The shift sheet names the tray family, not the individual tray or insert. It doesn't settle which of the two arrangements was present.

Tell support that limit explicitly. Ask the author to include the destination identity in a new capture, linked to the physical arrangement and the job that selected it.

Can the covering operator identify those two arrangements from the normal loading position?

Only by a small mark at the rear. The loader usually recognises them from where they are stored, which isn't a reliable handover.

Then the job needs a visible arrangement identity as well as a better export; otherwise you'll record a selection nobody can confidently match to the tray.

Luca's point is separate from inferring the old stop. Fix future identification without rewriting the uncertain history.

Agreed. We're adding a visible identity to each approved arrangement and having the author expose the selected one. The old occurrence remains uncertain.

Has the maintainer confirmed the next capture includes the events before the wait, rather than only the selected arrangement at the end?

Yes. It retains the request, receipt and destination condition leading into that step, with one occurrence reference.

I'd have the covering operator try the identification and export together, so the two improvements actually meet at handover

And leave the new wait unexplained until it occurs with that capture; an improved recorder is preparation, not a diagnosis.

The handover walkthrough is arranged. No new matched stop yet, so support still has the original export with its destination uncertainty clearly stated.