Digital Workflow in Manufacturing: Definition & MES Role 2026

What is a digital workflow in manufacturing?

A digital workflow — German digitaler Workflow or digitaler Prozessablauf — is a chain of manufacturing tasks, decisions and data handoffs executed through connected software instead of paper, spreadsheets or verbal coordination. It is event-driven, machine- and user-triggered, fully logged, and runs across ISA-95 levels: PLC and SCADA at Level 1/2, MES at Level 3, ERP at Level 4.

The term is often reduced to "paperless production." That is the surface effect, not the substance. A digital workflow only works when three things are true at the same time: data arrives in real time from the machine, business rules fire automatically on defined events, and every step writes back into the system of record. Based on building the SYMESTIC platform across 15,000+ connected machines in 18 countries, the failure mode is always the same — two of the three exist, the third is missing, and the workflow collapses into a dashboard with manual follow-ups.

Digital workflow vs. BPM vs. RPA vs. MES workflow

Four terms describe overlapping but distinct concepts. Treating them as synonyms is the fastest way to buy the wrong tool.

An MES workflow is a digital workflow specifically orchestrated by the Manufacturing Execution System — setup release, scrap confirmation, first-piece approval, shift handover, alarm escalation. It is the concrete, production-grade instantiation of the more general "digital workflow" concept.

The four building blocks that actually make a workflow digital

On paper, any tool with a form and an email trigger is marketed as a "digital workflow." In a production environment, four elements have to be present — miss any one and you end up with disguised manual work.

1. Real-time machine signal. Cycle counts, stops, setup states and alarms have to arrive from the PLC via OPC UA, MQTT or a digital I/O gateway. Without this the workflow runs on operator entries hours after the fact.
2. Event-driven rule engine. Conditions like "stop > 5 min → escalate to shift lead," "scrap reason X → trigger quality check," "first piece approved → release full batch." These fire automatically. No one clicks "run."
3. Bidirectional integration with the system of record. ERP releases the order; MES executes; MES confirms quantities, times and reasons back. Without the return path the ERP lives in a parallel reality.
4. Audit trail by default. Every event — user, timestamp, machine, order, reason code — is written immutable. This is not optional for ISO 9001, IATF 16949, GMP, or any serious traceability requirement.

Where digital workflows deliver the most measurable value

Five workflow families show the fastest, most reliable payback in discrete and batch manufacturing. They are also the ones we ship most often on the SYMESTIC platform.

• Downtime capture & escalation. PLC signals micro-stops automatically; operators categorise; shift lead sees aggregated losses in real time. Removes the 2-to-3× under-reporting typical of paper logs.
• Setup & changeover. Setup checklist delivered to the shop-floor client, first-piece approval captured digitally, batch released automatically once quality signs off.
• Order confirmation back to ERP. Cycle counts reconcile to work orders; quantities and times flow to SAP, Infor, InforCOM, Dynamics, proAlpha via REST or IDoc. End of manual shift reports.
• Alarm & maintenance handover. PLC alarm → MES classification → maintenance ticket with machine context. Mean time to acknowledge drops materially once the handover is automated.
• Shift handover. The outgoing shift sees exactly what the incoming shift will see: open alarms, active orders, scrap events, quality holds. No more "the last shift said…" conversations.

Hard-earned lesson from 12+ years building the SYMESTIC platform: we onboarded a customer who had already "digitised" their scrap workflow with a tablet app. Operators entered defects on the tablet, a rule engine routed alerts to quality, a dashboard showed trends. On paper, a textbook digital workflow. When we connected the presses directly, real scrap was 2.3× higher than what the tablet showed. Operators logged the defects they noticed; they did not log defects that happened during cycle time, during shift change, or when the line was under pressure. The workflow was digital end-to-end — except for the one step where a human decided whether to log or not. Lesson: a digital workflow that depends on voluntary manual input at the source is not a digital workflow. It is a dashboard with extra steps. The machine signal has to be the trigger, or the whole thing is cosmetic.

How SYMESTIC implements digital workflows architecturally

The SYMESTIC platform runs on Microsoft Azure as a microservice architecture. Shop-floor events arrive through IoT gateways — OPC UA for modern controls, MQTT for networked plants, digital I/O for brownfield machines without any digital interface. Events hit the ingest layer in real time, get enriched with order context from the ERP, and flow into the workflow engine where rules fire, notifications are dispatched, and confirmations are written back.

The ERP integration is bidirectional: SAP R/3 via ABAP IDoc, Infor/InforCOM, Microsoft Dynamics/Navision, proAlpha and others via REST or file interface. Every event is immutable and timestamped for audit. Gateways install in two to four hours per machine — no PLC change, no production stoppage, no LAN retrofit required for digital I/O. That is the architectural reason customers can start with one line and scale to a plant in weeks instead of quarters.

