简体中文界面。帖子、指南及政策正文保留原文;未对原文作自动翻译。

We planned 100, but the count includes repeat inspections

LiamBrooks0838 · 2026年8月3日 20:39 UTC

回复讨论
LI
LiamBrooks0838
I'm reconciling our batch of 100 for sample verification with Universal Robots UR5e in a fixture inspection station with item labels, using a molded housing. The dashboard looks nearly finished because its post-restart counter includes repeat inspections. Items can have failed checks followed by reviewed retries. I want the accepted-item count to be accurate without discarding that attempt history.

16 条回复

LI
LiamAllen0316

Can you follow one labelled item through its attempts? Compare successful-attempt count with its actual accepted disposition.

19
LI
LiamBrooks0838

I've traced one item in our saved history and found several successful attempts with the same identity. Our counter adds each success, inflating the apparent batch progress.

16
LI
LiamAllen0316

@LiamBrooks0838 Rebuild an offline total from item dispositions, keeping attempts underneath. Counting each accepted identity once should give you the right batch total.

20
EL
ElenaBennett0775

@LiamAllen0316 Only if accepted means the current disposition. An item that passed and was later rejected can't stay counted just because it passed once.

6
LI
LiamAllen0316

@ElenaBennett0775 You're right about that missing qualification. Use each item's current authorized disposition for the batch total, retaining earlier results without treating any historic success as permanent acceptance.

10
LI
LiamBrooks0838

For an item that fails and then gains reviewed acceptance, should the report count one accepted item while preserving both inspection attempts beneath it?

18
LI
LiamAllen0316

@LiamBrooks0838 Yes. Attempts explain its history; current disposition determines whether that item contributes to accepted quantity.

7
LE
LeahBell0669

@LiamAllen0316 My own report showed attempt count separately from accepted-item count, making the reinspection history visible enough to explain why those totals differed.

14
FA
FarahBrown0939

@LeahBell0669 Does a retry get a new item ID then? Or just a new attempt ID?

17
LI
LiamAllen0316

Same physical item, same item identity. The distinct inspection gets its own attempt identity, linked back to that item.

24
LI
LiamBrooks0838

The newer entries distinguish item and attempt identity, but some older results lack the item link needed to include them reliably in an accepted-item total.

19
EL
ElenaBennett0775

Avoid inferring unique items solely from event order; those unidentified results might contain repeated attempts against the same physical item.

13
LI
LiamBrooks0838

@ElenaBennett0775 I'll keep the unlinked results explicitly unresolved pending comparison with our physical traceability records, instead of inventing item identities to complete the total.

16
FA
FarahBrown0939

Do unresolved items count as rejected? Otherwise the report's got another number people have to understand.

8
LI
LiamAllen0316

Keep them separate. Unknown identity or disposition isn't evidence of rejection, just as it isn't evidence of acceptance.

13
LE
LeahBell0669

My batch had a manual counter correction with a note but no item link. It took us a while to stop treating the note as an extra item.

21

参与讨论

欢迎来到 Application Robot

所有人都可以阅读论坛。登录或注册后即可发起讨论、回复或上传照片。

忘记密码?

注册账号即表示你同意我们的 使用条款 社区准则。请阅读我们的 隐私政策 ,了解我们如何处理你的信息。