Most manufacturers already understand the value of ERP and SCADA.

ERP helps manage business-level operations: orders, inventory, purchasing, finance, customer commitments, and planning.

SCADA helps monitor and control machines, lines, utilities, and industrial processes in real time.

But between these two layers, many factories still rely on spreadsheets, paper forms, manual shift reports, disconnected quality logs, and tribal knowledge.

That middle layer is where production actually happens.

This is the space where MES, or Manufacturing Execution System, becomes important.

ERP knows what should happen. SCADA knows what is happening.

ERP systems are strong at answering business questions:

• What orders need to be produced?
• Which materials are available?
• What is the delivery schedule?
• What does the customer expect?
• What is the cost structure?

SCADA systems are strong at answering process questions:

• Which machines are running?
• Which alarms are active?
• What are the current temperatures, pressures, speeds, and counts?
• Which equipment is stopped?
• What is happening on the line right now?

The problem is that neither system fully owns the production execution layer.

ERP usually does not understand machine-level reality in enough detail. SCADA usually does not manage work orders, material genealogy, production routes, quality records, or operator execution workflows at the business-process level.

That gap creates operational blind spots.

What happens when the MES layer is missing

When there is no proper execution layer, the factory often fills the gap manually.

Operators write downtime reasons on paper. Supervisors update Excel files after the shift. Quality teams collect inspection results separately. Maintenance teams receive downtime information too late. Production planners work with outdated capacity assumptions. Managers see performance reports only after the losses have already happened.

The result is not only inefficiency. It is delayed visibility.

A line may be underperforming for hours before anyone understands the root cause. A batch may move through production before quality deviations are connected to specific materials or process parameters. A delivery promise may be missed because scheduling was based on theoretical capacity instead of real production constraints.

For teams evaluating this missing production layer, an Iotellect manufacturing execution system can connect scheduling, OEE, traceability, quality workflows, and real-time shop-floor data — not just display another dashboard.

What MES should actually do

A practical MES should help answer several questions during production, not after production is already finished.

1. What should be produced?

MES connects production orders with actual shop-floor execution.

It helps convert plans into work that can be assigned to lines, shifts, equipment, and operators.

This includes:

• Production orders
• Product definitions
• Routes
• Recipes
• Bills of materials
• Equipment-specific parameters
• Version control
• Change approval

In industries such as food and beverage, pharma, chemicals, electronics, and automotive, this structure is especially important because small changes in recipes, components, or process steps can affect compliance, quality, and traceability.

2. Can the factory actually produce it?

Planning is easy when every resource is assumed to be available.

Real production is different.

Machines have capacity limits. Operators work shifts. Materials arrive late. Setup time matters. Maintenance windows reduce available production time. Some products can only run on specific lines or equipment.

Finite capacity scheduling helps manufacturers move from theoretical planning to realistic production planning.

Instead of overloading resources, MES can help schedule work based on actual constraints.

3. How efficiently is production running?

OEE is still one of the clearest ways to understand production performance.

A useful MES should track:

• Availability
• Performance
• Quality
• Downtime
• Output
• Scrap
• Bottlenecks

But OEE alone is not enough.

The system also needs to explain why performance is poor.

Downtime reason codes, speed losses, scrap events, quality defects, and bottlenecks must be visible while there is still time to act.

A report tomorrow is useful for analysis.

A signal during production is useful for improvement.

4. What exactly went into each product?

Traceability is no longer only a compliance topic.

It is now a business continuity topic.

Manufacturers need to know which raw materials, components, batches, lots, machines, operators, process parameters, and quality checks were involved in each finished product.

When something goes wrong, the company should not need days to investigate.

It should be able to trace affected products, batches, or serial numbers quickly and accurately.

This is especially important in regulated and quality-sensitive industries such as:

• Pharma
• Food and beverage
• Electronics
• Chemicals
• Automotive manufacturing

5. Are quality checks connected to production?

Quality management becomes much stronger when it is built into execution rather than handled separately.

