AI tools are reaching the factory floor, quality office and maintenance workshop. The practical skills operations teams need to use AI well, from asking good questions to checking outputs, and how to build them without overwhelming people.
A maintenance supervisor was given access to an AI assistant connected to equipment manuals and maintenance history. For the first month, he barely used it: he was not sure what to ask, did not trust the answers and found it quicker to call a colleague. After a short, practical session using his own equipment and real problems, he began using it daily, to find troubleshooting steps, draft work orders and summarise shift handovers. The tool had not changed. His skills had.
As AI tools spread into operations, the difference between value and frustration is often skills, not technology.
Skills operations teams need
- Knowing what AI is good for: finding information, drafting, summarising, analysing patterns, and where it is weak: specific numbers, safety-critical decisions, unfamiliar situations.
- Asking good questions: clear, specific prompts with context, such as the machine, product, symptoms and what has already been tried.
- Checking outputs: verifying answers against sources, data and experience before acting.
- Understanding data: basic data literacy to read dashboards, trends and charts, building on skills such as SPC.
- Protecting information: knowing which data can be used with which tools, following company policy.
- Improving workflows: recognising routine tasks that could be simplified or automated.
Building skills practically
- Short, hands-on sessions using people's own work, not generic demonstrations.
- Role-specific examples: maintenance, quality, production planning and team leaders each use AI differently.
- Champions in each area who help colleagues and share tips.
- Simple guides: example prompts, do's and don'ts, and how to check outputs.
- Time to practise, with support.
Leading the change
- Explain why: how AI will help with real frustrations, such as searching for information, paperwork and repetitive tasks.
- Address concerns about jobs openly. In most operations, AI changes tasks rather than replacing roles, especially where labour is already scarce, as discussed in labour shortage strategies.
- Involve people in choosing and improving use cases.
- Recognise people who find good uses.
Measuring progress
Track usage, time saved on specific tasks, quality of outputs and user confidence. Collect examples of good use and share them.
Connecting to continuous improvement
AI skills fit naturally with improvement culture. Teams that already look for waste, solve problems and standardise good practices adopt AI more effectively, applying it to real problems rather than chasing novelty. AI becomes another tool in the improvement toolbox, as described in kaizen in the age of AI.
