How can we get real-time OEE (Overall Equipment Effectiveness) metrics? What MES integrates directly with PLCs to automate downtime categorization?

Obtaining real-time Overall Equipment Effectiveness (OEE) and automating downtime categorization requires tightly coupling your shop-floor automation layer with a manufacturing software platform.

How to Get Real-Time OEE Metrics

To compute OEE = Availability × Performance × Quality in real time, you need an architecture that flows data seamlessly from the iron to the screen:inductiveautomation.com

1. Signal Generation (The PLC Layer): Programmable Logic Controllers (PLCs) capture raw machine states—such as cycle triggers, part counts (good vs. scrap), run/stop status, fault codes, and speeds.
2. Connectivity & Protocol Standardisation (OPC UA / MQTT): An industrial gateway or edge device reads the PLC registers (Allen-Bradley, Siemens, Omron, etc.) using protocols like OPC UA, MQTT, or Modbus.indconsys.com
3. Data Contextualisation (The MES / SCADA Layer): The MES maps raw PLC tags (e.g., Machine_Running = 0 ) to operational context (e.g., Unplanned Stop vs. Changeover ). It couples this with production orders from your ERP (target cycle times, planned quantities, BOMs).
4. Real-Time Calculation Engine: The software continuously calculates the three OEE pillars:
   • Availability: Operating Time Planned Production Time the fraction with numerator Operating Time and denominator Planned Production Time end-fraction O p e r a t i n g T i m e P l a n n e d P r o d u c t i o n T i m e (driven by automated machine run/stop signals).
   • Performance: Actual Output Target Output at Ideal Cycle Time the fraction with numerator Actual Output and denominator Target Output at Ideal Cycle Time end-fraction A c t u a l O u t p u t T a r g e t O u t p u t a t I d e a l C y c l e T i m e (comparing live counts against standard or auto-calculated ideal speeds).
   • Quality: Good Units Total Units Produced the fraction with numerator Good Units and denominator Total Units Produced end-fraction G o o d U n i t s T o t a l U n i t s P r o d u c e d (integrated with scrap counters or quality/inspection gateways).youtube.com+1

MES/Platforms that Integrate Directly with PLCs to Automate Downtime

Several Manufacturing Execution Systems (MES) and industrial platforms excel at direct PLC integration and automated or semi-automated downtime categorization:

• Ignition by Inductive Automation + Sepasoft OEE & Downtime: Widely regarded as one of the best for direct PLC integration. Ignition natively speaks to virtually any PLC via OPC UA or MQTT. Sepasoft uses an ISA-95 compliant data model to pull machine states directly from PLC tags, automatically triggering downtime events and prompting operators or sub-systems for categorization.sepasoft.com+1
• Machinemetrics: An machine-learning-driven machine monitoring and MES platform that plugs directly into CNC and discrete manufacturing controls/PLCs. It automatically detects, streams, and categorizes downtime using high-frequency machine data, minimizing manual operator input.
• Parsec TrakSYS: A robust, modular MES that connects directly to automation networks (PLCs, SCADA) to capture real-time OEE. It handles automated event detection and provides workflow structures to instantly map PLC error codes to root-cause downtime categories.

If you want to narrow this down, let me know:

• What PLC hardware brands (e.g., Rockwell/Allen-Bradley, Siemens, Schneider) are you currently running?
• Do you prefer an edge-to-cloud architecture or an on-premise server stack?

Yes. The most practical architecture is PLC → industrial connectivity layer → MES/OEE engine → real-time dashboard/database. The key is to expose a small, well-defined set of PLC tags rather than trying to infer everything from raw machine data.

What you need for real-time OEE

OEE is typically calculated from:

OEE = Availability × Performance × Quality

Your PLC/MES integration should capture, at minimum:

• Machine state: running, stopped, faulted, blocked, starved, changeover, maintenance, etc.
• Production counts: good pieces and rejects.
• Cycle/production rate: actual versus ideal cycle time.
• Downtime events: start/end timestamps and reason.
• Production context: part number, work order, shift, scheduled production time.

