What would a generated freight invoice versus proof-of-delivery task contain?
A logistics scenario showing the records, rule text, allowed actions and held-out cause that a generated task pack could seed for a quantity disagreement between an invoice and a delivery record.
A freight forwarder's accounts payable agent receives a carrier invoice `FINV-3310` for 24 pallets. The proof-of-delivery record `POD-9072` shows 22 pallets signed for, with a damage note on a third consignment line.
A generated task pack could seed this case with the purchase order, the booking, the carrier rate card, the delivery scan events and the carrier's email thread. The held-out root cause is that two pallets were refused at the dock and returned, and the carrier billed the outbound leg for all 24 before the return was logged.
The agent's tools let it read all five records, check the rate card clause on refused goods, and take one of three paths: approve the invoice, short-pay with a dispute note, or hold and ask the operations coordinator. The policy text says short-pays need a reason code and a note to the carrier, and it forbids editing the invoice.
The correct path is a short-pay of the two pallets with the reason code `REFUSED-RTN` and a note citing `POD-9072`. If the agent approves in full, the variance is paid out. If it edits the invoice quantity, it alters the carrier's own record.
**What it shows:** a task pack can include both the data and the rule text, and a grader can check the action, the reason code and the note content separately.
A task pack can carry data, rule text and a held-out cause together, so graders can check action, reason code and note separately.
- Audience:
- Agent developers
- Content type:
- Scenario
- Buyer question:
- What can be generated
Content release: r-a220eb6e099b243b
Related questions
- What file layout and schema would a generated case set use?
- Can a pack include policies, tolerances and escalation rules, not just records?
- Which record elements should a dataset for this workflow include?
- What records and relationships can MirrorFirm generate for this workflow?
- What fields and file formats can a generated mismatch pack use?
- Can a release include mismatch cases that have no single correct cause?