Production Line KPIs: Benchmark Ranges for OEE, Line Balance and Schedule Attainment
The KPIs that run a conveyor or assembly line, how to measure each one, the levers that move it, and which benchmark figures have a cited source.
A conveyor or assembly line can be run from a short list of KPIs: overall equipment effectiveness (OEE), line balance efficiency, schedule attainment, machine utilization, bottleneck loss and, where process windows are tight, conveyor dwell time. Use the ranges below as reference points for spotting a gap worth investigating, not as targets. Your own trend, measured per shift and per SKU, says more than any single industry figure.
Every figure shows its basis. A superscript number links to a public source in the Sources list, and the text says when a cited figure is cross-industry or comes from packaging lines. A figure marked no cited source is kept exactly as originally published because no public source stating it was found, so treat it as illustrative.
Words in parentheses keep each figure's original framing, and Not stated means no figure was given for that column. The OEE rows are cross-industry or from packaging lines, not from a conveyor-specific study; the OEE section gives each source's scope.
Method: How the figures in this guide are sourced
A figure with a superscript number is stated by the cited source, listed under Sources with its publisher, location in the document and access date. Several cited figures are cross-industry or dated, and the text says so where they appear. A figure marked no cited source is kept exactly as originally published because no public source stating it was found. It is illustrative, not measured industry data. Worked examples are arithmetic, not benchmarks.
Overall equipment effectiveness
OEE is the headline line KPI. World-class discrete assembly runs 85 percent OEE and above3, 1; typical plants land at 60 to 65 percent1, 2, and anything under 50 percent signals a line in trouble no cited source.
The 85 percent target factors as roughly 90 percent availability, 95 percent performance and 99 percent quality no cited source. The Lean Enterprise Institute's worked example multiplies exactly those factors, 0.90 x 0.95 x 0.99 = 84.6 percent, without calling the result world class5. Nakajima's world-class factors, as tabulated by Raju and colleagues, are availability above 90 percent, performance above 95 percent and quality above 99.9 percent3.
The typical figures are cross-industry. OEE vendor Vorne observes that most manufacturers score closer to 60 percent1. A Chalmers University study of 23 companies and 884 machines found an average of 65 percent and a median of 70 percent2, and a Chalmers sample of 94 Swedish companies averaged 51.5 percent6.
Packaging lines show a similar spread. In 2009 machine data from 725 consumer packaged goods packaging operations, the middle half averaged 54.9 percent OEE and the top quarter 82 percent4.
A line stuck at 62 percent almost always has one factor dragging, usually availability in the low 70s no cited source. Measured samples do not always agree: across the 94 Swedish companies, average availability was 78.9 percent, and operational efficiency (utilization and speed rate combined) at 67.1 percent was the larger loss6.
Measure OEE per shift and per SKU, not as a plant average, because averaging hides the bad SKU.
How to measure
Multiply availability, performance and quality for each shift and SKU.
Track the three factors separately, so you know whether you lose to downtime, speed or defects.
Pull every factor from the same shift window and good-part counter.
What moves it
Availability work first, when downtime is the dragging factor.
Disciplined changeovers and a protected bottleneck.
Performance and quality work last, since those gains are usually smaller and slower.
Line balance efficiency tells you how evenly work is distributed, and it caps how lean the crew can be. World-class balance runs 90 to 95 percent no cited source; typical assembly lines sit at 75 to 85 percent no cited source. Below 70 percent, you are paying for stations that mostly wait no cited source.
Idle time is the complement of balance efficiency, so a line at 75 to 85 percent balance leaves 15 to 25 percent of paid station time idle, waiting on the bottleneck.
Measure it as total work content over stations times takt7. University line-balancing notes define efficiency the same way, as the sum of task times over the number of stations times the cycle time, and balance delay as 100 percent minus efficiency7.
The lever is rebalancing: move elements off the controlling station, split it, or add a parallel station. Test reassignments on paper before touching the floor, targeting under 8 percent idle per station no cited source.
How to measure
Sum the task times, then divide by the number of stations times takt.
Re-measure whenever the product mix or takt changes, not quarterly.
Log idle time for each station, not only for the line.
What moves it
Moving work elements off the controlling station.
Splitting the controlling station or adding a parallel one.
Schedule attainment measures whether the line hit the plan, and it separates good planning from good execution. World-class attainment is 95 percent and above of the daily leveled schedule delivered on time and in full no cited source; typical operations run 80 to 90 percent no cited source, and under 75 percent means the schedule is fiction no cited source.
Measure it as scheduled units completed on the scheduled day divided by units planned, not as raw output. Building the wrong SKU early does not count.
How to measure
Divide scheduled units completed on the scheduled day by units planned.
Count only the right SKU on the right day, on time and in full.
Track it daily on the line's tier board.
What moves it
Realistic takt setting.
Disciplined changeovers.
Protecting the bottleneck rather than simply pushing the line faster.
Machine utilization benchmarks depend on whether the asset is the constraint. A bottleneck should run 85 to 95 percent utilized no cited source; non-constraint stations deliberately run lower, often 60 to 75 percent no cited source, and forcing them higher just builds inventory.
Utilization is operating time over available time. It differs from OEE because it ignores speed and quality.
Read it against line balance: a non-bottleneck at 95 percent is overproducing, and a bottleneck at 70 percent is the first place to recover capacity no cited source. The lever on the constraint is downtime reduction; on non-constraints it is simply leaving idle time alone.