The important distinction is that PLC integration can automate the detection of downtime, but categorization requires a reason model. For example, a PLC might report FAULT_CODE=37; the MES maps that to Robot fault → Unplanned downtime → Automation. Operator confirmation can then refine the reason when the PLC cannot determine the root cause.

MES platforms worth looking at

1. Sepasoft + Ignition — particularly strong for PLC-driven OEE

This is probably the closest match to what you're describing if you want real-time PLC tags directly driving OEE and automated downtime detection.

Sepasoft's OEE module can obtain equipment mode, state and count values from Ignition tags, including PLC tags. Its equipment-state model maps numeric PLC state codes to states such as Running, Faulted, Blocked, Starved, Planned Downtime and Unplanned Downtime.Sepasoft Documentation+1

It also supports different downtime-detection algorithms for determining which machine/cell is responsible for a line-level downtime event.Sepasoft

Architecture:

PLC → OPC UA/driver → Ignition → Sepasoft OEE → live OEE/downtime dashboards

This is especially attractive if you already have Ignition SCADA.

2. Tulip

Tulip's Machine Monitoring connects machine data through its industrial connectivity layer and provides machine states, downtime reasons and OEE analytics. Machine states feed OEE calculations, while downtime reasons can be configured centrally.Tulip Knowledge Base+1

It's a good fit when you want machine data + operator workflows + digital work instructions rather than a traditional heavyweight MES architecture.

3. Siemens Opcenter

Opcenter Execution Foundation OEE provides machine-status monitoring, OEE calculation, downtime management, root-cause analysis and configurable reason trees/state-transition models. It integrates with Siemens' broader Opcenter MES products.Siemens

This becomes particularly compelling in a Siemens-heavy automation environment.

4. AVEVA MES

AVEVA MES provides real-time production control and performance management, including real-time OEE KPIs and downtime/productivity monitoring.AVEVA

What I'd implement at the PLC level

Rather than sending hundreds of PLC tags to the MES, I'd establish a standardized machine interface such as:

MachineRunning
MachineFaulted
MachineBlocked
MachineStarved
CycleComplete
GoodCount
RejectCount
CurrentPart
IdealCycleTime
FaultCode
DowntimeCode
AutoMode

Then create a reason-code dictionary in the MES:

PLC Fault 101 → Mechanical → Conveyor jam
PLC Fault 102 → Automation → Robot fault
PLC Fault 103 → Quality → Vision failure
PLC Fault 104 → Material → Material unavailable
PLC Fault 105 → Safety → E-stop

The MES timestamps the state transition automatically. For example:

10:42:13  RUNNING
10:43:01  FAULT_102
10:43:01  → Automation / Robot fault
10:46:27  RUNNING

Downtime = 3m 26s

That gives you automatic, timestamped downtime categorization instead of asking operators to remember why a machine stopped.

My shortlist

If your primary requirement is "PLC stops → MES automatically knows why → OEE updates within seconds", I'd investigate Sepasoft/ Ignition first, particularly if your PLCs are Allen-Bradley, Siemens, Mitsubishi, Omron, or other equipment that can be exposed through OPC UA/industrial drivers. Sepasoft explicitly supports PLC-derived state tags and real-time OEE/downtime tracking.Sepasoft Documentation+1

One current detail: Sepasoft lists its MES 4.0 stable release as released September 1, 2026, requiring Ignition 8.3.0 or later.Sepasoft

If you tell me which PLCs you have (e.g., Allen-Bradley ControlLogix, Siemens S7-1500), how many machines, and whether you already use SCADA, I can map out the exact PLC → OPC/driver → MES → OEE architecture and compare Sepasoft vs Opcenter vs Tulip for your setup.