What this looks like in the SYMESTIC deployment pattern

Neoperl correlates PLC alarms with stoppages and quality defects on fully automated assembly — the alarm-to-maintenance workflow removed 10 % of stops and 15 % of scrap. Klocke runs a validated-environment packaging workflow at the Weingarten site, scaled from one line to the full plant in three weeks on digital I/O gateways with a Navision ERP file-interface handoff. Schmiedetechnik Plettenberg runs bidirectional order confirmation against InforCOM — work orders and master data flow into SYMESTIC, quantities, stops and status flow back automatically. Carcoustics runs the same pattern at scale: 500+ machines across seven countries, MQTT via IXON gateways, SAP R/3 via ABAP IDoc. The architecture is the same; only the adapters differ.

FAQ

What is a digital workflow in simple terms?
A digital workflow is a sequence of tasks and decisions in a factory that is executed automatically by connected software — triggered by machine events or user actions, governed by rules, and logged end-to-end. The outcome is the same work done in seconds instead of hours, with full traceability and without paper, spreadsheets or verbal handoffs.

How is a digital workflow different from workflow automation or BPM?
Business Process Management (BPM) focuses on administrative processes (orders, approvals, tickets) with latencies of minutes to hours. A manufacturing digital workflow is triggered by shop-floor events — machine signals, alarms, cycle counts — and operates in seconds or sub-seconds. BPM tools are not built for OT data rates or for integration with PLCs, OPC UA and MQTT.

Do I need an MES to run digital workflows on the shop floor?
For anything beyond a single isolated process, yes. Digital workflows need real-time machine data, an event-driven rule engine, bidirectional ERP integration and an audit trail — that is the definition of an MES at ISA-95 Level 3. Standalone apps can cover one workflow but break as soon as orders, quality, maintenance and planning need to talk to each other.

What is the difference between a digital workflow and RPA?
RPA (Robotic Process Automation) simulates a human clicking through existing applications — it is a UI-level workaround when APIs do not exist. A digital workflow in manufacturing uses actual protocols (OPC UA, MQTT, REST, IDoc) and event data. RPA belongs in back-office automation, not on the shop floor.

How long does it take to implement a digital workflow?
For a single workflow on a single machine with a gateway already in place: days. For a plant-wide scale-up covering downtime, setup, order confirmation and alarms: weeks to a few months, depending on ERP integration complexity. Klocke scaled from one line to the full Weingarten site in three weeks on digital I/O gateways.

Can digital workflows work on old brownfield machines without a modern PLC?
Yes. Digital I/O gateways tap existing machine signals (stop, cycle, good/bad) without any PLC change. For machines built in the 1990s with nothing but analogue signals, the gateway converts them into structured MES events. Almost every machine can be connected — the common "our machines can't do this" assumption is rarely true in practice.

Are digital workflows audit-proof for ISO 9001 and IATF 16949?
When the MES writes immutable, timestamped events including user, machine, order and reason code, yes. Audit-proof is a property of the event log, not of the workflow itself. Shop-floor tablets that allow overwrites or deletions without trace do not meet the bar. The SYMESTIC platform writes append-only event logs by default.

How does SYMESTIC implement digital workflows?
SYMESTIC runs a cloud-native microservice platform on Microsoft Azure with IoT gateways for OPC UA, MQTT and digital I/O. The workflow engine reacts to real-time machine events, applies customer-specific rules, integrates bidirectionally with SAP, Infor, InforCOM, Microsoft Dynamics, proAlpha and others, and writes an immutable audit trail. See production metrics, production control, alarms, and the ISA-95 reference for the integration pattern (ISA-95).

Related: MES: Definition, functions & benefits · OEE: Definition, calculation & practice · Machine data collection (MDE) · Production data collection (BDE) · OPC UA · MQTT · ISA-95 · Smart Factory · Paperless manufacturing · Process data module.

About the author

Mark Kobbert

CTO at SYMESTIC. 12+ years building the cloud-native MES platform on Microsoft Azure — microservice architecture, IoT gateway development, real-time data processing for 15,000+ connected machines across 18 countries on four continents. B.Sc. Wirtschaftsinformatik (SRH Hochschule Heidelberg). Expertise: cloud-native MES architecture, Microsoft Azure, microservices, OPC UA, MQTT, IoT gateway development, edge computing, ISA-95 integration, ERP-MES integration, brownfield machine connectivity, real-time data processing, IT/OT convergence. · LinkedIn