A plant can finish commissioning a new robot cell while its job descriptions still describe the old production line. The equipment has changed. The operator's screen, the technician's diagnostic tools, the restart sequence and the supervisor's escalation decisions may have changed with it. Yet the training record might show little more than a completed course and a date.
Automation keeps changing the boundaries of the job
The International Society of Automation used International Automation Professionals Day on April 28 to discuss the future-ready automation professional. It's a useful description, although plants still have to decide what "future-ready" means for each role.
Consider a vision system added to a production line. An operator who previously inspected parts may now review exceptions generated by the system. A controls engineer may inherit responsibility for access permissions, data flows or model monitoring. The job titles don't have to change for the work to change.
Automation professionals already need a wider mix of technical skills and knowledge as industrial systems bring control engineering, networking, software and data closer together. The practical challenge is identifying which of those skills matter for a particular person, machine and shift. Sending everyone to the same general training course won't answer that question.
Training completed doesn't mean ready to work alone
Plants often treat course completion as a convenient stand-in for competence. It's easy to record and simple to audit. It can also give managers more confidence than the evidence supports.
A technician may pass a vendor's online module without having isolated the modified machine, diagnosed a live fault or followed the plant's local recovery procedure. An operator may understand an HMI demonstration but still need supervision when several alarms arrive at once.
Readiness needs observable definitions. Depending on the task, those might include:
- Understands the basic purpose and hazards but requires supervision
- Can perform the task correctly under supervision
- Can perform it independently within defined limits
- Can troubleshoot unusual conditions or train other employees
These levels shouldn't be copied blindly across every role. The evidence needed to operate a standard interface won't be the same as the evidence needed to modify a safety PLC or authorise a restart after maintenance.
Start with what changed
The most useful skills review begins with the equipment or process change, rather than the training catalogue. Plant managers should first identify the affected tasks. Did the project change how a person starts, stops, monitors, cleans, adjusts or troubleshoots the equipment? Did it introduce a new energy source, interface, access level or escalation path? Can someone now issue a command remotely that previously required physical access?
This analysis needs input from the people who understand the work from different angles. Engineering can explain the technical change. Operations knows how production behaves under pressure. Maintenance understands failure and recovery. EHS can identify changes to hazards and procedures. Experienced operators often spot awkward handoffs that don't appear in the project documentation.
Automation.com has previously argued for updating the skills and knowledge requirements attached to changing automation roles. At plant level, that update should produce a clear list of tasks, conditions and limits for each affected role.
"Understands the robot" is too vague. "Can place the cell in a safe state, identify the source of a routine fault and escalate non-routine faults without bypassing safeguards" can be assessed.
Define what proficiency looks like on the job
Once the requirements are clear, each one needs an observable standard. The International Society of Automation's overview of the Automation Competency Model explains that the model identifies the knowledge and skills needed in the automation profession. Its needs analysis matrix can help employers define role profiles and assess gaps, but plants still have to adapt those requirements to local roles and equipment.
That local detail is where a broad model becomes useful.
The Automation Competency Model provides a useful starting point for describing automation knowledge and skills. It shouldn't replace task analysis. No industry-wide framework can know the hazards, permissions and operating conditions of a specific line.
Clear behavioural anchors also make assessments fairer. "Experienced" can mean almost anything. "Performs the full startup procedure without prompting and identifies both conditions that require escalation" gives the assessor something concrete to observe.
Build an evidence chain, not a checkbox
The strongest evidence comes from the work itself. For a routine operating task, direct observation may be enough. Troubleshooting might require a controlled practical exercise or fault scenario. Higher-risk work may call for formal certification, documented supervision or assessment under an employer programme. The method should match the task.
A useful record identifies the requirement, expected proficiency, assessor, evidence, date and scope of the assessment. It should also state what triggers another review.
Make gaps visible across shifts and systems
Individual assessments solve only part of the problem. A plant also needs to know whether capability is available when and where production depends on it.
One shift may have three people who can operate a robot cell but only one who can recover it after a fault. Weekend maintenance may have no verified backup for a new safety system. A technically qualified employee might work at another site or cover a different line.
A role-based competency matrix for automation teams can map the proficiency required for each role against the evidence currently held for each employee. That makes shift-level gaps and single-person dependencies easier to see.
The matrix supports planning. It doesn't decide whether someone is legally certified, licensed or safe to perform regulated work. Those decisions still depend on valid requirements, suitable assessment and the applicable rules.
Used carefully, the view can help managers choose where to focus cross-training, supervision or recruitment. It can also stop a plant from discovering its coverage gap at two in the morning, when the only person who knows the system isn't answering the phone.
Reassess when the work changes again
Qualification isn't permanent when the task, equipment or hazard changes. OSHA's lockout/tagout rules provide a precise example. The agency's employee retraining guidance identifies several triggers, including changes in job assignments, machines, equipment or processes that present a new hazard and changes to energy-control procedures. Retraining may also be required when inspections reveal deviations or when an employer has reason to believe an employee's knowledge or use of a procedure is inadequate. The stated objective includes re-establishing employee proficiency.
That requirement applies specifically to lockout/tagout, so it shouldn't be presented as a universal rule for every workplace competency. The underlying management lesson travels well, though: a material change should reopen the readiness question.
A software update might alter a restart sequence. A retrofit could introduce stored energy that wasn't covered in the previous procedure. An incident review may show that employees understood the written steps but couldn't apply them under actual operating conditions.
Reassessment should follow the risk and the change. An annual calendar reminder is no substitute for noticing that the work is different.
Treat workforce readiness as part of commissioning
Commissioning proves that the system can perform. It should also confirm that the people responsible for operating and supporting it can do their jobs under the new conditions.
Before handover, plant leaders should be able to answer four questions: What work changed? Who is affected? What evidence demonstrates readiness? What event will trigger another assessment? If those answers aren't clear, the equipment may be ready for production while the organisation isn't.