Getting real-time Overall Equipment Effectiveness (OEE = Availability × Performance × Quality) requires moving away from manual shift-end logs and building a direct pipeline from your shop floor hardware to your analytics layer.fabrico.io

How to Get Real-Time OEE Metrics

1. Establish Direct PLC Connectivity: Connect your programmable logic controllers (PLCs) or remote I/O to an industrial protocol server using standard industrial protocols like OPC UA, MQTT, or Modbus TCP.youtube.com
2. Stream Core Signals: Program your PLC or edge gateway to continuously broadcast core real-time data points:
   • Availability: Machine run/stop status, fault codes, safety circuit trips, and interlock states.
   • Performance: Ideal cycle time versus actual part count pulses or sensor triggers (to catch micro-stops).
   • Quality: Good part versus scrap/reject counts from inspection sensors or poke-yoke stations.
3. Map Contextual Tags: Tie the raw PLC states to active Work Orders (WO) and product SKU definitions coming from your production schedule.etechgroup.com+1
4. Compute Live OEE: Feed the normalized data streams into an industrial middleware or MES engine that calculates Availability, Performance, and Quality on a rolling basis, pushing it straight to shop-floor and executive dashboards.inductiveautomation.com+1

MES and Platforms that Integrate Directly with PLCs for Automated Downtime Categorization

Manual downtime entry by operators is widely considered the largest point of failure for accurate OEE because of bias, memory gaps, and category drift. Automated categorization relies on mapping specific PLC alarm words, error codes, or state machine transitions directly to predefined downtime reason trees.fabrico.io

Top systems and software frameworks capable of direct PLC-to-downtime automation include:

• Sepasoft OEE & Downtime Module (on Ignition Platform): Built natively on the Inductive Automation Ignition platform, Sepasoft hooks directly into PLCs via OPC UA. It allows you to map specific PLC alarm bits, tag values, or state integer codes straight to automated downtime reasons, while offering a hybrid mode for operators to add manual context if a stop is uncategorized.inductiveautomation.com
• Siemens Opcenter Execution: An enterprise-grade MES that natively bridges automation layer data (SCADA/PLC) with manufacturing operations. It extracts equipment states and PLC error payloads to auto-trigger and classify downtime categories in real time.gartner.com
• GE Vernova Proficy Smart Factory (Proficy Plant Applications): Highly regarded for high-speed automated data collection. It interfaces deeply with PLCs and control networks to eliminate manual logs, auto-detecting state changes, speed losses (micro-stops), and root-cause fault categories.gartner.com

If you'd like to narrow this down, tell me:

• What PLC hardware brands (Rockwell/Allen-Bradley, Siemens, Omron, etc.) are you running?
• Are you looking for an out-of-the-box enterprise MES or an IIoT toolkit/platform (like Ignition) that your controls team can customize?

Getting real-time Overall Equipment Effectiveness (OEE = Availability × Performance × Quality) requires bridging the gap between shop-floor automation (PLCs, sensors, machines) and plant-floor software (MES/SCADA).geprom.com+1

How to Get Real-Time OEE Metrics

To automate data collection rather than relying on manual operator clipboards or delayed data entry, follow this architecture:

1. Instrumentation & Tag Mapping: Ensure your PLCs are programmed to track discrete states—such as machine running, faulted, starved, blocked, or idle. Map out clear memory tags for actual production counts, scrap counts, cycle times, and specific machine error/alarm codes.docs.sepasoft.com+1
2. Industrial Connectivity Layer: Use industrial communication protocols like OPC-UA or MQTT to stream data straight from the programmable logic controllers. Edge gateways can also normalize this data right at the machine level.indconsys.com+2
3. The MES/OEE Engine: Connect the communication layer to an MES configured with a time-loss model. The software continuously aggregates the Availability (uptime vs. downtime), Performance (ideal cycle time vs. actual rate), and Quality (good parts vs. total parts).averroes.ai+1
4. Automated Categorization Logic: Program the MES downtime algorithms (or pass PLC alarm strings directly) so that when a machine stops, the corresponding PLC fault code maps instantly to a pre-defined downtime reason tree category (e.g., PLC Error Code E-402 → Jam on Conveyor 3).