Instead of recording quality checks after the fact, MES can collect:

• In-process inspection data
• SPC measurements
• Defect information
• Operator confirmations
• Deviation records
• Electronic batch records
• Audit-ready production history

This reduces the risk of paper-based errors, missing forms, delayed reporting, and incomplete audit trails.

Why ISA-95 still matters

MES projects often become expensive because every plant describes production differently.

One site may define equipment one way. Another may structure lines, work centers, materials, and operations differently.

ERP integration then becomes painful. Cross-site reporting becomes inconsistent. Rollouts become slower than expected.

ISA-95 helps by providing a common structure for manufacturing operations and enterprise-control integration.

A good MES architecture should support consistent models for:

• Equipment
• Materials
• Personnel
• Production segments
• Operations
• Enterprise asset hierarchy

This does not mean every plant must become identical.

It means every plant should be modeled in a predictable way.

That consistency helps with:

• ERP integration
• Multi-site rollouts
• Standardized reporting
• Cross-plant analytics
• Template reuse
• Cleaner long-term maintenance

Without a clean data model, MES can quickly become another silo.

MES should not be a locked box

One of the biggest MES implementation problems is rigidity.

Some systems deploy quickly but are difficult to adapt. Others are flexible but require long custom development projects before they deliver value.

Manufacturing rarely fits perfectly into a standard template.

Every plant has specific workflows, exceptions, naming rules, approval steps, quality requirements, and reporting needs.

That is why modern MES architecture should allow teams to start with ready-made modules but still adapt the logic when needed.

The ideal balance is simple:

• Use standard modules for common needs.
• Customize only where the process truly requires it.
• Avoid rebuilding the entire platform from scratch.
• Avoid waiting months for vendor-side changes.
• Keep the production logic visible and maintainable.

This is especially important for system integrators, OEMs, and manufacturing IT teams that need to deliver repeatable solutions across multiple customers, sites, or production environments.

Deployment flexibility is now a requirement

Manufacturing environments are not all the same.

Some plants want cloud-based access across multiple sites.

Some require on-premise deployment because of security, latency, or regulatory needs.

Some need edge deployment directly on industrial PCs or local hardware near the production line.

Some need hybrid architecture where local nodes continue operating during connectivity loss.

MES should fit the infrastructure strategy, not force the factory into one deployment model.

The more distributed industrial systems become, the more important this flexibility becomes.

A plant should be able to keep production execution running locally while still giving central teams visibility across operations when connectivity is available.

MES, SCADA, BI, maintenance, and edge should work together

A common problem in industrial software stacks is fragmentation.

One tool handles SCADA. Another handles MES. Another handles reporting. Another handles maintenance. Another handles analytics. Another handles edge data collection.

At first, this looks manageable.

Over time, it creates:

• Duplicated tag databases
• Repeated integrations
• Inconsistent naming
• Middleware complexity
• Unclear ownership of data
• Delayed reporting
• Expensive maintenance

The long-term goal should be a cleaner architecture where production data can move naturally between execution, visualization, analytics, maintenance, and business systems.

When MES and SCADA share the same operational data model, many things become easier:

• Machine data can support OEE automatically.
• Downtime can trigger maintenance workflows.
• Quality deviations can be linked to process parameters.
• Production reports can use real-time and historical data.
• Dashboards can serve operators, supervisors, and managers from the same source of truth.

This is where MES becomes more than a production application.

It becomes part of the industrial operating layer.

Final thought

MES is not just software for reporting what happened on the factory floor.

At its best, it is the system that connects what the business planned with what production actually executed.

It links work orders, equipment, operators, materials, quality checks, performance data, and traceability into one live operational record.

For manufacturers, the question is no longer whether production data should be digital.

The real question is whether that data is connected, structured, and actionable while production is still running.

That is where the MES layer matters most.