I'm reviewing the operator screens for an FR10 sample station. The application crashed after sending an inspection request, then restarted with the plate marked pending. Its helper resends pending work automatically.
Pending also means not sent yet. Those are wildly different situations for the person standing there. What wording and recovery decision should I ask the developer to separate before I train anyone on this?
Controls can check the PLC's retained job register. The local file has no saved response. I'm asking for a visible 'outcome unknown' state which doesn't trigger another request.
Does the saved local request carry the same attempt identifier as the retained PLC record, so the person reviewing it can reconcile a particular attempt rather than just a plate name?
There is a shared attempt reference. Controls can match it, but only if the operator doesn't clear the local entry first. That clear button is another problem.
Our shift used to dismiss anything that looked like an error just to get the main screen back. I'd give them a way to view other information without clearing the unresolved attempt. Otherwise the interface is practically asking them to erase it.
Maintenance first, controls for reconciliation. Developer has separated viewing the status from clearing it. Unknown attempts stay visible and are excluded from automatic resubmission in the offline checks.
Nadia, try the revised screen with someone who missed this whole discussion. Ask them what happens to the plate next. If they say 'press restart and hope', the wording still needs work.