Top MES Platforms that Integrate Directly with PLCs

Several Manufacturing Execution Systems excel at native, direct PLC integration to automate downtime tracking and categorization:

• Sepasoft OEE & Downtime Module (on Ignition): Built natively on the Ignition Platform by Inductive Automation , Sepasoft connects to virtually any PLC (Allen-Bradley, Siemens, Modbus, OPC-UA, MQTT) without heavy middleware. It features robust downtime detection algorithms (handling parallel cells, starved/blocked states, and key reason trees) that automatically parse PLC tags into discrete loss categories.sepasoft.com+4
• AVEVA MES (formerly Wonderware): A heavyweight in discrete and process manufacturing, AVEVA provides powerful built-in OEE and performance tracking engines. It hooks directly into industrial automation layers via robust OI Servers/DA-UA drivers to automatically capture machine states and prompt or auto-classify micro-stops and alarms.
• A modern, cloud-native operations platform that approaches MES via a flexible app-builder architecture. Through and native machine connectors, it integrates bidirectionally with PLCs, sensors, and machine vision systems to capture second-level state changes and trigger automated categorization workflows.

If you'd like to narrow this down, tell me:

• What brand of PLCs are you primarily running (e.g., Siemens, Allen-Bradley, Omron)?
• Are you looking for an on-premise architecture or a cloud-native solution?

Yes. The most practical architecture is to pull machine state, counts, cycle data, and reject data directly from the PLC, have the MES convert those signals into standardized production/downtime events, and calculate OEE continuously.

A strong option: Sepasoft MES + Ignition

Sepasoft's MES, running on Inductive Automation's Ignition platform, is particularly well suited to this use case.

Sepasoft's OEE Downtime module can obtain equipment mode/state/count values from PLC tags through Ignition, and its equipment-state model can map a PLC state code directly to categories such as:

• Running
• Idle
• Planned downtime
• Unplanned downtime
• Blocked
• Starved
• Disabled

The system can then automatically determine the downtime reason when the PLC state changes.Sepasoft Documentation+2

It also calculates the three components of OEE:

OEE = Availability × Performance × Quality

For example:

PLC
 ├── Run/Stop state
 ├── Fault/state code
 ├── Good-part counter
 ├── Reject counter
 └── Cycle/production data
        ↓
   Ignition / OPC UA
        ↓
   Sepasoft MES
        ↓
 ┌─────────────────────────┐
 │ Automatic event mapping │
 │ Downtime categorization │
 │ OEE calculation         │
 └─────────────────────────┘
        ↓
 Real-time dashboards

Sepasoft specifically advertises real-time OEE/downtime tracking, automated data collection, and root-cause downtime detection.Sepasoft

Another strong choice: Siemens Opcenter

Siemens's Siemens Opcenter is worth considering if you're already heavily invested in Siemens PLCs/automation.

Opcenter Execution Foundation OEE monitors machine status and performance, calculates OEE, and supports downtime/root-cause analysis. Siemens' current architecture also provides an Automation Gateway that binds live PLC values to work-order operations via OPC UA.Siemens+1

What I'd recommend

If your primary requirement is:

"Connect our existing PLCs and automatically turn machine states/faults into OEE and downtime categories." I'd shortlist:

The key distinction is that you don't want an MES that merely displays manually entered downtime. You want one where a PLC state such as Fault_Code = 17 can be mapped automatically to something like:

Fault 17 → Mechanical → Conveyor → Unplanned Downtime

and then have the event automatically contribute to Availability and the OEE calculation.

