One governed data layer before the first agent

"Can you deliver 12,000 units by 15 November?" depends on ERP, MES, planning tools, spreadsheets and people. An agent on top of all that only makes the fragmentation conversational. Build one governed data layer for the decision first, and the agent becomes the easy part. This is the pattern we follow; the details change by plant.

Stojan VeličkovićPractice Lead for Cloud, Software and Data PlatformsPublished 15 October 2026 · 2 min read · Manufacturing

01

What one question touches

InputUsually lives inWhat we often find
Open orders and promised datesERPPromise dates edited by hand, with no history
Capacity and line statusMESDifferent line and product codes in each plant
Material on hand and inboundERP or WMSStock counts that lag the floor
Planned maintenanceCMMSNot linked to the production plan
Shift staffingPlanning spreadsheetsOne person's file, updated when they remember
The real constraintsPeople's headsNever written down

02

What "governed" means here

—

One definition per term.What "available capacity" means is agreed with operations and written once, not in each prompt.

—

One owner per source,who decides which version is right when two systems disagree.

—

Lineage.Every number the agent uses traces back to a system and a timestamp.

—

Access.The layer applies the same permissions as the people asking.

—

Freshness targetsper source, matched to how often the decision is made.

03

The build order

01

Pick one decision,such as the delivery date promise. Not "all plant data".

02

Map the inputsit needs, and name an owner for each.

03

Reconcile the codesfor plants, lines and products into one model, in the warehouse or lakehouse you already run.

04

Compute in code.Capacity, material and schedule checks run as planning and optimization logic, not inside the language model.

05

Expose a few governed views or APIs.

06

Only then add the agent.It asks, calls the views, and explains the answer.

04

Shop-floor data comes in carefully

Historian and SCADA data is read-only to this layer. Following the Purdue model, data flows up from the plant network through a demilitarized zone, and nothing flows back to the line. Where data has to stay on site, the processing runs at the edge.

05

Start small, let it grow

The layer is scoped to the first decision. Each next use case reuses what is already reconciled, so it costs less, which is how one system makes the next one cheaper. For a first decision, that is a matter of weeks, not a data programme.

The language model is the interface. Planning, forecasting and optimization compute the answer.

Practice Notes · for CTOs and tech leadsMore in Insights →

Ask how this applies to your own stack.

Schedule a call with a partnerFree until 11 December