I have spent a lot of time on composites and plastics floors over the years, and if you want to see where a manufacturing process quietly loses track of itself, watch the demolding step for a shift.
I have spent a lot of time on composites and plastics floors over the years, and if you want to see where a manufacturing process quietly loses track of itself, watch the demolding step for a shift.
Demolding is the point where the part comes off the tool. In composites it is the moment after cure, when a layup finally separates from the mold it was built on. In plastics it is the same idea at a faster clock speed, injection or compression parts popping loose every cycle. It looks like the simplest step in the whole process. It is usually the one where the paper trail breaks.
Here is why. Everything before demolding is attached to something that does not move much. The mold sits in a press or an oven. The layup sits in the mold. You know where it is because it has nowhere else to be. The second the part comes free, that certainty ends. Now you have a loose part, and somewhere nearby you have a mold that is either going back into the cycle immediately or heading off for cleaning, inspection, or repair. Both of those things need to be tracked separately from here on, and most shops are still tracking them with a travel card, a grease pencil mark, or someone’s memory of which cavity that part came out of.
The stakes are not abstract. If a part comes off a mold with a known defect history and nobody flags it, that part can travel three more process steps before anyone catches it. If a mold goes back into rotation without its post-cycle inspection because the routing sheet got separated from it during demold, you find out when the next part fails, not before. And when a customer asks which mold produced a specific serial number, six months after the fact, the honest answer on a lot of floors is “we think it was cavity 3, but check the log.”
None of that is a knock on the people running demold. It is a fast, physical, hands-full step, and it is the worst possible place to ask someone to also do careful data entry. That is exactly why it breaks.
I want to be straight about this rather than oversell it. RFID is good at answering two questions: where is a thing, and which thing is it. It is not a substitute for process control, and it does not measure cure state, resin flow, or whether a part came out clean. A press controller or a process monitoring system owns that.
What RFID, done right, can do at a demold station is close the identity gap. A tag on the mold, read automatically as it enters and leaves a defined area, gives you a location and a dwell time without anyone writing it down. That is a real, honest use of zone-level tracking, and it is worth saying clearly what “zone-level” means: a reader covering a demold bay or cleaning station can tell you a mold arrived and roughly when, typically settling to a confirmed read within a few minutes, not the instant it crosses a threshold. That is fine for knowing which molds are sitting where and for how long. It is not a stopwatch, and I would not pitch it as one to anyone running a cycle-time study.
Passive tags on parts or travel cards, read at a fixed station right at demold, are a better fit if what you actually need is a hard checkpoint: this specific part, off this specific mold, at this specific time, logged without a form. That pairing, a zone-level view of where the molds live and a hard scan event at the part level, is closer to how the problem actually breaks down on a real floor than any single technology on its own.
If a plant’s honest need is faster than that, sub-minute cycle correlation tied directly to a press signal, that is a process-control integration question, not an RFID question, and I would rather say so than stretch a technology past what it does.