NOTQIN — OT Track

OEE

2 min read

OEE — Overall Equipment Effectiveness

Definition

Single composite metric of how productively a piece of equipment is being used vs. its theoretical maximum during planned production time. Originated in TPM (Total Productive Maintenance, Nakajima 1980s); formalized in ISO 22400-2.

Formula

OEE = Availability × Performance × Quality

Availability = Run Time / Planned Production Time
Performance  = (Ideal Cycle Time × Total Count) / Run Time
Quality      = Good Count / Total Count

World-class benchmark ≈ 85% (varies by industry). Most Moroccan SMEs run unmeasured; when measured first time, 40–55% is typical.

Required data

  • Planned production time (per shift / per line)
  • Downtime (sum of stoppages during planned production time)
  • Ideal cycle time (engineering spec per part)
  • Total count (parts produced including rejects)
  • Good count (parts produced meeting spec)

Possible data sources

  • PLC tags (machine status, part counter, reject counter)
  • HMI / SCADA event log
  • MES production orders for planned time + ideal cycle time
  • Manual operator entry for stoppage reason codes (Stage 3 escalation)

Factory example

Auto Tier-2 stamping press, 8-hour shift:

  • Planned production time: 480 min
  • Downtime: 90 min (changeovers, jams, breaks)
  • Run time: 390 min
  • Ideal cycle time: 12 sec/part
  • Total count: 1700 parts
  • Good count: 1632 parts (68 rejected)
Availability = 390/480       = 81.25%
Performance  = (12 × 1700)/(390 × 60) = 87.18%
Quality      = 1632/1700     = 96.00%
OEE          = 0.8125 × 0.8718 × 0.96 = 68.0%

Economic impact

  • Lost output (revenue, missed shipments)
  • Overtime to recover
  • OEM scorecard hits (IATF 16949 OTD, PPM)
  • Hidden capacity = avoided capex if recoverable

Common mistakes

  • Using scheduled time instead of planned production time → inflated availability
  • Ignoring micro-stops (<1 min) → fake high performance
  • Reject count from final QC only → quality artificially high (in-process scrap missed)
  • Reporting OEE per plant instead of per machine/line → averages hide the bottleneck (the whole point)
  • Comparing OEE across different products/mixes → meaningless without same ideal cycle time
  • Calling 60% “bad” without context — industries vary; the trend matters more than the absolute

See also

Graph View

Backlinks