Olivia_Blair · 13 Dec 2025, 06:21 UTC
Define the interrupted stages before moving the save instruction. The design must distinguish a block still at source, held by the tool, placed, and not yet confirmed. One pocket number cannot describe all of those.
9 pointsControls suggested saving after release instead. That leaves the opposite gap, doesn't it?
11 pointsYes. A stop after release but before the save can leave the stored position behind. The recovery design needs to handle uncertainty, not just move it to another line
20 pointsDoes startup show the saved number as a confirmed empty pocket?
17 pointsIt says next pocket. No uncertainty shown.
9 pointsWho is expected to reconcile the interrupted transfer when the operator cannot establish where the block is, Olivia? That responsibility belongs in the design as well as the stored states.
2 pointsYasmin's wording point matters. I would have controls hold an uncertain transfer for the agreed reconciliation, with the last stored information identified as such. Reopening must not convert it into a fresh request.
20 pointsAnd reconciliation is more than choosing a screen value. The physical recovery and access need assessment with the integrator before those choices become operator instructions
9 pointsDesignated maintenance lead, Pavel. Agreed method still outstanding. We're not handing this to production yet.
21 pointsCan the software test a stop at both sides of release without moving hardware?
17 pointsThe developer can test the state logic with simulated observations and interruptions. That is useful coverage, but it will not establish how the real placement is confirmed or how physical recovery is carried out.
6 pointsOffline cases now include both gaps. Also reopening with the block state unknown.
9 pointsCases written, or passed?
22 pointsOlivia, include a repeated interruption during reconciliation in the discussion with the maintenance lead; people can be called away before an agreed recovery has finished.
25 pointsWritten when I posted. Passed now, with unknown remaining held after reopening. Physical method still under review.
15 pointsDoes the test also reopen after the second interruption Pavel describes? A hold that survives one restart may still be cleared accidentally by a later path.
4 pointsKeep the operator trial separate from those software results. Someone unfamiliar with the fault should understand why the station is held, without being invited to guess the missing placement
7 pointsWhat will it display if the tray identity is missing too?
13 pointsThis discussion is closed to new replies after six months without activity. Last activity: .