Rolled out a custom MES across nine plants in 14 months, lifting OEE by 11 percentage points and surviving the inevitable shop-floor reality check.
Anonymized delivery dashboard
OEE lift
+11 pts
Plants rolled out
9
Operator NPS
+46
Before
Constraint
Build
Controlled cutover
After
Measured gain
Every engagement is framed around the business situation, the constraint that made it hard, and the decision that turned delivery into a controlled path to value.
Situation
Two prior MES initiatives had stalled — one because the vendor product did not match the production model, one because the operators refused to use it. The third attempt had to ship to the floor, not just the steering committee.
Constraint
Two prior MES initiatives had stalled — one because the vendor product did not match the production model, one because the operators refused to use it. The third attempt had to ship to the floor, not just the steering committee.
Decision
Designed plant-by-plant with operators, proving the workflow on the floor before scaling to the next site.
Build
Every screen was prototyped with the operators who would use it, in the language they used. Devices were chosen with input from the plants, not procurement.
Plant 1 ran for ten weeks before plant 2 went live. Each rollout fed concrete changes back into the platform before the next plant was scheduled.
Outcome
OEE across rolled-out plants lifted by an average of 11 percentage points.
All nine plants live within 14 months of programme start.
What changed after launch
Operators became active contributors to the rollout, and each plant improved the platform before the next go-live.
“This was the first rollout where operators felt the system was built with them rather than delivered to them.”
The work was not abstract modernization. It changed day-to-day behavior, ownership, and the evidence leaders used to make decisions.
Before
After
The delivery plan made the system boundary explicit, then used rehearsals, gates, and telemetry to optimize safely before launch.
Delivery architecture
Manufacturing control loop
Discover
Every screen was prototyped with the operators who would use it, in the language they used. Devices were chosen with input from the plants, not procurement.
Launch
Plant 1 ran for ten weeks before plant 2 went live. Each rollout fed concrete changes back into the platform before the next plant was scheduled.
Every screen was prototyped with the operators who would use it, in the language they used. Devices were chosen with input from the plants, not procurement.
Plant 1 ran for ten weeks before plant 2 went live. Each rollout fed concrete changes back into the platform before the next plant was scheduled.
After launch
Operators became active contributors to the rollout, and each plant improved the platform before the next go-live.
Java
Spring Boot
PostgreSQL
Kafka
React
Azure
Kubernetes
A similar problem?
A senior engineer will follow up within one business day with an opinionated take on the shape of the work.
