MES for Multi Site Manufacturing: One Standard Across Every Plant

Key takeaways

• A multi site MES works when every plant measures losses the same way.
• Set a group standard for states, loss codes and the time basis; let plants keep local detail below it.
• Compare plants on own losses, not raw OEE, or the ranking will be unfair.
• One industry survey found 93% of manufacturers have an MES, but only 28% run it enterprise wide.
• Roll out plant by plant, with the first plant's setup as the group template.

The short answer: MES for a multi site group is mostly about one shared standard. Same states, same loss codes and the same time basis in every plant, so a minute of downtime means the same thing everywhere.

This guide is for group operations, CI and engineering leads with two or more plants. Fabrico, a computer vision enabled Manufacturing Performance Platform, gives one view across plants on top of line level OEE. The method below applies to any system.

Why is MES across many plants harder than in one?

Each plant grew its own habits. One counts changeovers as planned stops, another counts them as downtime. One sets the micro stop cutoff at 2 minutes, another at 5. Put their OEE side by side and you compare rules, not performance.

The data backs this up. In one industry survey of 1,560 decision makers in 17 countries, 93% had an MES, but only 28% ran it enterprise wide and 23% were fully integrated. Integration was the top buying need, named by 44%.

What should the group standardise, and what stays local?

The rule of thumb: standardise anything that goes into a group report. Leave anything that only the line uses to the plant. Our guide to standard loss definitions across plants goes deeper.

How do you compare plants fairly?

Raw OEE punishes plants with long runs of small batches and rewards plants with one product. Compare them on what each plant can fix:

1. Own losses. Split each machine's time into own stops, starved and blocked. Rank plants on own stops.
2. Loss in money. Price each loss at contribution per unit lost at the bottleneck. A minute on a fast, high margin line weighs more.
3. Trend, not level. A plant that climbs steadily from a low base is doing better work than one that sits still at a high number.

See how to roll up OEE across sites for the math of weighting lines and plants.

A worked group comparison

Sample group (illustrative numbers, not a customer): three plants, one bottleneck line each, a weekly view.

By raw OEE, Plant B looks worst. By money, Plant C loses the most, because its line is the fastest. A group that ranks on raw OEE sends help to Plant B first and misses the biggest loss. Ranked by money lost, Plant C goes first.

How do you roll MES out across plants?

• First plant as the template. Prove it on one line, then the whole plant, and write down the setup.
• Next plants copy the template. Each adds local detail codes only below the group codes.
• One owner per plant and one group owner for the standard.
• A monthly group loss review on own losses in money, not a league table of raw OEE.

If you are still choosing between a single plant tool and a group tool, read single plant vs multi site OEE.

What does the group see each week?

Keep the group view short. Each Monday the group operations lead should see three things per plant: own stop minutes against last week, the top 3 losses in money, and the open actions with owners. Plants see the same view for their own lines, so nobody is surprised by a number in a group meeting.

The group view is a starting point for questions, not a verdict. When one plant fixes a loss, the group asks which other plants have the same machine and the same loss, and copies the fix. That is where a multi site MES pays back: a fix found once and used many times.

Mistakes groups make

• Letting each plant pick its own time basis, then comparing OEE.
• Forcing one detailed code list on every plant, so operators stop using it.
• Ranking plants on raw OEE and blaming the plant with the hardest product mix.
• Buying a group system before one plant proved the routine works.

Where Fabrico fits

Fabrico is a computer vision enabled Manufacturing Performance Platform. OEE and light MES come first: automatic downtime logging that keeps own stops, starved and blocked time apart, and OEE live by line and shift.

• One view across plants, with OEE live by line and shift.
• Own stops apart from starved and blocked, so plant rankings are fair.
• Losses priced in money and ranked across the group.
• The operator confirms the reason in one tap, with a camera clip as evidence where a camera watches the line.

In ISA-95 terms Fabrico works in the Level 2 to Level 3 space; it is not an ERP. There is no free plan and there are no tiers: each plant gets a quote. See our light MES page for the full picture.

Frequently asked questions

What is a multi site MES?

An MES that runs in several plants on one shared standard, so the group can see and compare losses across sites in one place.

Should every plant use the same reason codes?

Use the same top level codes everywhere. Let each plant add detail codes under them.

Is raw OEE a fair way to rank plants?

No. Product mix, batch size and line speed differ. Rank on own losses, priced in money, and on the trend.

Which plant should go first?

The plant where a lost hour costs the group the most and the team is ready. Its setup becomes the group template.

Does a multi site MES replace ERP?

No. MES sits at ISA-95 Level 3, between ERP at Level 4 and machine control at Levels 0 to 2.

See your group in a demo

Bring one line from two plants. We will show both on one screen, on the same loss rules.

Request a demo · Contact us