A 60% OEE score is fairly typical for manufacturers today, and the gap to the 85% world-class benchmark is usually split roughly evenly across unplanned downtime, micro-stops and slow cycles, and scrap/rework [S6].
Module lines built around injection molding, assembly workstations, and packaging cells follow the same Availability x Performance x Quality multiplication, with each factor landing near 90% in mature plants but dropping into the 70 to 80% range wherever downtime categories are not actively tracked to root cause [S2][S7].
How OEE Is Calculated on a Module Line
OEE multiplies Availability, Performance, and Quality, where Availability is run time divided by planned production time, Performance is actual speed divided by ideal cycle speed, and Quality is good parts divided by total parts started [S5].
Planned production time excludes breaks, lunches, safety meetings, and scheduled maintenance, so a two-shift day with two 15-minute breaks and a 30-minute lunch collapses from 960 minutes to 840 minutes of planned run, against which every unplanned stop is benchmarked [S2]. A 1 hour 35 minute unscheduled event on top of that 14-hour window drops Availability to 89% before performance and quality are even considered [S2].
Typical OEE Bands and World-Class Reference
Plants commonly sit in the 55 to 65% band, well-run lines climb into the 70 to 80% band, and the 85% threshold is treated as the world-class reference set by ISO 22400-1:2014 and reinforced through TPM and Lean programs [S5][S6].
The same standard treats 100% as a theoretical maximum: zero defects, zero slow cycles, zero downtime, which no operating module line sustains because changeovers, restarts, and minor stoppages are structural to discrete manufacturing [S9]. Scoring above 85% usually requires disciplined SMED changeover programs, predictive maintenance on the bottleneck asset, and inline defect capture rather than end-of-line inspection [S5].
Availability Losses: Unplanned Downtime and Changeovers
Unplanned downtime, driven by equipment failures, power outages, and supply disruptions, is the single largest contributor to low OEE on most module lines, and it drags the Availability factor into the 80 to 90% range even on otherwise healthy assets [S3].
Slow maintenance reaction compounds the problem: a 30-minute fault that should clear in 10 minutes doubles its OEE cost, which is why MTTR, MTBF, and MTTF are tracked alongside Availability rather than as separate metrics [S3]. On the planned side, changeovers sit inside Availability in the strict OEE definition and are the second-biggest availability loss after breakdowns, with mold changes, tooling adjustments after quality failures, and material changeovers all counted as downtime events [S8]. OEE tracking software that simply bins every stoppage as "downtime" understates the changeover penalty and hides the SMED opportunity [S5].
Performance Losses: Micro-Stops, Slow Cycles, and Long Changeovers
Performance losses come from micro-stops, reduced running speed, and cycle degradation, and they routinely account for one-third of the total OEE gap on packaging and assembly cells where the ideal cycle is rarely sustained [S5].
Micro-stops are typically caused by starved or blocked conditions, sensor faults, or product jams that clear in under a minute, and because they fall below typical operator-reporting thresholds they are invisible to legacy downtime tracking even though collectively they can drop Performance from 95% into the 80% band [S1][S3]. The OEE Downtime Module approach is to track blocked and starved times separately so a true equipment fault is not blamed for a starvation event two cells upstream, which keeps the root-cause attribution clean enough to act on [S1].
Quality Losses: Scrap, Rework, and First-Pass Yield
Quality captures scrap, rework, and first-pass yield losses, and on module lines with manual assembly steps the Quality factor commonly sits in the 92 to 97% range, with the worst offenders being start-up scrap after a changeover and cosmetic rejects at the vision station [S5].
Long production launches are a specific quality drain, often driven by manual communication with the lab or by changeover sequences that have not been optimised, with the typical pattern being 30 to 90 minutes of elevated scrap rate after every product switchover [S3]. Tracking Quality as a live metric rather than a shift-end tally lets operators see reject rate climb before the batch is complete, which is the difference between scrapping 20 parts and scrapping 200 [S1][S5].
Detection Logic: Why Root-Cause Attribution Matters
Effective OEE downtime detection algorithms distinguish simple equipment state, initial reason, parallel cells, and key reason, and the choice of algorithm determines whether the system reports the true root cause of a stoppage or simply the first cell that tripped [S1].
The key-reason algorithm is the most informative for a module line where the bottleneck cell can sit anywhere in the line and where backup or starved conditions are routine, because it walks the cell graph to find the cell that is both stopped and not a downstream consequence of an upstream stop [S1]. For cells where no single root cause is appropriate, tracking cycle times and surfacing blocked versus starved events separately is a better fit, and on a conveyor sorting line the distinction between a sensor-induced stop and a recirculation jam is what allows maintenance to dispatch the right skill set [S1].
Common Calculation Mistakes That Inflate or Deflate OEE
Seven recurring errors corrupt module-line OEE: inconsistent definitions of planned production time, cycle times set above the true ideal, scrap counted as downtime, downtime counted as planned stops, micro-stops excluded from the data set, single-shift bias in the data, and benchmarks that compare apples to oranges across product mixes [S7].
A typical trap is using the takt time of the slowest product family as the ideal cycle for the whole line, which inflates Performance on faster products and hides the real gap on the slow ones [S7]. The remediation is to recalculate ideal cycle per SKU and to define Planned Production Time against the published shift schedule, not against observed machine-on time, which prevents the silent reclassification of breakdown minutes as "scheduled" [S4][S7].
Legacy MES vs Plug-and-Play OEE Software: Cost and Agility
Most legacy MES platforms were deployed 7 to 15 years ago, and changing a downtime reason code or an OEE dashboard typically costs $10K+ per change request with several weeks of lead time, which is why supervisors keep returning to their Excel sheets [S5].
Plug-and-play OEE systems clip onto existing machines via current sensors, direct Ethernet, or operator mobile apps, and they go live in weeks without touching the existing MES, which is the only practical way to get fresh root-cause data into the hands of second-shift supervisors [S5]. For module lines that already include molding cells, the same sensor layer feeds both the MES order record and the live OEE dashboard, which is the architecture that closes the loop between a quality alert and the next planned changeover [S1][S5].
The next node to track is the adoption curve of plug-and-play OEE layers on legacy module lines through 2026, and a second signal is whether the key-reason algorithm becomes a default option in mainstream MES OEE modules rather than a specialist configuration. For plants already running module lines, a practical first step is a two-week time study that separates micro-stops, changeovers, and breakdown events and feeds the categories into a root-cause register rather than a single "downtime" bucket.
See also our earlier report, Argon supply tightness in 2026: shielding gas procurement for welding and steelmaking.