Queueing theory explains why a busy constraint needs a buffer in front of it. As the arrival rate approaches the service rate, work in process and time in the system both rise, slowly at first and then rapidly8.
How to measure
Divide operating time by available time for each station.
Label each station as constraint or non-constraint before you read its number.
Bottleneck loss converts directly to output, because the controlling station sets the whole line rate.
For example, a bottleneck at a 34 second cycle against a 30 second takt caps the line at 3,600 / 34 = 106 units per hour instead of 3,600 / 30 = 120, a loss of about 12 percent before any downtime.
World-class lines hold the bottleneck cycle within 5 percent of takt no cited source; typical lines run 10 to 20 percent over no cited source.
The lever is targeted: offload work from the constraint. A one second reduction at the bottleneck adds output, while the same second saved anywhere else adds nothing.
How to measure
Time every station's cycle and compare the slowest with takt.
Re-measure at every product mix change.
What moves it
Offloading work from the constraint.
Keeping the constraint from being starved or blocked.
Find the controlling station by comparing every station's cycle time with takt. Cycle Time
Conveyor speed and dwell time
Conveyor and dwell KPIs matter where process windows are tight. Dwell time is zone length divided by conveyor speed, so a drive that drifts off setpoint changes dwell and can push a cure or cool zone out of spec.
For example, a drive running 5 percent fast shortens dwell by 4.8 percent, and one running 5 percent slow lengthens it by 5.3 percent.
Benchmark dwell time compliance at 98 percent of parts within the process window no cited source; typical lines without closed-loop control drift to 92 to 95 percent no cited source.
How to measure
Calculate dwell as zone length over measured speed.
Audit dwell against the recipe for each zone.
Log the share of parts that stayed within the process window.
What moves it
Closed-loop speed control on the drive.
Buffer sizing, so upstream stops do not starve or flood a fixed-length zone.
OEE, attainment and utilization are shift-level and SKU-level metrics. Rolling them into weekly plant averages buries the specific loss you need to fix.
Line balance and bottleneck loss change with product mix, so re-measure them at every mix change, not quarterly. Pull all metrics from the same shift window and good-part counter so OEE, throughput and schedule attainment reconcile against each other.
A daily tier board showing OEE, attainment and the current bottleneck by station is worth more than a monthly report that averages the signal away.
Sequence the improvement levers
Sequence the levers by payoff, because chasing every KPI at once dilutes effort. Attack availability first when OEE is under 65 percent no cited source, since downtime usually dwarfs speed and quality losses and a 10 point availability gain flows straight to output.
Rebalance next when idle time exceeds 15 percent no cited source, freeing crew cost without new equipment. Protect and offload the bottleneck third, converting balance headroom into real throughput, and only then chase performance and quality points, which are smaller and slower.
Set stepped targets: move OEE from 62 to 70 to 78 over successive quarters rather than declaring 85 percent and stalling no cited source. IndustryWeek makes the same point, that a target should not be a world-class benchmark but a line drawn for a given piece of equipment, most often the constraint9.
For pace, the U.S. Environmental Protection Agency estimates that most companies can realize a 15 to 25 percent increase in equipment efficiency rates within three years of adopting total productive maintenance10.
How to build your own baseline
Pick one KPI, write its definition down, and keep it fixed.
Measure per shift and per SKU from one good-part counter.
Collect several months of data before you set a target.
Compare against your own trend first and published ranges second.
Record what changed on the line whenever the KPI moves.
Sources
World-Class OEE: Set Targets To Drive Improvement. Vorne Industries (OEE.com). Sections 'What Is World-Class OEE?', 'TPM and World-Class OEE', 'Some Perspective on World-Class OEE'. Accessed September 30, 2026.
Anatomization of the overall equipment effectiveness (OEE) for various machines in a tool and die shop. Journal of Intelligent Manufacturing and Special Equipment, Vol. 3 No. 1, pp. 97-105 (Emerald, CC BY 4.0), DOI 10.1108/JIMSE-01-2022-0004. published 2022. p. 101 (PDF p. 5): sentence "Seiichi Nakajima included the above four “world-class” numbers." and Table 1 "The world-class OEE factor"; p. 103 (PDF p. 7): "worldwide threshold of >85% set by the Japan Institute of Plant Maintenance (JIPM)". Accessed September 30, 2026.
Assembly-Line Balancing (lecture notes). King Saud University (faculty course notes, dated 4/20/2015). published April 20, 2015. Slide page 2 (Figure 9.12 worked example) and page 8 ('Step 6: Compute efficiency and balance delay'). Accessed September 30, 2026.
Little's Law as Viewed on Its 50th Anniversary. Operations Research, Vol. 59, No. 3, pp. 536-549 (INFORMS), J. D. C. Little, MIT Sloan; DOI 10.1287/opre.1110.0940. published 2011. Section 1 (Introduction), PDF p. 1: statement of L = λW; Section 4 (operations management), PDF p. 10: WIP = TH x CT (Hopp and Spearman); healthcare staffing discussion, PDF p. 11: queue growth as arrival rate approaches service rate. Accessed September 30, 2026.
OEE: Learn How to Use It Right. IndustryWeek (Ellis New). published August 19, 2014. Section 'Key 3: There is no absolute that works as an OEE benchmark or target'. Accessed September 30, 2026.