INDUSTRY FIELD GUIDE
Other poultry cohort and inventory records for AI
An inventory correction can refer to a cohort without documenting an actual transfer.
NAICS 112390 · Other Poultry Production
An inventory correction can refer to a cohort without documenting an actual transfer. USDA APHIS publishes poultry industry materials. A producer can simulate a small event sequence to test whether a model distinguishes the recorded count from an observed movement. USDA APHIS poultry industry manual
Create two imaginary cohorts with similar labels. Add a dated count correction for one and a verified transfer for the other. Ask which entries changed the practice inventory and whether the records support a transfer. Include a case with no independent verification so the answer must acknowledge uncertainty.
Use no actual animal health records, facility details, supplier identities, live footage or worker data. Use independently written values and have a reviewer check the mock sequence.
Describe a cohort records exercise
Original Fiund exercise
A corrected count is not itself a recorded transfer
Two imaginary poultry cohorts have artificial counts and original event cards. No animal handling, health information, facility detail or live footage is involved.
| Record | Source | What it records |
|---|---|---|
| PC1 | Cohort A count | Day 1 inventory 20. Reviewed correction replaces it with 18 and says earlier count error; no movement event associated. |
| PC2 | Cohort A2 transfer | Distinct cohort A2 begins at 30; verified fictional transfer of 4 to destination D on day 2, leaving 26 at source. |
| PC3 | Unverified note | Says 'A moved 2' without date, source document or reviewer confirmation. No destination receipt supplied. |
Question for the model
Which inventory changes are supported, and which records actually document a transfer rather than a count correction?
Show the reference answer and checks
PC1 supports replacing A's erroneous day-1 count with 18, but explicitly supplies no movement. PC2 documents a four-unit transfer for separate cohort A2 and its resulting source count of 26. PC3 is an unverified, undated note and cannot explain PC1's correction as a confirmed two-unit transfer. Keep corrected counts, documented movements and unresolved notes distinct rather than assuming every numerical change represents an observed movement.
- Turning a count correction into a transfer event.
- Merging similarly named cohorts or using an unverified note to create a destination history.
Fields to preserve
- Cohort and reference date
- Defines the count.
- Correction reason versus movement type
- Distinguishes different causes of record change.
- Verification and source/destination reference
- Supports or limits a transfer claim.
Test another case
Add a signed mock correction confirming PC3's date but not its cohort identity. Require the identity conflict to remain open.
Use independent fictional values and reviewer keys. Exclude actual health, supplier, worker and facility information; no animal decision is made.
Discuss a contribution
Describe the sources you control, the task they could support and what has already been checked. Use the existing-data assessment for an archive or the collection brief for a new recording project. The contributor agreement defines Fiund’s licensing authority and compensation. A guide or fictional exercise does not establish an accepted collection or a sale.