Skip to content

Industry field guide

Fossil generation reporting-basis data for AI review

Gross and net figures can describe the same interval while representing different quantities.

Sector
Utilities
2022 NAICS
221112
Updated

What this guide tests

After the correction, which gross and net periods are documented? Is a January-to-February net change established?

See the specimen and fields · Work through the fictional records

Task and source context

Gross and net figures can describe the same interval while representing different quantities. EIA’s electricity survey guidance and Form 923 data show the importance of reporting definitions. A generator can test whether a model labels each quantity correctly and preserves the relevant reporting period. EIA guide to electric power data · EIA Form 923

Invent two monthly records and mark one as net, one as gross, and a third as unverified. Add a correction that changes the period for only one record. Ask which values can be compared and what basis must be confirmed. Grade units, source, and period before arithmetic. Keep operational advice outside the evaluation.

Use synthetic values and aliases. Do not export plant configuration, control details, forecasts, contracted fuel terms, equipment settings or live telemetry.

Describe a reporting-basis sample

Industry research

The U.S. Census Bureau’s industry definitions establish which activities these codes cover. The workflow sources in this guide describe the task or record type. Fiund’s pilot below uses fictional material to test a specific decision within that scope.

221112 · Fossil Fuel Electric Power Generation

This U.S. industry comprises establishments primarily engaged in operating fossil fuel powered electric power generation facilities. These facilities use fossil fuels, such as coal, oil, or gas, in internal combustion or combustion turbine conventional steam process to produce electric energy. The electric energy produced in these establishments is provided to electric power transmission systems or to electric power distribution systems.

U.S. Census Bureau, 2022 NAICS Manual, p. 113 ↗

Build a pilot sample

Set up this pilot

Prepare synthetic gross and net report columns across two periods. Correct a single gross entry, then test whether the model preserves measure definitions and refuses an unsupported net trend.

Prepare the test packets

  1. Start with the worked case. Use the fictional record types below as a guide to the source files you need. Give each packet a stable ID and save its source versions with it.
  2. Change one fact. Make one packet with enough evidence to answer the question and one where the evidence points to a different answer. Keep the other fields fixed so a reviewer can tell what changed.
  3. Remove the decisive evidence. Give the reviewer a third packet with a plausible near-match in its place. Have them write what can still be established and what evidence would resolve the rest.

Fields the reviewer needs

Measure definition
Keep the gross or net label with the amount.
Effective reporting period
Use the corrected period for the specific record.
Correction target
Preserve which extract the correction changes.

Use Plan a sample of business data for AI to turn the specimen into a brief with a defined task, source and review owner.

Fictional exercise

Keep gross, net and corrected periods attached to figures

Original demonstration units illustrate survey-data association. The records contain no real output, forecast, plant configuration or equipment setting.

Records supplied to the evaluator
Record IDSourceWhat it records
GROSS-G1Practice monthly extractG1 records 120 demonstration units, labeled gross, for January.
NET-N1Earlier practice extractN1 records 100 units, labeled net, and initially names January.
CORRECT-N2Confirmed period correctionThe correction states that N1's period is February. It does not change G1's period or supply January net output.

Question for the model. After the correction, which gross and net periods are documented? Is a January-to-February net change established?

Show the reference answer

G1 remains a January gross figure of 120. N1, as corrected by N2, is a February net figure of 100. The package has no January net figure, so it cannot establish a net change between the two months. Comparing January gross with February net would mix both period and reporting definition. Preserve the correction's limited scope.

Download the fictional records and answer key (JSON)

How to check the result

Ask for the answer, the record IDs that support it and any fact the evidence leaves unresolved. Have a reviewer compare the response against the reference before deciding whether the sample is ready to expand.

Errors to flag in this case

  • Treating gross and net values as one interchangeable measure.
  • Keeping N1 in January after its explicit period correction.
  • Applying N2 to unrelated G1 merely because the records share a folder.

Change the inputs and run it again

Add an authorized January net figure with the same stated unit and definition. A comparable difference becomes calculable only after that missing source is supplied.

Score the decision and the evidence separately

Save the reviewer’s expected answer, supporting record IDs and unresolved fields in a separate answer key. Score the model on the decision, its citations and its handling of the deliberate gap. A correct answer citing a different case counts as an evidence error.

Before expanding collection, have a second reviewer run the same checks on the held-out packet. Keep related versions, views and exports of the same case together so they stay on one side of the training and evaluation split.

Check permission and scope

All values are artificial. No generation performance, fuel decision, forecast or operational advice is provided. Real facility and commercial records remain excluded.

Use a controlled simulation or an authorized export with topology, customer details and control-system information removed.

Describe an authorized archive through Fiund’s existing-data assessment, or describe a new capture capability through the collection brief. The contributor agreement covers the terms for participation.

Sources and further reading

Resources for this workflow

How the examples work

Company and program reviews

Use these reviews to compare sourcing, capture or evaluation approaches. Confirm the applicable program, accepted sources and terms with the provider.

All company reviews

Related industry examples

← Utilities