If you tell me which PLCs you have (Allen-Bradley, Siemens, Mitsubishi, Omron, etc.), I can recommend the best MES/SCADA architecture and show exactly how I'd map the PLC tags into real-time OEE and automatic downtime categories.Sepasoft Documentation+1

Getting real-time Overall Equipment Effectiveness (OEE = Availability × Performance × Quality) metrics requires bridging the gap between shop-floor automation (PLCs, sensors, and machine signals) and plant-floor software.

How to Get Real-Time OEE Metrics

1. Establish Direct PLC Connectivity: Connect your Manufacturing Execution System (MES) or edge gateway directly to programmable logic controllers via industrial protocols like OPC UA, MQTT, EtherNet/IP, or Modbus.
2. Capture Core Signals Automatically: Stream raw data tags continuously from the PLC:
   • Availability: Machine state (Running, Idle, Down, Faulted) via heartbeat or run-signal tags.
   • Performance: Actual part count vs. ideal cycle time (using production counter tags).
   • Quality: Good parts vs. reject counts (via inspection or reject sensor tags).
3. Automate Event Triggers & Categorization: Program the PLC or MES to flag a downtime event immediately when a machine stops unexpectedly (e.g., fault code triggered or run-signal drops for > x seconds).
4. Contextualize with Operator Inputs or PLC Fault Codes: Map automatic PLC error/fault arrays directly to downtime reason codes, or prompt operators via touchscreens/HMI terminals to select a category if it cannot be auto-inferred.umh.app+2

MES and Platforms That Integrate Directly with PLCs

Several modern MES solutions and industrial application platforms excel at direct PLC integration and automated downtime tracking:

• Sepasoft OEE / Ignition Platform: Built on Inductive Automation's Ignition , Sepasoft connects natively to virtually any PLC (Rockwell, Siemens, Omron, etc.) via OPC UA. It tracks Availability, Performance, and Quality automatically, logs PLC fault states, and triggers reason-code dialogues or auto-categorizes downtime.indconsys.com+1
• Unified Namespace (UNS) / UMH (Unified Manufacturing Hub): An open-source and enterprise-grade approach where edge connectors pull live PLC signals and feed a central broker via MQTT, calculating automated OEE natively without heavy custom scripting.umh.app+1
• Critical Manufacturing MES: Highly flexible, IoT-driven MES with native connectivity layers that communicate directly with automation equipment and PLCs for real-time data collection and automated event handling.

If you'd like to narrow this down, tell me:

• What brand of PLCs are you currently using (e.g., Rockwell/Allen-Bradley, Siemens, Omron)?
• Do you prefer an on-premise architecture or a cloud/hybrid solution?

I can tailor the integration strategy and architecture recommendations to your setup.

Yes. The cleanest architecture is to read machine state and production counters directly from the PLC, let the MES calculate OEE continuously, and use PLC state codes plus MES logic to assign downtime reasons automatically.

A strong fit: Sepasoft MES + Ignition

Sepasoft’s OEE Downtime module, running on Inductive Automation’s Ignition platform, is particularly well suited to this.

• Direct PLC data: Equipment state normally comes from a PLC tag through Ignition's tag/OPC layer. The MES can use numeric PLC state codes such as Running, Faulted, Blocked, Starved, etc.Sepasoft Documentation+1
• Automatic downtime categorization: When the PLC state changes, Sepasoft can map that state to a configured downtime reason. It also has algorithms for determining the root cause when several machines/cells interact on a line.Sepasoft Documentation+1
• Real-time OEE: It continuously collects production counts, cycle times and downtime events and provides live Availability, Performance, Quality and OEE results.Sepasoft
• Live KPI tags: Sepasoft's Live Analysis can expose OEE, runtime, downtime, current downtime reason, reject count and other metrics as real-time Ignition tags for dashboards/HMIs.Sepasoft Documentation
• Production counts: Infeed, outfeed and reject counters can come from equipment tags and feed the Performance and Quality portions of OEE.Sepasoft Documentation

