I was asked to rank two handling proposals, one using an FR10, from their demonstration videos. The first used a clean machined tray. The second used our returnable tray, including the awkward corner pockets.
Both suppliers selected a run in which the spacers were placed successfully. Neither supplied setup time or a record of operator interventions. The internal comparison sheet nevertheless has a winner highlighted in green, based on a small difference in the visible cycle.
I am not comfortable signing that sheet. What common trial would let us compare the proposed systems without pretending those videos held everything else constant?
Which demo won, the arm or the tray?
ThomasBrown0886 · 5 May 2025, 20:11 UTC
15 replies
Replying to ThomasBrown0886
Agree the input, accepted output and included manual work first.
19pointsReplying to KaiBrown0889
The machined tray is included in that proposal, but someone must transfer our incoming parts into it; the video begins after that work is done.
10pointsReplying to ThomasBrown0886
Then time that transfer too.
5pointsReplying to ThomasBrown0886
We had a similar comparison where the faster cell needed a person sorting parts off-camera. Once that work was included, the decision looked rather different
13pointsReplying to OmarBrown0926
Was the sorting part of your normal incoming process, or work introduced by the proposed cell?
14pointsReplying to ThomasBrown0886
Introduced by the cell. Our existing process didn't need it. That was the bit we'd missed
8pointsReplying to OmarBrown0926
Exactly the cost to include.
21pointsReplying to ThomasBrown0886
Our lab compared two dispensers using little piles of hand-selected parts. Both looked wonderful. Then the technician who actually prepared the work asked where the rest of the bag had gone. We'd removed nearly every awkward sample before the trial. Nobody meant to cheat, we just wanted the demo to go smoothly. I'd let the operators choose a representative incoming set and write down any preparation. Also ask for the unedited run and intervention notes. The rejected and interrupted bits can tell you more about owning the equipment than another minute of successful transfers.
10pointsReplying to RosaChan1075
Keep the awkward samples in.
8pointsReplying to KaiBrown0889
Careful with that. Representative awkward samples, yes. Every damaged tray you can find, no. Otherwise you've replaced a flattering demo with a punishment test that doesn't describe the workshop either.
3pointsReplying to LucyBrown0915
Fair correction. Representative is the point.
9pointsReplying to LucyBrown0915
We have a proposed incoming set selected with production, including the corner-pocket variation we routinely see; both suppliers will receive the same input description and acceptance criteria.
11pointsReplying to ThomasBrown0886
Will you watch a changeover too?
12pointsReplying to RosaChan1075
Yes, including tray preparation and the operator's involvement. The ranking is withdrawn until those trials provide a comparison we can defend.
10pointsReplying to ThomasBrown0886
That green cell can wait.
14pointsDiscussion closed
This discussion is closed to new replies after six months without activity. Last activity: .