Most OPC UA rollouts nail security and connectivity but leave scrap and rework data semantically unmapped, forcing quality engineers into monthly spreadsheet archaeology. Here’s how to model it right the first time.
Read MoreDay: July 30, 2026
Bolt-On AI or a Separate Time-Series Store: What Your Historian Can Actually Support
Every major MES vendor now sells an AI copilot or predictive-quality module, but the historian underneath it was built for trending and compliance reporting, not model training. Here’s how to tell whether yours can carry the load.
Read MoreYour Predictive Quality Model Works. Your MRB Process Wasn’t Built for This Many Holds.
Predictive quality models are moving from advisory dashboards to closed-loop MES actions that auto-trigger holds — and plants are discovering their disposition workflows were never built to handle the volume. Here’s how to redesign hold logic before it becomes the bottleneck.
Read MoreLevel 3 vs Level 4: The ISA-95 Boundary Vendors Keep Asking You to Forget
As vendors pitch MES/ERP “convergence” as a feature, the ISA-95 Level 3/4 boundary is getting harder to see. Here’s how to re-draw it, and a three-question test for any convergence pitch.
Read MoreNaming IEC 62443 in Your RFP Isn’t a Security Requirement — It’s a Trap
Citing IEC 62443 in a purchase order without target security levels and evidence requirements is a paper shield — here’s how to write SL-T clauses that let you actually reject a bad bid.
Read MoreYour Predictive Quality Model Isn’t Wrong Yet — It’s Just Not Talking to the Recipe Anymore
As predictive quality models spread across more lines in 2026, the real risk isn’t accuracy at go-live — it’s silent drift after a tooling change nobody told the model about. Here’s how to build detection and sign-off into the MES itself.
Read More