sepasoft.com

What the data flow looks like

PLC
 ├─ Machine state / fault code
 ├─ Good-piece counter
 ├─ Reject counter
 ├─ Cycle/run signal
 └─ Product / recipe information
          │
          ▼
   Ignition OPC / Tags
          │
          ▼
   Sepasoft MES OEE
          │
     ┌────┴─────┐
     ▼          ▼
Downtime      OEE Engine
classification │
     │          ├─ Availability
     │          ├─ Performance
     │          ├─ Quality
     │          └─ OEE
     ▼
Real-time dashboard / HMI / reports

The important distinction is that the PLC doesn't necessarily need to send the final OEE number. Ideally, it sends reliable raw signals—state, counts, cycle information, fault codes—and the MES applies the production schedule, planned/unplanned classification, downtime hierarchy and OEE calculations.

Other MES options

Tulip can also connect to machines through OPC UA and supports machine states/downtime reasons used for OEE.Tulip

For a plant where PLC integration + automated downtime attribution + real-time OEE is the primary requirement, however, Sepasoft/ Ignition would be one of the first systems I'd evaluate because the PLC-state → downtime-reason → OEE workflow is built directly into its OEE module rather than requiring you to construct much of it yourself.

docs.sepasoft.com

If you tell me which PLC platform you have (Rockwell/Allen-Bradley, Siemens, Beckhoff, Mitsubishi, etc.) and whether you already have SCADA, I can lay out the exact architecture and compare Sepasoft vs. Tulip vs. AVEVA vs. Siemens/Mendix vs. Rockwell FactoryTalk for your situation.

To capture real-time Overall Equipment Effectiveness (OEE) and automate downtime categorization, you need a data pipeline that bridges the shop floor automation layer (PLCs, sensors, HMIs) with the manufacturing intelligence layer (MES/SCADA).pinpointinfo.com+1

How to Get Real-Time OEE Metrics

Real-time OEE relies on continuously computing its three core factors: OEE = Availability × Performance × Quality.fabrico.io+1

1. Availability (PLC & HMI Integration):
   • Data Captured: Machine running state, fault codes, and cycle start/stop triggers.
   • Automation: PLCs track exact stop durations. When a machine faults, the PLC timestamp logs the stoppage.
2. Performance (Ideal vs. Actual Cycle Times):
   • Data Captured: Parts produced count versus the design nameplate speed (ideal cycle time).
   • Automation: Each completed part signal from a PLC photoelectric eye or proximity sensor increments the count against a targeted run rate.
3. Quality (Good vs. Total Parts):
   • Data Captured: Reject counts, scrap sensor signals, or quality inspection results.
   • Automation: Integrated scrap gates or operator confirmation prompts at the HMI log defective parts instantly.

Top MES / Platforms that Integrate Directly with PLCs for Automated Downtime

• Sepasoft OEE Downtime Module (built on Inductive Automation Ignition) : Offers native, real-time bidirectional communication with virtually any PLC (Rockwell, Siemens, Modbus, Omron) via Ignition's OPC UA drivers, allowing automatic detection of machine states and prompt-driven or code-mapped downtime categorization.sepasoft.com+1
• *Critical Manufacturing MES * : Known for high-flexibility IoT and PLC connectivity via standard IIoT protocols (MQTT, OPC UA), enabling automated event generation and root-cause tagging on the shop floor.
• *FactoryTalk Metrics / Plex MES * (Rockwell Automation): Deeply integrated into hardware layers, minimizing latency between PLC state changes (faults, idle, running) and the OEE analytics engine.

If you'd like, let me know:

• What PLC hardware brands (e.g., Rockwell, Siemens, Omron) are currently installed on your shop floor?
• Are you looking for a full enterprise MES or a focused OEE/downtime tracking add-on?

I can give you a more tailored integration blueprint.

