Automation engineering guide · Operator and support interface

HMI, Alarm, Recipe and Diagnostic Design for Supportable Machines

Designing operator interfaces that expose machine state, first-out faults, interlocks, recipes, trends and maintenance evidence without bypassing control ownership.

An HMI is part of the control-system interface, not decoration placed over PLC tags. It should explain current state, available actions and reasons why an action is unavailable. Good diagnostics reduce recovery time without giving the operator unsafe or ambiguous controls.

Design around operator tasks

Structure screens around running, changeover, setup, fault recovery, quality checks and maintenance rather than copying the PLC tree. Keep navigation, colours, units, command confirmation and role behaviour consistent. Show mode and machine state persistently when they determine which actions are permitted.

Treat commands as transactions

An HMI command should pass through a defined PLC interface. The PLC validates permissions and state, acknowledges acceptance or rejection, and owns completion. A momentary screen button must not write directly to a physical output. Where an operation has consequences, show the object, action and relevant condition before confirmation.

Build actionable alarms

Alarm text should name the affected asset, describe the detected condition and provide a useful first action. Record event time, activation, acknowledgement and clearance. Preserve first-out information and suppress only consequential alarms that are genuinely redundant.

Alarm priority should reflect required response, not the preference for a colour. Review alarm rates, standing alarms and chattering conditions during commissioning and production support.

Control recipes and parameters

Separate product recipes, machine configuration, calibration data and engineering limits. Validate ranges and cross-field dependencies in the PLC or controlled service, not only in the browser. Record recipe identity and revision with the produced batch or cycle where traceability is required.

Make the active values and requested values distinguishable. Define when changes take effect and how partial downloads or communications loss are handled.

Provide useful support evidence

  • State and mode history around the event
  • First-out fault and active interlocks
  • Trends for relevant process values, setpoints and outputs
  • EtherCAT and device communication health
  • Software, recipe and parameter revision
  • Clearly indicated simulation, forcing or maintenance bypasses

Remote support should receive the minimum necessary access through an approved security route. Diagnostics should aid investigation without normalising uncontrolled remote operation.

Primary technical references

References are provided for software architecture and implementation planning. Validate the versions, licences, support matrix and regulated-use requirements applicable to the final deployment.

From technical concept to production system

Apply this technology through an Axiotech engineering work package

Axiotech can connect the compute, AI or analytics platform to the machine controls, data contracts, validation evidence and lifecycle-support model required for industrial use.

Engineering support

Apply this guidance to your machine or control system

Axiotech can assess an existing TwinCAT project, define the software and controls architecture, implement a bounded work package, or support commissioning and lifecycle recovery.