{
  "title": "Rice milling data for AI fraction and grade-report reading",
  "source_page": "https://fiund.com/industries/rice-mill-fraction-report-data",
  "authorship": "Original fictional Fiund editorial exercise. No actual operating records or third-party source assets are included.",
  "example": {
    "title": "A matched separation worksheet supports fractions while a lookalike tray image does not",
    "context": "An original mill fixture uses inert kernel-like shapes, textual image annotations and arbitrary masses. No rice product, actual grade, inspection service or certificate is represented.",
    "records": [
      [
        "RMF1",
        "Sample A worksheet",
        "Parent test portion 50 practice mass units. Reviewed separation gives whole fraction 35 and broken fraction 15, exact A label."
      ],
      [
        "RMF2",
        "Sample A report",
        "Original report explicitly cites A worksheet and uses whole test portion as denominator for both fractions. No grade determination supplied."
      ],
      [
        "RMF3",
        "Lookalike image",
        "Same tray background and scale, but readable sample B. A few visible whole shapes; no A link or complete mass evidence."
      ]
    ],
    "question": "Which fractions follow from A's matched records, and can the image substitute for its sample worksheet or establish a grade?",
    "answer": "RMF1 and RMF2 support 35/50=70% whole and 15/50=30% broken by the stated practice mass basis for A. RMF3 belongs to B and shows only some visible objects; its background and scale cannot verify A's fractions or whole-lot composition. Preserve the named denominator and sample reference rather than substitute visual counts for mass or turn this original calculation into an official grade determination.",
    "failureModes": [
      "Using a similar tray background as proof of sample identity.",
      "Mixing visible count and mass fractions or omitting the denominator when reporting percentages."
    ],
    "fields": [
      [
        "Sample, image parent and worksheet",
        "Defines the matched evidence."
      ],
      [
        "Whole/broken mass and denominator",
        "Supports fractions."
      ],
      [
        "Reviewer scope and grade evidence",
        "Limits the conclusion."
      ]
    ],
    "holdout": "Change the tray pattern and ask for broken mass as a share of whole mass. Require the new 15/35 basis to be explicit rather than reuse 30%.",
    "boundary": "Use original inert props and dummy records. Exclude actual grades and proprietary reports; qualified staff retain official inspection decisions."
  }
}