Skip to main content

OEE Quality

Quality dashboard

The OEE Quality screen focuses on the Quality component of the OEE equation. It helps quality engineers and production supervisors track reject rates, identify scrap patterns, and understand quality loss impacts on OEE.

What is quality?​

Quality measures what fraction of produced pieces meets quality standards:

Quality % = Good Pieces / Total Pieces x 100

Where:

  • Total Pieces = Good Pieces + Reject Pieces
  • Good Pieces = pieces that pass quality inspection
  • Reject Pieces = pieces that fail quality inspection (scrap, rework, etc.)

A Quality of 100% means every piece produced was a good piece. Any rejects reduce the quality percentage.

Types of quality losses​

Quality losses are categorized by the reason for rejection:

Startup rejects​

Pieces produced during startup, warmup, or changeover that do not meet quality standards. These are expected to some degree and are often tracked separately from production rejects.

Production rejects​

Pieces rejected during normal production. These indicate process issues and are the primary target for quality improvement.

Key metrics​

MetricDescription
Quality %Overall quality ratio for the selected period
Total pieces producedAll pieces (good + reject)
Good piecesPieces that passed quality checks
Reject piecesPieces that failed quality checks
Quality loss (seconds)Time spent producing rejects (reject count x ideal cycle time)
Quality objectiveTarget percentage from equipment configuration
Top quality reasonsRanked list of reject reasons by count or impact

Reject reason hierarchy​

OpSee uses a hierarchical reason classification for quality rejects, similar to the downtime reason hierarchy:

LevelPurposeExample
Level 0Top-level categoryMaterial Defect
Level 1Sub-categorySurface Quality
Level 2Specific reasonScratches
Level 3Detail (optional)Conveyor belt scratches

This hierarchy is configured in the Reason section of the Admin module.

tip

Use the reject reason hierarchy to build Pareto charts. Focus improvement efforts on the top 20% of reasons that cause 80% of the rejects.

Analyzing quality data​

Quality over time​

Review quality trends to spot patterns:

  • Does quality drop at the start of a shift (startup rejects)?
  • Does quality decline after a changeover?
  • Is quality consistently lower on certain products?
  • Does quality degrade over time (indicating tooling wear)?

Correlation with other metrics​

Quality issues often correlate with other factors:

  • Speed: running too fast can increase reject rates. Check Performance metrics.
  • Downtime: frequent stops and restarts can cause startup rejects. Check Downtime patterns.
  • Material: track material batches against quality data using the Material module.

Quality impact on OEE​

Quality is often the smallest loss factor in OEE (many factories achieve 98-99% quality), but its impact is multiplicative. A 2% quality loss applied to an 85% availability and 90% performance reduces OEE from 76.5% to 75%.

Where reject records come from​

Every reject record on this screen comes from one of two places, and each record says which:

  • A waste counter. The equipment's quality configuration can name several waste counters, one per point on the line where pieces are rejected. OpSee turns each counter increment into a reject record stamped with that counter's name, so you can tell the filler's eject gate apart from the packer's camera.
  • A waste zone. Where rejects happen between two piece counters and nothing counts them directly, a zone settles the difference when a trigger fires — a confirmed empty line, an order change, or the Line empty button. Its records are named after the zone.
  • Manual entry. Waste that no counter sees — a reject bin emptied by hand, for example — is entered by an operator on the quality screen and shows as Manual.

Filtering or grouping by the counter name is the quickest way to see whether a bad quality figure comes from one station or from the whole line. Counters are configured per equipment in OEE Settings.

caution

Waste entered by hand for pieces a counter already counted is counted twice. If a reject bin sits behind a counter that already sees those pieces, leave it to the counter.

Waste zones on the quality screen​

If the line has zones, the quality screen shows a panel with one line per zone.

Line empty. When a zone is settled by an operator trigger, the panel offers a Line empty button. Press it when the stretch of line the zone covers has actually run dry — that is what makes the figure exact. At that moment OpSee reconciles the zone: everything the upstream counter passed but the downstream counter never saw is reject, because there is nowhere else for it to be.

Press it when the line is not empty and the pieces still sitting in the buffer are recorded as rejects.

Status. Each zone shows when it last settled and whether that settlement was exact, or Not baselined yet for a zone still waiting for its first exact event — until then it writes nothing.

What is not shown, deliberately. The unsettled gap. Between settlements that number is buffer plus rejects with no way to separate the two, so showing it would invite reading a full accumulation table as scrap.

Why zone waste appears in steps​

A zone's waste does not tick up through the shift. It appears in a step at each confirmed empty line, and nothing in between.

That is the honest shape of the measurement. Between empty lines the difference between the two counters is buffer plus rejects, and nothing can separate them — a full accumulation table would otherwise be reported as scrap. So the figure you see after a vide-ligne covers everything since the previous one, attributed to the moment it was confirmed rather than spread across the hours it accumulated in.

If a zone shows no waste yet, it means the line has not been confirmed empty since it started measuring — not that nothing was rejected.

Configuration​

Quality targets and reject tracking are configured in OEE Configuration:

  • Quality objective: the target quality percentage (e.g., 99%)
  • Reject reasons: the hierarchical reason codes available for classifying rejects
  • Quality data source: how reject counts are collected (manual entry, PLC signal, vision system, etc.)
info

For accurate quality calculations, ensure that both good and reject piece counts are being collected by the data acquisition system. If only total pieces are counted, quality will always appear as 100%.

Prerequisites​

Next steps​