MES Case Studies — Real Data
OEE trajectory with ProMOS®
observed in the field
Four real implementations of the ProMOS® Production module — from plants leaving paper behind to legacy MES migrations. No generic promises: month-by-month evolution, with the numbers we observed.
Request a Demospecialized nutrition
average across all lines
coming from manual management
in a legacy MES migration
About this page
Where these numbers come from
ProMOS®, the MES by H2W SYSTEMS®, records production for dozens of plants in 8 countries: production entries, stops and efficiency, shift by shift. This lets us track how each customer’s OEE evolves — from the first month of implementation to years of mature operation.
Below, the general pattern we observe and four real cases, anonymized, each in a different situation: a plant leaving paper behind, a plant in its second year of use, a new line without MES, and a migration from another MES to ProMOS®.
Observed pattern
The typical trajectory of H2W customers
For plants coming from manual management, the pattern repeats with variations by industry and management maturity.
Typical cumulative OEE gain for plants coming from manual management using the H2W MES
The window where the MES makes the raw difference is between month 6 and month 24. Before that, it is a learning curve; afterwards, the system becomes infrastructure and growth depends more on management than on the tool.
For plants migrating from another MES to ProMOS®, the pattern is different: OEE preserved from month 1, with no new learning cycle — operators already understand the concept of production entries. The extra gain comes from the real customization ProMOS® allows and from adding other controls to the same operator routine — such as quality and safety controls.
Case 1
The fastest learning curve in the portfolio
Scenario: specialized nutrition industry, 2 high-speed filling lines. ProMOS® implemented in August/2023. A concentrated, disciplined operation — small team, management rituals already well established before the implementation.
Performance / Reliability of the main line — first 5 months of ProMOS®
| Month | Performance / Reliability | Efficiency | Volume | Comment |
|---|---|---|---|---|
| Month 1 | 73.9% | 88.7% | 1.7 M units | Start — baseline |
| Month 2 | 87.4% | 91.6% | 2.0 M units | Learning curve begins |
| Month 3 | 88.2% | 92.3% | 2.4 M units | Peak — +14 pp in 3 months |
| Month 4 | 77.2% | 86.2% | 2.2 M units | Variation driven by product mix |
| Month 5 | 84.3% | 88.1% | 1.4 M units | Stabilizes |
| Following year (average) | 77.8% | 84.0% | 2.2 M units | Mature regime |
| 2 years later (average) | 74.9% | 87.0% | 2.5 M units | Volume +30% |
Evidence: +14 percentage points of Performance in 3 months on the main line — the fastest curve we have observed in the portfolio. Over 2 full years, produced volume grew 30% with no proportional increase in capacity.
Case 2
Broad evolution from year 1 to year 2
Scenario: long-life dairy filling plant, more than 15 lines (aseptic filling, form-fill-seal, HDPE, carton, pouch), all running ProMOS® for over a year. Analysis of the full base year vs. the following year.
| Indicator (whole plant) | Base year | Year +1 | Δ |
|---|---|---|---|
| % Operational Utilization | 45.7% | 40.2% | −5.5 pp |
| % Performance / Reliability | 76.5% | 76.7% | +0.2 pp |
| % Production Efficiency | 76.5% | 88.5% | +12.1 pp |
The plant ran 6% less volume in year +1, but with fewer shifts and more efficiency per worked hour. Management used ProMOS® data to concentrate production and cut idle capacity.
Every line present in both years gained efficiency (minimum +5.96 pp, maximum +22.96 pp). It was not a one-off gain from an exceptional line — it was broad, distributed evolution.
Evidence: +12 percentage points of Production Efficiency in 12 months (plant average) — with broad coverage, not depending on outliers.
