Production and operational data strategy for manufacturers. Covers why MES and ERP figures don't reconcile, who owns the handoff, and how to govern work order traceability, yield, scrap, and machine data without replacing systems.
When production data from MES and cost data from ERP do not reconcile, month-end becomes a manual exercise. When yield and scrap are defined differently in two systems, variance reports are contested rather than acted on. When work order status is out of sync, scheduling and cost reporting both suffer.
These are governance problems — ownership, definitions, and handoffs — not integration problems. Production and operational data strategy addresses how shop-floor data flows from capture to decision, covering work orders and batch traceability, machine data integration, yield and scrap reporting, and the MES-ERP boundary.
Production operations run across ERP, MES, SCADA or IoT, and often spreadsheets for operator logs or quality checks. Each system records for its own purpose; none serves as the source of truth.
Data lives in silos. Work order status sits in MES. Cost and variance data sits in ERP. Machine utilisation sits in SCADA or IoT. Yield and scrap are sometimes in MES, sometimes in spreadsheets, sometimes in quality systems. When a question spans two systems — such as the true cost of a production run or the root cause of a quality defect — the answer requires manual reconciliation.
Definitions vary between systems. A “completed work order” in MES may mean the operator marked it done. In ERP, it may mean the material has been backflushed. A “machine outage” in SCADA may be categorised differently from downtime logged in MES. The same term means different things in different places.
No one owns the handoff. Operations owns MES data. Finance owns ERP cost data. The data that falls between them — production variance, scrap reconciliation, material consumption accuracy — belongs to no one. These gaps are where errors accumulate.
Work orders and batch records are the primary unit of production accountability. They link planned production to actual output, materials consumed, labour time, and quality results. Traceability — the ability to follow a product or batch back through its production history — depends on work order and batch data being complete, consistent, and linked across systems.
Governance questions include: What is the authoritative work order record? Where does it live? How does it flow from ERP (planning) to MES (execution) and back? Who owns its accuracy? When definitions diverge between ERP and MES, which system is correct?
Machine data — cycle times, utilisation, stoppages, condition signals — supports efficiency reporting, predictive maintenance, and capacity planning. In many environments, this data is captured by SCADA, IoT platforms, or machine-specific systems that do not integrate cleanly with MES or ERP.
Governance questions include: Who owns machine data? How are stoppages categorised and attributed? What definitions of “uptime,” “availability,” and “OEE” are used, and are they consistent with finance and operations? How is machine data reconciled with production output for variance analysis?
Yield and scrap data drives cost control, process improvement, and material variance reporting. When yield is recorded in MES, scrap in a separate system, and variances calculated in ERP, the three rarely align. Month-end reconciliation becomes a negotiation rather than a fact.
Governance questions include: Who owns yield and scrap data? At what point is it captured — at the machine, at quality inspection, at packing? How is it categorised? How does it flow to ERP for cost and variance reporting?
The handoff between MES and ERP is the critical governance boundary. ERP sends work orders and planned quantities. MES reports actual production, material consumption, and labour. ERP uses this data for cost allocation, inventory updates, and variance reporting.
Each handoff is a point where data can degrade. Work order status may not sync. Material consumption may be estimated rather than actual. Timing differences may create temporary mismatches. Effective production data strategy includes defining how this handoff works, who monitors it, and how exceptions are resolved.
Many manufacturers reconcile production data, material consumption, and variance figures using spreadsheets maintained by individuals. The logic is not documented. When the person who maintains it leaves, the reconciliation process leaves with them. Finance and operations may use different reconciliation approaches, producing figures that do not match.
Batch and work order traceability is required for quality, regulatory compliance, and customer requirements. When traceability data is fragmented across MES, quality systems, and ERP, linking a defect or complaint back to production conditions becomes a manual investigation. Root cause analysis lacks a reliable data foundation.
OEE, utilisation, and efficiency metrics are often calculated differently by operations, finance, and maintenance. The same production run can yield different figures depending on who calculates and which system they use. Performance management and capacity planning rest on numbers that cannot be validated.
Real-time production dashboards can create a false sense of control. When the underlying data — work order status, machine status, yield — is inconsistent or ungoverned, dashboards display compiled noise rather than actionable truth. Decisions made from these dashboards inherit the underlying data problems.
Production and operational data strategy does not require a new MES or ERP. It requires deliberate decisions on five things.
Assign data ownership for production domains. Name an owner for work orders, batch traceability, machine data, and yield/scrap. Define what ownership means: accuracy, completeness, timeliness, and dispute resolution at the MES-ERP boundary.
Define the authoritative record. For work orders, batches, and production status, which system is correct when MES and ERP disagree? The answer must be defined in advance, not negotiated at month-end.
Establish quality controls at capture. Implement validation in MES and at the point where production data is recorded. Prevent errors at entry rather than correcting them in variance reports.
Document the MES-ERP handoff. For each data flow between systems, document what is expected, who monitors it, and how exceptions are handled. This is not an integration project — it is an ongoing governance responsibility.
Standardise definitions across functions. Uptime, availability, OEE, yield, scrap, and variance should have agreed definitions. Operations, finance, and maintenance must use the same language.
Production data sits within manufacturing data strategy. It feeds supply chain data integration, asset and maintenance data oversight, and data-driven decision frameworks. Without governed production data, capacity planning and variance reporting lack credibility.
The month-end variance reconciliation example illustrates a manufacturer where MES-ERP misalignment was adding seven days to the close. The fix was governance, not a system replacement.