{
  "title": "Hay bale and forage sample data for AI evaluation",
  "source_page": "https://fiund.com/industries/hay-lot-forage-sampling-data",
  "authorship": "Original fictional Fiund editorial exercise. No actual operating records or third-party source assets are included.",
  "example": {
    "title": "A composite sample describes its defined lot rather than a neighboring cutting",
    "context": "Two invented hay stacks and artificial lab cards illustrate sampling scope. The package contains no feed recommendation, real forage report, buyer account or physical bale handling.",
    "records": [
      [
        "HY1",
        "Stack and sample map",
        "Stack A cutting 1: composite sample CA from approved practice bales A1–A4. Stack B cutting 2 has separate lot ID LB."
      ],
      [
        "HY2",
        "Results",
        "Lab card RA links CA to arbitrary quality index 10. Local moisture reading M-A2=12 practice percent applies only to bale A2 at the stated time."
      ],
      [
        "HY3",
        "Unrelated card",
        "Lab card RB is cutting 3 from another lot; same quality index units. No composite result for LB supplied."
      ]
    ],
    "question": "Which result is associated with Stack A's defined composite, and can the local reading or unrelated card characterize Stack B?",
    "answer": "HY1 and HY2 link RA's quality index to composite CA from Stack A's specified cutting and contributing bales. M-A2 is a local moisture observation with a different measurement meaning and scope. HY3 belongs to another cutting and lot, not LB. Stack B therefore lacks a linked composite result in this packet. Preserve sample method and lot definition before discussing representativeness; the exercise does not determine feed suitability.",
    "failureModes": [
      "Treating a single-bale moisture reading as a lot-level quality analysis.",
      "Transferring a nearby or same-unit result across cuttings without a sample-to-lot link."
    ],
    "fields": [
      [
        "Lot, cutting and contributing bales",
        "Defines the composite's population."
      ],
      [
        "Sample method and result type",
        "Separates local from composite evidence."
      ],
      [
        "Lab reference and unmatched lot",
        "Exposes missing associations."
      ]
    ],
    "holdout": "Add an LB sample collected from one bale only. Require it to remain a single-bale result rather than silently call it representative of the stack.",
    "boundary": "Use original mock cards and approved photographs if needed. Exclude commercial tests and buyer data; lab reuse rights are separate, and staff retain feed decisions."
  }
}