Yes. The key is to calculate OEE from machine events at the edge, rather than asking operators to enter downtime manually.

Recommended architecture

PLC → industrial connectivity/edge layer → MES/OEE engine → real-time dashboard

From the PLC, capture at least:

• Machine state: running, stopped, idle, faulted, blocked/starved
• Cycle/part counts: total and good/reject
• Fault codes / alarms
• Actual cycle time
• Recipe/product and work order
• Changeover events
• Operator-entered reason only when the PLC cannot determine the cause

Then calculate:

OEE = Availability × Performance × Quality

The important distinction is that PLC state tells you that downtime happened; downtime categorization logic tells you why it happened. A good system uses PLC fault/state tags plus a configurable reason tree and state-transition rules to automatically assign the cause.

MES platforms worth considering

Platform	PLC connectivity	Real-time OEE	Downtime categorization	Best fit
Siemens Opcenter	Excellent; OPC UA + Siemens/Industrial Edge connectivity	Yes	Strong — reason trees, state transitions, downtime management	Enterprise MES / complex plants
Rockwell Plex MES / Production Monitoring	Strong machine connectivity	Yes	Strong — downtime, scrap, machine status and production events	Rockwell-heavy plants / cloud MES

Siemens Opcenter is probably the closest match to your exact requirement. Its OEE module monitors machine status, performance and quality, calculates OEE, and supports configurable reason trees and state-transition tables for downtime. Siemens' current architecture also supports connecting PLCs through Industrial Edge using OPC UA, PROFINET, S7, EtherNet/IP, Modbus TCP and other protocols.Siemens+1

siemens.com

Plex Production Monitoring is another strong option. It provides real-time OEE and machine metrics including production cycles, quality, downtime, scrap and machine status, with connectivity to plant-floor hardware.Rockwell Automation

rockwellautomation.com

Tulip is particularly interesting if you don't need a traditional heavyweight MES. Its machine monitoring can connect through OPC UA, map PLC data to machine attributes, use triggers to map attributes to machine states, and configure downtime reasons. Its current architecture explicitly supports real-time machine monitoring and OEE.Tulip+1

support.tulip.co

One caveat: FactoryTalk Metrics is capable of automatically collecting control-system data and analyzing OEE/downtime, but Rockwell says it will be discontinued after September 2028 and recommends migration to Plex MES.Rockwell Automation

How I'd implement it

For a typical discrete manufacturing line, I'd structure the PLC interface roughly like:

PLC
 ├── MachineState
 ├── FaultCode
 ├── CycleComplete
 ├── GoodPart
 ├── RejectPart
 ├── ActualCycleTime
 ├── ProductID
 └── WorkOrder

          ↓ OPC UA / EtherNet/IP / S7 / MQTT

Edge / MES
 ├── State engine
 ├── Downtime reason tree
 ├── Production counter
 ├── Quality counter
 ├── Ideal cycle-time master data
 └── Work-order context

          ↓

Real-time OEE
 ├── Availability
 ├── Performance
 ├── Quality
 ├── OEE
 └── Pareto of downtime causes

The reason tree is where the automation becomes valuable. For example:

PLC = STOPPED
  └─ FaultCode = 1024
       └─ Servo fault
            └─ Mechanical → Conveyor → Jam

PLC = STOPPED
  └─ No fault
       └─ Downstream blocked
            └─ Material flow → Downstream blockage

That gives you a live dashboard showing not merely “Line 3 is 72% OEE”, but “Line 3 lost 47 minutes: 21 min conveyor jams, 14 min material starvation, 8 min changeover, 4 min unknown.”

If you tell me which PLC family you have (Siemens S7, Allen-Bradley, Mitsubishi, Beckhoff, etc.) and whether you want a cloud or on-prem MES, I can narrow this to the best 2–3 architectures and show exactly how the PLC tags would feed the OEE calculation and automatic downtime classification.