Practical resource

OEE data quality: define the inputs before trusting the percentage

Overall equipment effectiveness (OEE) combines availability, performance, and quality. The percentage is only interpretable when planned production time, stop rules, ideal cycle time, total count, good count, and comparison context are defined consistently.

Illustrative operator inspecting a sample beside stainless batch-processing equipment
01

Understand

Input definitions

A common structure is OEE = Availability × Performance × Quality, but organizations vary in schedule, exclusions, ideal-rate, rework, and good-count rules. Document the local definition.

InputWorking definition to approveCommon ambiguity
Planned production timeTime intended for production under the reporting policyBreaks, maintenance, no-demand time
Run timePlanned production time minus counted stop timeMicrostops and threshold rules
Ideal cycle timeApproved reference time for one unit in the stated contextProduct, cavity, speed, and constraint differences
Total countUnits processed under the chosen boundaryRework, partial units, multi-cavity output
Good countUnits meeting the defined quality stateLater rejects, rework, release timing
02

Understand

Common data errors

01

Equipment and business clocks are misaligned

02

Stops overlap or remain open across shifts

03

Unknown downtime is forced into a specific cause

04

Ideal cycle is copied across unlike products

05

Good quantity changes after the reporting cutoff without restatement

06

Manual and machine counts are combined without reconciliation

03

Apply

Compare like with like

Segment by relevant product, routing, machine or tool configuration, shift definition, scheduled time, and quality rule. Show changes in context next to the number. A rising OEE can reflect a changed product mix or exclusion rule rather than an operational improvement.

04

Review

Validation checklist

01

Trace a sample from dashboard to source events

02

Test boundary times and overlapping states

03

Expose unknown and missing data

04

Reconcile counts with the approved production record

05

Version calculation rules

06

Have operations and quality owners approve interpretations

01

Start with one process

Which workflow currently costs your team the most time?

Bring one normal example and one exception. Use them to frame the systems, decisions, controls, and evidence a sensible next step needs.

Discuss your operation