Case 3
First implementation, coming from manual management
Scenario: Brazilian frozen bakery industry, 7-8 lines. Before ProMOS®, no robust MES — partial data entry, manual records, an “official” OEE full of noise. Implemented in 2022.
| Period | OEE | Availability | Performance | Quality | Comment |
|---|---|---|---|---|---|
| 1st half (implementation) | 55.1% | 67.1% | 84.0% | 99.5% | Baseline — partial data entry |
| 2nd half | 65.4% | 72.4% | 90.5% | 99.8% | +10 pp in 6 months |
| 1st full year | 68.9% | 72.0% | 96.5% | 99.2% | Peak — mature operation |
| Year of the new line | 60.8% | 69.6% | 88.7% | 98.5% | New line without MES on day 1 pulled the average down |
| Following year | 61.9% | 68.8% | 91.7% | 98.2% | New line climbed to 65% |
First insight: +10 OEE points in 6 months of implementation — the window where the gain does not come from physical improvement, it comes from operators no longer skipping small stops and managers seeing it in real time. The component that rose the most was Performance (84% → 96.5% in year 1), exactly where manual data entry distorts the real number the most.
Second insight: two years later the plant commissioned a new line — without ProMOS® from day 1, even though the system was already running on the other lines. It repeated the learning cycle: started at 50%, climbed to 65% in 12 months. The plant’s average OEE dropped temporarily because the weighted average included a line still in its learning curve.
Evidence: +10 percentage points of OEE in 6 months in the first year. And the MES gain is not transferred between lines by osmosis — each line needs to start with the system from day one to shorten the curve.
Case 4
Legacy MES migration to ProMOS®, with no disruption
Scenario: Brazilian plant of a multinational dairy company, 23 lines, running for many years on a proprietary MES developed in-house by another multinational in the industry. Operators experienced in the legacy tool, mature processes, reliability already stabilized at ~70%.
The migration to ProMOS® was done in a short window. Reliability did not drop during the transition — it kept the mature regime of the previous MES from the very first month.
Reliability month by month over the 10 months post-migration — no transition step-down
| Window | Reliability | OEE | Status |
|---|---|---|---|
| Pre-migration (legacy MES, mature regime) | ~70% | ~68% | Baseline |
| Month 1 on ProMOS® | 70.8% | 68.0% | No disruption |
| Month 3 | 72.9% | 70.7% | Peak of the period |
| Month 7 | 65.8% | 63.1% | Seasonal dip — not migration |
| Month 10 | 69.6% | 67.7% | Stabilized |
| 10-month average | 70.5% | 68.8% | Stable |
Evidence: the migration from a legacy MES to ProMOS® preserved reliability at ~70% and OEE at ~69%, with no operational disruption and no production loss. Fast adoption by operators experienced in another tool.
Cross-case analysis
Where the biggest improvement lever is
Crossing the 4 cases, Performance is the component that rises the most — not Availability, not Quality.
In Case 1, Performance jumped +14 pp in 3 months. In Case 2, the aggregate jump was in Production Efficiency (+12 pp), while pure Performance stayed practically flat — the plant eliminated unproductive shifts instead of improving machines. In Case 3, Performance went from 84% to 96.5% in year 1, while Availability rose only 5 pp and Quality was already 99%+.
For plants where Availability and Quality are already reasonable (most mature CPG, dairy and bakery operations), the biggest promise of the MES is capturing micro-stops and reduced speeds that manual data entry never sees. That is exactly what distorts estimated OEE the most.
Practical lessons
Three things that accelerate the curve
The operator builds the result — not just receives it
In ProMOS® results are per shift and per operator, and part of the data is entered by the operator themselves. There is no magic number displayed ready-made: the operator builds the result every shift and goes home knowing whether the day was good or bad — and whether their efficiency was above or below the line average.
Configure stop reasons with the plant team before go-live
Generic configuration produces poor data entry in the first month — and a poor first month delays operator engagement.
Daily production meeting reading the dashboard together
Plants with this routine reach month 12 at +12-15 pp. Plants that only email a weekly report reach it at +6-8 pp.
Want to understand what your trajectory would look like?
Every plant has its starting point. Plants leaving manual management see gains within months; plants migrating from a legacy MES preserve their current regime and gain flexibility. Book a demo and let’s discuss the data from your most critical line.
Request a Demo