Planned Production Time. Captures downtime from changeovers, jams, material replenishment, and breakdowns.

Performance = Net Cycle Rate / Ideal Cycle Rate. Captures speed losses — running below rated speed, minor stops under 5 minutes, and idling.

Quality = Good Units / Total Units. Captures scrap, rejects, and rework.

Multiply the three together. A line with 80% availability, 90% performance, and 95% quality runs at 68% OEE. That is typical for a well-run packaging line in steady state.

OEE benchmarks for packaging operations:

World-class: 85% (rare in packaging outside of high-volume beverage and tobacco)
Good: 70-80% (well-managed multi-SKU operations)
Typical: 55-70% (most factories in steady state)
Poor: below 55% (mismatched equipment, operational issues, or chronic quality problems)

First-year OEE runs 10-15 percentage points below steady state. Commissioning, debugging, and operator learning curve cost real production. Budget for it.

The OEE number matters because it is the only spec that captures the total cost per unit. A 180-IPM machine at 65% OEE produces fewer saleable units than a 140-IPM machine at 80% OEE. Slower, simpler machines often win on total economics. For how this plays out in dollars, see our 5-year TCO analysis.

MTBF and MTTR: Reliability Metrics

Mean Time Between Failures (MTBF) and Mean Time To Repair (MTTR) are the two reliability metrics that determine how much unplanned downtime a machine will generate. Together they define the machine's availability contribution.

MTBF is the average operating hours between unplanned stops. Modern servo-driven packaging machines typically deliver 600-1,200 hours MTBF in steady state. Legacy mechanical lines run 300-600 hours. A line running 5,000 hours per year at 800 hours MTBF expects roughly 6 unplanned stops per year.

MTTR is the average time to restore the machine to running condition after an unplanned stop. Packaging machines typically run 30-90 minutes MTTR depending on complexity, diagnostic accessibility, and spare parts availability. Complex lines with poor diagnostics can run 2-4 hours MTTR for non-trivial faults.

Annual unplanned downtime = (Operating Hours / MTBF) x MTTR. For 5,000 hours, 800 MTBF, 60 minutes MTTR: 6.25 hours per year — excellent. A legacy machine at 400 MTBF and 90 minutes MTTR: 18.75 hours. Worse, but not catastrophic.

The catastrophic case is 200 MTBF with 4-hour MTTR — 100 hours of unplanned downtime per year. These machines exist and typically result from poor supplier selection or poor maintenance practices.

MTBF and MTTR are not on most datasheets. Ask suppliers directly and require customer references that can speak to actual reliability. For more on supplier evaluation, see our upcoming guide on supplier questions.

Changeover Time: Hidden Cost

Changeover time is the time between the last good unit of one format and the first good unit of the next format. It is the most under-reported cost on packaging machine datasheets.

Datasheets typically report changeover time as "10-30 minutes" or similar. This is usually best-case changeover with all parts pre-staged and an experienced technician. Real-world changeover times run 2-4x the datasheet number.

The cost of changeover adds up quickly. A line running 8 SKUs with 45-minute average changeover, 5 changeovers per day, 250 days per year: 562 hours of changeover time annually. At an operating cost of $200 per hour (labor, depreciation, overhead), that is $112,000 per year — equivalent to a full-time operator.

Quick-change tooling and SMED (Single-Minute Exchange of Die) methodology can reduce changeover by 40-70%. A 45-minute changeover routinely drops to 15-20 minutes after a focused SMED project. The investment in quick-change format parts typically pays back in 12-18 months on multi-SKU lines.

Servo-driven machines have a structural advantage here. Format recipes stored in the HMI recall axis positions, timing, and recipes in seconds. Mechanical machines require manual adjustment of cams, guides, and stops. The servo premium of 15-25% on purchase price often pays back through changeover reduction alone within three years on lines running 4+ SKUs.

Reading Between the Lines: Marketing vs Reality

Datasheet language has telltale signs of overstatement. Recognize the patterns.

"Up to" speeds. "Up to 180 ppm" means the machine has hit 180 ppm once under ideal conditions. Sustained production speed is lower.

"Industry-leading" reliability. Without MTBF data, this is a marketing claim. Ask for MTBF and customer references.

"Quick changeover." Compared to what? Without a number, meaningless. Demand specific changeover times for specific format changes.

"User-friendly HMI." Every supplier says this. The question is how many screens deep the operator has to go for common adjustments, and whether recipes are parameter-driven or require manual tweaking.

"Energy-efficient." Compared to what? Demand kW draw at idle, at typical production speed, and at rated speed.

"Compact footprint." Including or excluding maintenance access? The footprint number on the datasheet often excludes the 3-foot clearance needed on each side for maintenance.

"Open architecture controls." Check whether the controller is a standard industrial platform (Allen-Bradley, Siemens, Beckhoff, B&R) or a proprietary supplier controller. Proprietary controllers lock you into the supplier for integration, programming changes, and many spare parts.

Red flags: datasheets without MTBF, OEE, or changeover numbers. Suppliers who will not provide customer references running similar products. Warranties without performance guarantees. Refusal to run a trial on your product before purchase.

Spec Questions to Ask Suppliers

A short list of questions separates credible suppliers from marketing-driven ones. Ask all of them, in writing, and require written answers.

Speed and throughput:

What is the documented CPM in production at existing customers running a similar product?
What is the typical first-year OEE for this machine?
Will you guarantee a minimum sustained production rate under defined conditions?

Reliability:

What is the field MTBF for machines delivered in the last 3 years?
What is the typical MTTR for top 5 fault types?
What spare parts do you recommend carrying on-site, and what is the total value?

Changeover:

What is the documented changeover time between two defined formats?
Are format recipes stored in the HMI or manually adjusted?
How many changeovers per day is the machine designed for?

Integration:

What industrial network protocols does the controller support (EtherNet/IP, Profinet, EtherCAT, Modbus TCP)?
Can the HMI export production data to MES/ERP systems via OPC UA or equivalent?
Is the controller programmed in IEC 61131-3 languages or a proprietary system?

Lifetime costs:

What is the annual spare parts consumption as a percentage of purchase price?
What software licensing fees apply annually?
What is the typical service labor cost per year for a line at this complexity?

Support:

What is the guaranteed response time for remote and on-site support?
Where are the nearest service technicians and spare parts depots?
What training is included in the purchase price, and what costs extra?

Credible suppliers answer these directly with numbers. Suppliers who dodge or give qualitative answers either do not know their own field performance or do not want you to know. Both are disqualifying for a 7-figure purchase decision.

The machine selector tool bakes these questions into the matching process. For a complete cost picture that uses real-world OEE, MTBF, and changeover data, see our 5-year TCO guide.