• • • AI • • •
FIX YOUR DATA BEFORE YOU CHASE AI
BY TIM CROWE, CO-FOUNDER AND CEO, WRXFLO “ D
on’t buy AI for the shop floor; build the shop floor that earns AI.” Tim Crowe, Co-Founder and CEO of WrxFlo, on why
connected data must come before the algorithm. AI is now part of almost every conversation
about manufacturing. But for many plants, the real challenge isn’t deploying AI, it’s giving it something reliable to work with. On the factory floor, the problem is rarely a lack of technology. It’s a lack of context.
When the system is
someone’s memory Manufacturers have digitalised the business in pieces. Some teams run on spreadsheets, others on paper, others on legacy systems that don’t talk to each other. Valuable knowledge sits in people’s heads, especially experienced operators, planners
and maintenance staff. That works until someone leaves, goes on holiday, or a shift changes and the business discovers how much of its decision- making relied on memory instead of a system. That’s why AI fails when it’s bolted onto a
broken process. Feed it incomplete, disconnected or poorly structured data, and it will still produce an answer, just not always the right one. In manufacturing, that’s risky. A system that guesses quickly loses the trust of operators and managers. Once trust is gone, adoption gets much harder.
Foundations before algorithms The better approach: build the digital foundation first. Before asking what AI can do, manufacturers need to ask whether they’re capturing the who, what, when and why of daily operations. Are failure reasons logged consistently? Are quality issues traced back to the machine, material or operator action? Are planning decisions stored where they can be reused? If not, AI has little to draw on. This matters because the frontline drives
everything else. A small error at machine level can snowball into a bigger one in supervision, planning and executive reporting. Even good decision-
18 ELECTRICAL ENGINEERING • JULY/AUGUST 2026
making suffers when the input data is poor. Clean, connected operational data sharpens visibility across the organisation and gives every level a stronger base for smarter decisions.
Why context matters
on the factory floor AI is moving so quickly that many manufacturers feel pressure to act before they are ready. But the real challenge is not deciding whether AI has potential. It is deciding whether the business has enough structure, context and consistency in place for AI to be genuinely useful. On the factory floor, many of the hardest
problems are not technical in the traditional sense. They are operational. A machine goes down overnight and the reason is buried in an email, a text message or someone’s memory. A seasoned technician retires and takes years of setup knowledge with them. A quality trend starts to drift, but nobody can immediately tell whether the issue sits with the machine, the material or the operator. In each case, the work still gets done, but the context needed to make the right decision is missing.
electricalengineeringmagazine.co.uk
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
