My FR5 puck-loading sequence works uninterrupted, but a stop between pick, place and advance leaves pocket memory at odds with the cell. I need a recovery design the next shift can use, not another counter adjustment.
Write down what the saved value means first. Next pocket to pick, last pocket commanded, or last placement confirmed? Those are different records. A counter named pocket often ends up serving all three until an interrupted run exposes the disagreement.
I would also list what the maintainer can establish about the puck without relying on that value. Recovery needs an evidence source for its physical state. The last command alone is insufficient.
Then don't rename it completed pocket. Keep its actual meaning and add the missing transfer state explicitly. Who decides an uncertain placement on the shift when you are absent?
Jo, adding a state helps describe the uncertainty, but it doesn't give the shift a permitted recovery action. The cell owner and responsible designers need to define that, including when the operator stops and calls maintenance.
Agreed. I'd use an offline table of interruptions to find the missing decisions before touching the live routine: before pick, after pick, after release is commanded, after placement is established, and around saving the advance. Some rows should say cannot establish, escalate. That's useful information, not an incomplete table you should force into a green state.
Retain that row as uncertain. If the restart screen offers continue there, a user may reasonably believe the system has checked the placement when it has not.
Also try replacement of the tray while stopped. Even a correctly saved number belongs to a particular tray arrangement, not every tray subsequently put in the station.
Who is defining it, Sara? I would bring the people who perform the approved access and recovery work into the review, not just whoever can change the program. They need a usable boundary and an escalation route.
When the draft returns, test restart during recovery as well. Otherwise you can fix interruption of production and leave a second, less documented interrupted sequence inside the recovery procedure.
The review has defined who confirms source, tool and destination state under our approved access procedure. Offline checks now hold ambiguous cases instead of advancing, including interruption during recovery. Live return-to-service testing is still outstanding.
Has the shift accepted the wording for those held cases? An accurate internal state can still be misunderstood if the prompt sounds like permission to resume normal movement.