Comparing OEE Across Sites Without Hiding Product Mix

Use the working record

Download the editable worksheet. Record the actual evidence and responsible reviewer against each relevant row. The empty results are inputs for your project, not completed tests. Add the equipment identity, operating period and reference method before assigning an outcome. Retain unsuccessful checks and the evidence from any repeat after correction.

Give the comparison one declared purpose

Decide whether the report supports local improvement, production allocation or management review. Those questions may need different views. A plant producing short specialist batches should not be judged only against a plant running a stable high-volume product. Preserve the question beside the chosen denominator so the number remains interpretable when the report reaches another manager.

Start with comparable equipment and reporting boundaries. An individual press, a complete packaging line and a department containing parallel machines are different units. Document where output is counted, which constraint is represented and whether upstream starvation or downstream blockage remains inside the measured period.

Reconcile the production calendar first

Define planned production time for every participating site. Record the treatment of breaks, changeovers, planned maintenance, unavailable orders and additional shifts. A schedule revision made after a difficult shift can materially change the apparent result. Keep its author, effective time and reason rather than silently replacing the original calendar.

A common definition does not require identical working hours. The group can compare different shift patterns if the included minutes and exclusions remain explicit. Show planned time, running time and excluded time together. A very high percentage from a lightly scheduled machine should not obscure low total output or unused calendar capacity.

Version the product rate reference

Identify the product, equipment configuration and approved ideal cycle or rate. Tooling, active cavities, pack size and count units can change the relationship between machine cycles and saleable items. Store the reference that applied during each segment, including its source and approval. A revised standard should have an effective date rather than retrospectively improving every earlier shift.

For mixed products, calculate fully productive time within compatible segments. Multiplying a shift-wide average cycle time by the total number of units can hide a change in mix. If products use different count units, convert them through a documented method or keep their contributions separate until they are expressed as time.

Distinguish a mean of percentages from a combined result

Consider this hypothetical example, using already verified fully productive time. Plant A has 120 planned minutes and 96 fully productive minutes; Plant B has 480 planned minutes and 240 fully productive minutes. Their OEE results are 80% and 50%. The simple average is 65%, while the combined time-weighted result is 336 divided by 600, or 56%.

Neither calculation describes a customer result. The example shows why the report must name its aggregation method. An unweighted mean answers what the average site percentage was. The combined ratio answers what share of the combined planned production time became fully productive time. A group should choose the measure that matches its decision and keep the underlying contributions visible.

Examine quality at the same physical boundary

Define the first-pass good quantity and the point at which rejects are recorded. A site that excludes reworked units from its reject count may appear stronger than a site that records the initial failure correctly. Separate the original quality outcome, rework activity and eventual saleable disposition. The comparison should preserve the agreed first-pass definition.

Quality records can arrive later than production counts. State whether the report is provisional, when it closes and how an authorised correction is shown. Keep the original publication and revised result traceable. A late laboratory or inspection decision should not become an unexplained change in last week's management report.

Use product views to explain the group view

Show each site's result by product family or another technically meaningful grouping. Compare similar configurations over representative periods before attributing a difference to operator practice. Volume, product complexity, available tooling and scheduled changeovers can all affect interpretation. The local production team should explain the operating context using records rather than a narrative added only after a disappointing ranking.

An improvement action should name the loss being addressed. Reducing a verified changeover interval is a different task from correcting a count conversion or revising an unrealistic rate. Keep data-quality corrections separate from physical operating changes so a reporting improvement is not presented as additional manufactured output.

Accept the report with a reproducible worked period

Choose one period from each site and retain its source counts, calendar, product references and quality records. Have an independent reviewer reconstruct the calculation outside the application. Investigate missing records, overlapping intervals and inconsistent units before accepting the group total. A matching dashboard percentage alone does not establish that every contributing input was appropriate.

The handover should include the permitted reporting views, calculation version, exceptions and correction procedure. If a site cannot supply a compatible input, keep it visibly excluded or qualified until the reporting owner approves a method. This makes expansion to another factory a controlled data-mapping task rather than a copy of an unexplained percentage.

Connect the decision to the project

Use Oee management, Why production dashboards fail, Manufacturing to place this task within the wider installation. Bring a recent operating record and identify the decision that the proposed measurement should support. A focused discussion should establish the first boundary, the responsible site participants and the evidence needed before expanding the scope.

Agree who maintains the point register and who can change interpretation rules. Preserve the previous version when a device, product, reporting period or responsibility changes. During handover, demonstrate an export that another authorised person can reconcile without relying on the dashboard operator to explain missing fields.

Discuss the scope with addanode using the actual site information, available interfaces and proposed acceptance record.

Sources and method

The Vorne OEE calculation guide supplies the calculation background. The two-plant figures above are explicitly hypothetical; the proposed reporting and acceptance procedure is an engineering method, not an industry performance benchmark.

Frequently asked questions